"How can a wargame be realistic?" Topic

Of course, it's all made up. Some gamers love it, and they're not judging the results on realism, they've accepted ease-of-play rules art ahead of some claims to military science/accuracy.

FlyXwire:
That is fine. Leggo enthusiasts like the M-4 tank in the picture I provided rather than the 'realistic one'. Terrific. For wargamers they rarely know what was done for playability and what was done for historical accuracy. The designer of F&F claimed both for his design. Where each was provided and how much is the question? Having generic artillery in F&F, was that a playability issue or is the combat impact of such batteries in play historically similar to actual historical evidence? Who knows? The designer does, and you and I don't.

The game experience of knowing What and Where you are simulating something in history is far different from 'wondering' and imagining.' One thing gamers have become good at is rationalizing game mechanics and results that have no explanation from the designer.

But, but, but…….if the particular science doesn't really exist for these historical features in the first place, then when putting pen to paper, the art of rules writing will fill in the blanks, so to speak.

That is a common problem with ALL simulation designers, from research [remember the Galaxies colliding] to the military. There are methods for dealing with absent information and data. I also believe that a lot of historical 'unknowns' are simply a lack of research on the part of designers.

I'm conflicted – that the act of playing a wargame often leads to choosing between what you like about gaming, versus what you know about history, or what you can surmise is contained in any set of rules – play mechanisms and stuff just made up.

Yes, you can 'surmise…' That shouldn't be something the customer should have to do. As for the conflict, I played F&F quite a bit knowing that is had some very wonky history. It is a choice. This isn't about what gamers want. They are welcome to want playability and history in any quantity they like. This issue is knowing what they are buying and playing… how much history and playability levels are they getting. When designers claim BOTH, shouldn't gamers expect both to be clearly delineated? At the moment, there is no 'ingredients label', just hype. [unsupported or proven claims]

That is not right and is and has been detrimental to the historical wargaming hobby.

Wolfhag, more than once you have bemoaned the fact that few wargamers, particularly at conventions, aren't interested in 'realism.'

Most have a non-existent or limited experience of how realism is experienced in a simulation, and rarely can identify it when designers provide so little information.

On the other hand, I think you are asking far too much of our hobby. Let me explain with the example of the RC airplane hobby. I have mentioned this before. It still applies.

The hobby has levels of hobby involvement exemplified by their categories of models: Free Flight, Semi-Scale, Scale, True Scale.

Free flight are planes that usually are pre-made and don't resemble real planes. They usually have limited controls. They generally found with the occasional weekend flyer.

Semi-Scale are planes that have some of the 'look' of real planes, historical and current, but still are not to scale, wings and such being out of scale. Many are built by the modelers themselves. The next two categories are almost always built by the hobby enthusiasts.

Scale These planes are physically scale on the outside. Details are missing, but in overall form, to scale.

True Scale These are incredibly detailed, cockpit, rivets, dropping bombs and flying at scale speeds.

Each category fulfills the wants of particular hobbyists. The last category, True Scale is fully dedicated to 'realism' in all its forms.

Should the "True-Scalers" bemoan the fact that the "Weekend Free Flight" folks aren't interested in 'realism?' Do they? No. Because it is clear which is which need and desire. Some, like my dad moved from Free-flight to True Scale as most everyone else in the latter category. Others are quite happy to stay at with Free-flight planes…and everyone is satisfied. That is how hobbies general provide for entry level and growth.

What do the RC modeler magazines display most of the time? The last two category planes: Eye candy.

And at what level do most people participate in the hobby? [Free Flight] The least? [True Scale] So, Wolfhag, would it be any surprise or a problem when 'most' wargamers, particularly at conventions, don't seek a whole lot of 'realism'?

Our hobby problem is that we have no 'leveling up' process for gamers in regards to realism. Command & Colors Napoleonics, LaSalle, Shako, Age of Eagles, Napoleon's Battles, Empire V, etc. etc. etc. all have one thing in common: They ALL claim to provide the gamer with the same ability to use Napoleonic battlefield tactics in a historically faithful wargame.

I don't know about you, but if I believed Command & Colors Napoleonics offered the same historical experience as Shako or Empire, I know which I'd play. That is the situation for a good many wargamers. Others look at the claims and write it all off as hype and it's all fantasy.

I read all this with some amusement Perhaps I'm an anomoly, but the spectacle weighs as much, if not more, that realism. While I don't want a game to be a farce, I know from my military background (21+ years) that a tabletop will NEVER emulate realism. Perhaps institutions like West Point use wargames in military studies, but their approach is different to a casual gamer. I've been gaming for almost 50 years, with more rulesets than I remember. I started with Grant's The Wargame. Went full circle and came back to it early this year, with only modifications to some command and control, but nothing onerous. I'm having fun with the the spectacle on a 12x6 table. Nearing my winter years now and happy with the hobby as it is for me. I don't want to integrate my military career with moving toy soldiers across the table.

McLaddie, your points are valid, but rules aren't perfect shoe-ins for reality.

As PB12 points out above, as the hobby greys, we're perhaps as interested in getting the game on, and enjoying the spectacle for what it is.

I am in a similar position to PB12. Wargaming for over 50 years, gaming different periods and using many different rules for the variety of periods played. All I want now is a game where the rules I use give me results which appear to give results that conform generally to what my knowledge of the period suggests.

One aspect that has not been mentioned are the facilities available to gamers.
If you are only able to game once or twice a week and have to set up and shut down your game on each occasion, then I would suggest the style of rules you may have to use will be different than where you have a permanent facility and a game can be continued from week to week unbroken. I would suggest that the desire for greater realism is better served when the latter gaming facilities are available.

Perhaps I'm an anomoly, but the spectacle weighs as much, if not more, that realism. While I don't want a game to be a farce, I know from my military background (21+ years) that a tabletop will NEVER emulate realism.

PB12: I doubt you are anomaly. I have been wargaming since the 1970s. As a teenager, my first miniature army was a Napoleonic Naples 30mm army I molded myself. I love the spectacle as much now as then.

However, with your military background, you don't think the Sherman tank model top above 'emulates' reality to some extent? Just a little bit?

While I don't want a game to be a farce,…

Neither do I. Maybe you are more forgiving than I am. A farce, compared to what? History/reality? If you are making that comparison, that is getting close to the technical simulation term for 'realism.'

For me, a farce would be a set of rules like Fire & Fury where the Confederates take 1.5 hours game scale to cross fields that took the real combatants in Pickett's Charge 20 minutes. Or maybe Nappy Nappy [among other rules sets] where the movement rates provided make it impossible for units to ever accomplish what the original combatants did.

Or how about most all Napoleonic command processes provided in our wargames which are the exact opposite of the historical command dynamics? Do those count as 'farcical?' It makes it even more farcical when the designers claim these rules are 'historically accurate', simulate, recreate, represent, etc. etc.

I play and enjoy wargames that have no more historical realism than the Leggo Sherman tank above. If the term and the expectations for historical accuracy and 'realism' dropped out of the hobby's lexicon tomorrow, I would have no problem with that. It would be more honest and, if nothing else eliminate so much purposeless discussions, gamers guessing what the designers meant with a rule and gamers talking about 'reasonable results' which mean very little of anything historically or in comparison to reality.

When I know how simulations work and hobby designers continue to claim they are faithfully representing history with their rules… I think it is reasonable to have those expectations for their rules and be dissatisfied when they fail to the point of farce.

That's it. B.S. and farcical claims don't do our hobby any favors.

McLaddie, your points are valid, but rules aren't perfect shoe-ins for reality.

FlyXwire:

If you felt my points were valid, why are you saying "rules aren't perfect shoe-ins for reality."

The word perfect was never used nor implied. A participatory simulation/wargame can do certain things to emulate certain parts/aspects of reality. That can be objectively established. Knowing these specifics are an important part of a participant's experiencing that reality.

Perfection isn't the issue. The issues are: Being functional, achieving stated design objectives, and providing evidence of such for the consumer.

All I want now is a game where the rules I use give me results which appear to give results that conform generally to what my knowledge of the period suggests.

Chad47:
I have been gaming as long as you have. I appreciate your views. From my experience, I have found that my conclusions about rules that give results that conform generally to my knowledge are:

1. Weakened the more I have learned of Napoleonic warfare,

2. Far too often what I *think* the game mechanics and results are portraying does not sync with what the designer says those game mechanics were designed to portray, and

3. Satisfying "Results" can be achieved with a game system that has nothing to do with history. A number of our hobby rules sets have proven that. There are a myriad of game design ways to achieve the same game results. We tend to be interested in the process that mimics the historical dynamics and decision-making. Pick three different rules sets and see if that isn't true.

Little Wars Just interviewed Dick Bryant titled
"The Courier & Early American Wargaming"

YouTube link

Boy, do I feel old: "Early American Wargaming"

Putting it a different way, if we can't crack a smile or joke around at the game, there might be something amiss (my personal journey perhaps)? Though as of late, all I see are dour faces around the tabletop – has me wondering what's so darn serious with the hobby!

Clearly we are not on the same page. Like many posters I have been playing since about 14 and am now 69. I would never consider a War game a hobby for mainly a laugh and a joke. That would take it out of thew realms of a serious game like tennis, golf or Chess to name but a few.
As I get older I want more correlation between what I read and what happens on the war games table. If you are too busy laughing and joking then really you are under no pressure and not playing to the best of your abilities.

Now there is nothing wrong with that but your fundamental aims are almost irreconcilable with mine. We would never play the same games.

I love history and want my games to be a medium for understanding more.

Recreating history is not my thing, creating a secnario that may reflect issues of something I have read is. That does not take months. It takes at best a few hours. I would personally consider it a complete waste of tame to painting the exact figure.

Much money on our behalf is spent on modular terrain so battle maps can be reproduced credibly in a short space of tame and reformed to something completely different next week. My leopard 2 tanks will do for Ukraine or 80's cold war, the visual aspect is about 3D display not whether its got the right unit patch on it. Anyway I am be far too pee-occupied on working out when to yell the code word for the FDF than looking at detail on a tank that cannot be seen from standard game viewing distance, made even less serious if at the time its position is only marked on a map.

Clearly we have different aspirations never mind different rules. No rules will cover a massive difference in aspirations.

"Clearly we have different aspirations never mind different rules. No rules will cover a massive difference in aspirations."

Indeed we do.

So really this thread is about done. The question was technically wrong. No offense TP01, Its taken this entire thread to find out, not how realistic can it be, but do you

a) know what realistic means for your period, without which its a none starter.

b) At what point do you want to compromise, as you don't want a serious game, but a game with lots of laughs and jokes that primary is a vehicle to display detailed (possibly over detailed) figures and terrain at the expense of a credible model.

McLaddie,

Welcome back.

However, most designers have chosen to ignore this possibility, instead creating a legion of command points, staff points, chit pulls, etc. etc. that have very little to do with the dynamics of command. They actually reverse the command dynamics, having the 'limits' or 'command failures' identified first and then the commanders doing a resource allocation, usually every twenty to sixty minutes in game scale time. They basically chose which orders fail. 'Real commanders' did the opposite. They issued their orders to everyone, only later to find out which ones weren't followed, etc.

Yes, there have been dozens of discussions about all of the permutations of parsing the action in the game. I have not seen one yet that approaches what actually happens on the battlefield but there are many players that have heavily bought into the different versions of IGYG, unit activation, cards, and initiative determination rules. While not very realistic or historically accurate they do seem to be a necessary evil for the sake of playability and most players are already familiar with the game mechanics. After all, it is a game.

In the end, for the majority of miniature players, the visuals will determine the level of realism and almost any arbitrary system to parse the action will be sufficient. But that's not me. I like to recreate the nuanced differences between different vehicle and gun systems and how they can be a strength or weakness. For some gamers rating a Panther as a 9 and a Sherman as a 7 is good enough.

I agree with pbbishop12 that a tabletop will never be able to recreate what a real crew experiences on a real battlefield, nothing can. Realistically crews made split-second decisions with less than perfect intel and no god's eye view of the battle and the action is simultaneous. You'll never be able to recreate that in a game. However, I do think you can recreate the historical performance of vehicles and guns because you can get the historical values. You can also let players make the same Risk-Reward Decisions as real crews.

In the past, I designed a game for the "Charge of the Light Brigade" and the ACW "Battle of the Crater." The goal was a historical recreation of the battles at a convention. We had a table 20 feet long with excellent quality of 20mm miniatures and terrain with 6 players per side. I researched the battles and the tactics, movement, rates of fire and historical causality rates and tactics. There were a few Risk-Reward Tactical Decisions and leadership options players had. I did not refer to any published ACW rules. The players had a great time and both games had historic outcomes which translated to the players as "realistic." However, I designed both games that would almost guarantee a historic outcome unless players did something really stupid. I don't think that's very realistic.

In that iconic tank battle in Koln the Panther should have had the Pershing dead. His gun was already pointed down the street and intersection the Pershing was going to move into. The Panther commander survived the war and in an interview, he said he "hesitated" because he had not seen a Pershing and thought it may be a German tank. A hesitation of a couple of seconds meant his destruction. The Persing gunner said he pumped three rounds of rapid fire into the Panther because he did not know what the gunner was doing and the Panther's gun was pointed directly at him.

The goal of my design was to recreate what Otto describes. Whether it is realistic or not is up to the players, I don't make claims about realism because it is too contentious. I can claim a fairly high level of historical recreation of things like target engagement times, turret traverse speed, rates of fire, and gunnery tactics because they are taken from training manuals, combat footage, first-person accounts and after-action reports. I can't take credit for the OODA Loop either. Traditional rules that tell you what you can do and when are replaced with the player simply "looping back" to observe the results and issue his next move or shoot order, just as they would do in training or combat. The "design" aspect of the game has been documenting historical actions, making them somewhat variable with a die roll, and making sure they interact correctly with all of the other aspects. I didn't create new rules, I just document the action and create a data card for each vehicle model.

My design is a game of seconds and there are historical things that did occur that affect timing that are outside of the player's actions or control. Things like a round misfire or jam, a gunner having a hard time acquiring the target and taking additional time engaging it, a crewman hesitating, loaders can slip or drop a round resulting in additional seconds to fire again (especially if they don't have a turret basket), the equipment can break down, etc. Otto Carius described many of them in his book. So most of the SNAFU results I use are historical occurrences that I've come across that are outside of the player's actions. The fact that they rarely occur is not the point. The point is that it does give a portrayal of historical realism and eventually, they will happen. That potential threat is relayed to the players each time someone takes a shot.

This post I recently put up also gives details on all of the things that can go wrong with a tank:
TMP link

There is no way to know historically how frequently these things occurred so it's left up to the designer. Players can choose not to use it or make their own, there is nothing I can do to stop them. One or two additional die rolls in a 2-3 hour game are not going to slow anything down. The chart is made before the game so we are not taking time or "trying" to simulate an occurrence, we just roll the dice. There is no effort on the player's part other than a potential die roll. If you are playing a version of IGYG with set turn lengths and player choice of activations I don't think this is something that you could realistically integrate into that type of system.

UshCha: As far as when do I have a "proper" game, I don't take myself that seriously. Right now I'm trying to enjoy the time I have left and spend it with friends.

Wolfhag

Wolfhag

I don't take myself that seriously

Playing half halfheartedly means you get less than half the fun. You don't have fun in any hobby unless you give it your full attention. Try kite buggying with half a mind and it hurts as you inadvertently eject, way less than half the fun. But each to their own.

Wolfhag

I don't take myself that seriously

Playing half halfheartedly means you get less than half the fun. You don't have fun in any hobby unless you give it your full attention. Try kite buggying with half a mind and it hurts as you inadvertently eject, way less than half the fun. But each to their own.

As to the IGO UGO if the time steps are correct it ends up as good as the real world, that's how fluid analysis works.

So really this thread is about done. The question was technically wrong. No offense TP01, Its taken this entire thread to find out, not how realistic can it be, but do you

UshCha:

TP01 did a good job of highlighting the issues regarding 'realism' in wargames.

What he didn't ask was "how realistic something can be," but "How can a wargame be realistic?"

That is the basic technical question: How something is accomplished, not how much is included. The latter is a goal by goal question, depending on the design's purpose and ANY conclusions, simple or complicated, could all end up proving to be 'realistic' / 'credible.'

a) know what realistic means for your period, without which its a none starter.

You are imagining a 'right' answer about what a period simulation needs, an amount of content required to be 'realistic', a how much question. That is not a technical definition of 'realism.' That is a designer's goal/interpretation of what he or she needs in the way of content to achieve the goal the designer has chosen out of the billions of aspects of reality possible. Much like the two Sherman Tanks or a True Scale versus a Semi-Scale RC plane or Wolfhag's tank rules. The technical inclusion of 'realism' as a methodology is the same for all of them and the test for 'credibility' is the same.

b) At what point do you want to compromise, as you don't want a serious game, but a game with lots of laughs and jokes that primary is a vehicle to display detailed (possibly over detailed) figures and terrain at the expense of a credible model.

I think if you get rid of the notion of 'compromise' as though there is one optimum model for every topic as well as linking that to a singular 'credible' model, you would be better off in achieving 'realism.'

Phil Sabin designed a two page, 20 unit/counter game called "BlockBuster" which is included in his book Simulating War Obviously it is a simple game. There are no grenade launchers or artillery, etc. or even combat factors included in the rules. Yet, in modeling *some* of the basic tactical issues involved in Urban Warfare, it is a "Credible Model" and considered a viable simulation. Why because it was 'tested' against reality, including the conclusions of a number of army officers who were veterans of such urban combat before and after playing the wargame.

You have very strong feelings about what YOU see as important for a 'credible' simulation/wargame. Terrific but your view of what amount of *should* be in a simulation of combat is not a definition of 'realism' nor of a simulation, only one set of design decisions/objectives for your system, your goal regarding 'realism.'

It is like insisting that the Galaxy Collision simulation must include every Black Hole in the galaxies be included in its program to be 'realistic', a 'credible model.' It worked/was proven without that information.

How well the simulation matches the aspects of reality chosen to be modeled is what makes the simulation 'credible.' Not how much.

Wolfhag:
Sometimes I think you are trying to drive me crazy.

While not very realistic or historically accurate they do seem to be a necessary evil for the sake of playability and most players are already familiar with the game mechanics. After all, it is a game.

There is nothing 'necessary' about the design choices designers make, anymore than it is 'necessary' for you to use nothing but D6s in your rules because they are more 'familiar' to gamers. [which you don't] The familiarity certainly is a factor with ease of entry, but the reason designers have constantly chosen the command mechanisms they have is because either they don't know the way historical command dynamics worked [And boy have I seen that more than once] OR they found the challenge too much trouble and faking it is easier.

Realistically crews made split-second decisions with less than perfect intel and no god's eye view of the battle and the action is simultaneous. You'll never be able to recreate that in a game. However, I do think you can recreate the historical performance of vehicles and guns because you can get the historical values. You can also let players make the same Risk-Reward Decisions as real crews.

Which is 'realistic' and you have made a magnificent effort to back your rules with evidence, making that connection… doing all the work to establish your rules as a simulation, only to back off because such a claim would be 'controversial.' Not if you have the evidence to support it. And lots of designers claim historical accuracy and 'realism' and never bother to prove any of it. You have, but shy away from claiming what you've worked so hard to achieve.

The goal of my design was to recreate what Otto describes. Whether it is realistic or not is up to the players, I don't make claims about realism because it is too contentious.

Whether it is 'realistic' or not, is NOT up to the players. You designed the wargame, they didn't. You identified the 'realism' you wanted in your system, they didn't. This isn't back to one opinion is as good as another or "I like" etc. etc. etc. You built a wargame to illustrate particular aspect of historical combat. Either you succeeded or you didn't. You present the evidence and how you did it, which you have, you have 'sort of' tested your system to prove if it does present players with a game experience that matches the real experiences of tankers in specific ways. For players to determine whether your wargame is realistic means they have to provide counter evidence. They don't have to like it, but players don't get to decide based on whatever feels good.

In the past, I designed a game for the "Charge of the Light Brigade" and the ACW "Battle of the Crater." The goal was a historical recreation of the battles at a convention.

And everyone expected the specific 'realistic' outcomes from the start, so no one was disappointed. Play it a dozen times and get the same 'reasonable results.'

There is no way to know historically how frequently these things occurred so it's left up to the designer. Players can choose not to use it or make their own, there is nothing I can do to stop them.

And of course, the military never tried to find out. And of course, simulation designers never, ever had the problem of too little data, and couldn't figure out how to still demonstrate 'credible' mechanics and results.
There are proven methods to address those issues. It certainly isn't your job or anyone's to 'stop' players for making up whatever they want to. They are free to do that and have.

We are talking about how to to make a wargame realistic, objectively and provable. Making it evident to the players.

You have gone 90% of the way and then..? Lots of work, lots of creativity, lots of evidence, but nope. "I don't make claims about realism because it is too contentious." Your wargame isn't realistic unless someone else says so-- regardless of what they know or any evidence they may think they have, and even when you present them with the background about what they are experiencing in playing your wargame.

It doesn't have to be contentious…just supported with evidence.

The goal of my design was to recreate what Otto describes….I can claim a fairly high level of historical recreation of things like target engagement times, turret traverse speed, rates of fire, and gunnery tactics because they are taken from training manuals, combat footage, first-person accounts and after-action reports.

[Can you now? Obviously you don't think that is enough]

I can't take credit for the OODA Loop either.

[Why should you? You are recreating those military issues, not inventing them.]

Traditional rules that tell you what you can do and when are replaced with the player simply "looping back" to observe the results and issue his next move or shoot order, just as they would do in training or combat. The "design" aspect of the game has been documenting historical actions, making them somewhat variable with a die roll, and making sure they interact correctly with all of the other aspects. I didn't create new rules, I just document the action and create a data card for each vehicle model.

[You had to put those data points into a game system to capture that 'realism,' so don't go blowing smoke to avoid controversy]

I'll leave with this analogy, Horse Power instead Historical Accuracy and Realism, both technical issues:

Designers create V8 engines to produce 450 HP.

Designer #1: I don't want to claim it produces 450 HP, too controversial. That is for the customers to decide.

Designer #2: In my opinion, my engine produces more than 450 HP. Everyone has a right to their opinion.

Designer #3: My engine does produce 450 HP. If you use my engine, you will know it does.

Customer #1: It feels like 450 HP, it has the get-up-and-go I want from an engine, so it does.

Customer #2: It can't possibly produce 450 HP unless it has fuel-injected Duel carburetors, so I know it can't produce that HP.

Customer #3: After spending hundreds of dollars and hours putting the engine in his car, when asked if it does produce 450 HP, the customer shrugs, saying "Who cares? It's just a car engine. It gets me to the store."

Customer #4: Does it produce 450 HP? Don't get so serious. Just drive. If you don't like it, get another engine.

The claims over the years: [Just a sampling of Black Powder rules}

Johnny Reb
"The game is by nature complex, since we are simulating the complex interaction of many factors to produce a realistic effect." [Introduction, page viii]

Shako
"It is important to simulate how these systems operated, and to represent the fundamental differences between them…consequently, Shako simulates the difficulties inherent in moving and maintaining order within large formations. The system of Orders used in Shako emphasizes the pre-battle planning and battle management necessary to fight such battles." [Introduction]

Grande Armée
"…If the defender is doing what he is supposed to be doing, then all combats will be gruelling 1:1 slugfests. That may not be as much fun as the way other games let you do it, but that's the way it was in the Napoleonic wars." [Page 38]

Fire and Fury
"The ebb and flow of Civil War conflict has been recreated in Fire and Fury, an innovative game system using miniature armies to recreate battles of the American Civil War. The system, the result of five years of development, emphasizes playability without sacrificing historical accuracy." [Introduction]

Regimental Fire & Fury
"It took years of refining the rules and much playtesting to find the balance between playability and historical accuracy players expect in a Fire and Fury game. " [Foreword and Acknowledgements]

Piquet [2nd edition]
"In these notes, I hope I can make it clear to critics and fans alike what I was attempting to do with Piquet; why I believe it provides a new and refreshing approach to gaming, and why it is both a game and a simulation." [Designer's Notes]

Black Powder
"Naturally, we wish our game to be a tolerably convincing representation of real battle; however no pretense is made to simulate every nuance or detail of weaponry, drill or the psychology of warfare." [pages 4-5]

Command and Colors Napoleonics
"The scale of the game fluctuates, which allows players to effectively portray epic Napoleonic battles, as well as amaller historical actions… The Napoleonic tactics you will need to execute to gain victory conform remarkably well to the advantages and limitations inherent to the various Napoleonic National Armies of the day and the battlefield terrain features on which they fought…" [Rules Introduction. {For those who, like me, play this on the table top, but board game designers claim the very same things as tabletop designers.]

Le Feu Sacré
"The rules are mostly concerned with command, control and leadership. The actual combat elements are, although historically accurate, deliberately simplified to allow focus on what really determined victory on many Napoleonic battlefields – Leadership, and Le Feu Sacré." [Introduction]

LaSalle
If you have a club, or a larger collection and gaming area, Lasalle can also be used to simulate historical battles of the Napoleonic Wars, such as Quatra Bras, Albuera, Saalfeld, Eggmühl. [Description, Honor Games website]

Don't concern yourself with such claims, though you are supposedly spending money based on them. The products are, after all just games played for fun.

One of the issues here is that there is no common definition in usefull tems of what a player wants.

For me most commecial wargames became joyless. There relationship to what I read was so tenious that ther was no point. I wanted a game that at least bore some resemblence to the real world situation in certain aspects.

Other folk have diffrent opinions on what makes a good game. Some don't want, need or care about a credible game, they want to display toys and move them about look good and if wolfhag is correct even have an unrealated sound track to improve their experience. So be it.

McLaddie you missed customer #5
Show me the calibration data that shows it produces 450hp if it looks good I will buy. If I get it home and get suspicious I will send it to a specialist and get it tested. If it fails I will need my money back as the system is being sold fraudulently.

Our own rules go as credible as we require. They are simple, it does not have 1:1 correspondence to bits of reality, it is by definition a usefull approximation like all simulations. ITS FUN for us unlike many other games which have too tenious a link to the real world. It highlights the issues and lets us understand more of the real world which is even more fun.

If you don't understand the period and tactics you won't ever know if the game is credible or not. But in that case you proably don't care. Many players I have met fall into this category. In trying to give them a credible game, to be honest you are proably flogging a dead horse. If they are not prepared to read up to start with, why would they play the game often enough to learn by their mistakes in an uninformed way? Typically such folk don't want the complexities of planning and backup planning required of a credible game. In reality a credible game is just not their thing. Lob a few die, don't want to spend too much time thinking and instead drool over the minis is all they want or need and nobody has the right to deny them their pleasuers.

However they do forgo a right to be heard on a discussion on simulation, the same as they have no right to be heard in a discussion on football when all they want to talk about is golf.

Thought a bit more about flogging a dead horse. The answer is If folk arn't interested in simulation its definitely a no go.

A number of NATO countries can task orgaise there troops in IFV's. Our army lists cover the Task orgaisations. For a given scenario a Player given the scenario will need to play the higher command level and select what task orgaisation he considerd is required. You are not going to get the disinterested to even conduct such a mundane task, especially if he has no idea of the various merits of the organisations relative to situation he is faced with.

McLaddie you missed customer #5
Show me the calibration data that shows it produces 450hp if it looks good I will buy. If I get it home and get suspicious I will send it to a specialist and get it tested. If it fails I will need my money back as the system is being sold fraudulently.

UshCha:
Exactly. I am glad you caught it. It is the obvious response to such a claim: Prove it, Check it. I didn't miss it though. I left it out because the Hobby continually misses that most reasonable and obvious response.

If a designer claims:
"…Shako simulates the difficulties inherent in moving and maintaining order within large formations. The system of Orders used in Shako emphasizes the pre-battle planning and battle management necessary to fight such battles."

The obvious questions are "Does it?" "Where and how?" "Based on what information?"

Or

"The Napoleonic tactics you will need to execute to gain victory conform remarkably well to the advantages and limitations inherent to the various Napoleonic National Armies of the day and the battlefield terrain features on which they fought…

The question is "How do you know that? "Where is that demonstrated in play?"

The designer is the only one who can reasonably answer those questions. Supposedly, they 'proved' those claims to themselves to make such statements… but they aren't telling.

So the most reasonable response of Customer #5 is ignored, making it impossible to actually verify any of those claims. As I said: Mystery Meat.

Which brings me to your statement:

Our own rules go as credible as we require. They are simple, it does not have 1:1 correspondence to bits of reality, it is by definition a useful approximation like all simulations. ITS FUN for us unlike many other games which have too tenious a link to the real world. It highlights the issues and lets us understand more of the real world which is even more fun.

If you don't understand the period and tactics you won't ever know if the game is credible or not.

When you say "It highlights the issues and lets us understand more of the real world which is even more fun" you are categorically stating that your game has a 1:1 relationship to reality Somewhere. It is only a question of where and how you know that to be true.

If you aren't going to publish the wargame, or aren't interested in informing gamers how your game system provides that real world understanding, there is no need to do anything. You know the rules inside and out, having designed them and they meet your needs. Terrific.

But you do claim your game system has a 1:1 relationship to some aspects of history/reality.

As I've said before, if game designers continually claim simulation qualities for their wargames and gamers continue to want, discuss, and believe they are getting such benefits from those wargames, there is a distinct desire for what only simulations can provide even if gamers deny wanting or caring about simulations. That disconnect is easy and almost unavoidable when designers never justify their many claims of 'realism' and historical recreation.

McLaddie,
I appreciate your feedback.

I think what I'm trying to do in my writeup and designer notes is to show the historical accuracy of the game by comparing the values I use and the outcome to the historical research I used to arrive at those values. While we all know not all sources are reliable they are real-world factors I use instead of an abstract game mechanic I dreamed up.

Ideally, I convey this information to potential players and they arrive at the level of realism for themselves without me needing to state it. The historical and realistic level of detail for determining the factors for rates of fire, target engagement, fire control, and the different risk-reward decisions tankers use is unknown to the vast majority of players I run into.

For example, I started using terminology from the manuals. I quickly found out most players were clueless of these and the game was more of an education to bring players up to speed. I needed to present these concepts in an intuitive and common sense way.

What I've found is that players can experience "realism" in a game with exceptional figures and terrain using almost any set of rules, I've seen it time and again as a player too. Very few players are aware of the realistic tactics and weapons platform performance and bring with them a lot of false biases that other games have created.

Such as a Panther is better than a Sherman because a Panther is a 9 and a Sherman is a 6. They are not aware of the fact that a Sherman traverses its turret at 25 degrees per second and a Panther D at about 6 degrees per second and the Sherman has a commander override to engage a target without the gunner. The Sherman gunner has a roof-mounted periscope with a 60-degree field of view to engage more quickly while the Panther gunner is limited to about half of that through their gunsight and no roof periscope. The Sherman's rate of fire is over 50% quicker than the Panther too.

I recently played three meeting engagement scenarios between Sherman 75's and German Panther and Tiger I with a 6 vs 6 ratios.

I showed the Sherman player before the game how to use the Reverse Slope Defense, peeking around a building and firing, and then reversing to a new position (Shoot & Scoot). The slow engagement times of the Germans meant the Shermans were normally firing without getting return fire and being hull down. They were firing WP which screened or suppressed the Germans for 60 seconds (a lifetime in the game) and did set a Panther on fire.

When the Germans stopped to shoot the US player called in a 155mm barrage that immobilized one Panther. This forced the Germans to move closer to the Shermans and attempting to shoot on the move had no hits and made the Sherman fire more accurate.

The German player complained that the Sherman was shooting much quicker than the Panthers and Tigers and the Shermans were winning which is not realistic. I gave him the historic references and explained why the Shermans were doing better and he understood, it was a learning experience. From my experience, very few players are knowledgeable enough to give a realism rating to a game. Neither was I before I started working on the Time Competitive OODA Loop system. There can be a big difference between "feeling" something and reality. My wife reminds me of that all of the time.

I also model the Panther mantlet shot trap and the ability of an HE round hitting the turret to jam it or cave in the hull roof. A smart German player won't stick around under a barrage of HE rounds.

In scenarios where the Germans start concealed and open fire first at long range when the Shermans get to a location that was already sighted in on the Shermans get a very bad beating, there is not much they can do other than get out of the German LOS and call artillery in on the German position to force them to move. In those scenarios, we'll give the Americans a 4 or 5 to one advantage.

What I'll be doing in the future is once the game is over I'll ask the players to rate the "realism" of the game on a 1-10 scale and why.

Wolfhag

I thought what you were doing Wolfhag was explaining what drove your game design.
Leading with not claiming realism is a reasonable thing to me. Unless you then go on to claim or imply realism. I would avoid that myself.

Personally, I describe my attempt at ww2 rules as providing some ww2 tactical flavour.
Some aspects are kind of like things I feel are significant aspects of real world. Some are just there to increase the interest / enjoyability.
It's not intended to be a real world model.
Head east for that experience.
Ask nicely and someone may give you a rifle.

Wolfhag:

Sorry about being late getting back to you.

The goal of my design was to recreate what Otto describes….I can claim a fairly high level of historical recreation of things like target engagement times, turret traverse speed, rates of fire, and gunnery tactics because they are taken from training manuals, combat footage, first-person accounts and after-action reports.

So, that is an effort to recreate a historical combat environment, focusing on specific aspects of that reality. Now, have you succeeded?

What I'll be doing in the future is once the game is over I'll ask the players to rate the "realism" of the game on a 1-10 scale and why.

I'd like to suggest that you won't get the confirmation information you want with that 'generic' question. You are offering specific aspects of WWII tank warfare, not simply 'realism.'

With your evaluations, you want to confirm that players did have the experience you describe above.

You will have two types of players, ones who don't know what your wargame is specifically designed to provide in the way of a gaming experience. And you don't know how much they do or don't know about WWII tank warfare. Their responses won't tell you what you want to know. As you point out

What I've found is that players can experience "realism" in a game with exceptional figures and terrain using almost any set of rules, I've seen it time and again as a player too. Very few players are aware of the realistic tactics and weapons platform performance and bring with them a lot of false biases that other games have created.

It's that lack of awareness, of mentally creating their own 'realism' from wherever that doesn't help you know that your game provided the experience of specific aspects of real tank warfare you want to confirm your game does provide.

I know you make a point of telling participants what your wargame does before hand, what to expect as a game experience and the game mechanics and systems that provide it. It doesn't have to be a lecture, just a 3 minute run-down of what you've done. You have to create expectations if they are going to clearly look for and recognize what you are doing. As you say:

Ideally, I convey this information to potential players and they arrive at the level of realism for themselves without me needing to state it.

Then your questions afterward have to focus on that 'level of realism', not in general.

*Give them real world examples of tank warfare such as you have done with your August 2, 400pm examples. Ask them if and possibly where they experienced that. Even a 1-10 rating would work.

*Give them real world tank situations and ask them simple questions about what to expect, what they would do, and/or their situational concerns. The closer they respond like a tank commander would, the players have experienced and learned what you were after for them.

This is a fun 'quiz-like' survey to learn what you want to know.

*Ask them after playing the game, what advantages would they want to have or create in encountering an enemy tanks?

Asking them if the game providing a sense of realism, is weak for your purposes unless you ask them for specific game events or processes that produced those impressions.

Now, players will have all sorts of experiences that are uniquely theirs, and they will show up in the surveys, but when you collect 50 or more, you will see very clear trends that will confirm your success [or not]. Results that verify your success can be included in your Designer's Notes.

You have a well-articulated, specific set of design goals, you have collected an impression amount of evidence describing the reality of WWII tank combat. You interpreted this historical data and sources with your game system to recreate those experiences.

Now, you need to confirm that you have indeed provided that kind of experience for the players.

One of the critical elements in experiencing that 'realism' in a wargame is knowing what is and isn't being provided. They need to know to experience the full benefits of your simulation. If players go into the experience not knowing what to expect, they will apply expectations from everywhere but your purposes.

Andy:
Your comments are the view of many wargame designers and gamers.

Leading with not claiming realism is a reasonable thing to me. Unless you then go on to claim or imply realism. I would avoid that myself.

Well, the question is what it takes to have 'realism' as wargame content.

Personally, I describe my attempt at ww2 rules as providing some ww2 tactical flavour. Some aspects are kind of like things I feel are significant aspects of real world. Some are just there to increase the interest / enjoyability.

So you're 'kind of like' significant aspects of the real world is not 'realism?'

I think as a wargamer and possible customer, many would ask what 'real world aspects', where in the rules, what tactics? How do you know they are significant? AND, which are significant real world aspects and what are just fun mechanics having nothing to do with history, the real world or WWII tactics??

It's not intended to be a real world model.

So, what is it, a model of significant aspects of some unreal world?

The 'kind of like' doesn't save your rules from being an attempt at a 'real world model', whether it's just a little bit of reality, sort of, kind of like or a lot. ALL real world models/simulations are 'like' the real world, not the real world.

That 'kind of like' the real world, 'kind of like history', 'kinda of like' war, 'kinda like tactics' creates a 'kind of like' historical wargame which produces a kind of like historical wargame hobby. In is no wonder many wargamers have expressed a difficult time telling the difference between historical wargames and fantasy--or simply stating there is no difference.

You labeling your efforts at history and WWII tactics as 'flavor' says it all. That is easily the most meaningless wargame design term ever passed around.

It is that historical flavor of 'kind of like' wargame designs, dimorphous and ubiquitous, useless in the extreme that exemplifies far too much of our hobby.

Wargame designers have and are claiming to create real world models of some aspects of history, war and combat [model:
a representation of…], but dodge any evidence for, justification or proof of said model…or that they are even creating a model at all.

Andy, create whatever rules you want for whatever reasons, but at least admit what you are doing in regards to modeling, 'significant aspects' of history and war, etc. etc.

Head east for that experience.
Ask nicely and someone may give you a rifle.

It is silly digs like this that make me believe many know know they are dodging the intrinsic design issues and blowing smoke and 'flavors' to obscure them.

Head east for that experience.
Ask nicely and someone may give you a rifle.

He does poor design himself and denigrates others and lets face it clearly has no understanding of Simulation. Clearly no rules I would want to play.

OK guys. Andy O'Neal has been a great contributor for years. Just like everyone else posting on TMP I normally disagree with them. Including some of Andy's posts but I respect his opinion because it is constructive and a reason based on how current games are played.

I hang out here on TMP to listen to differing points of view and attempt to understand them because it's the only way I might gain some knowledge. I don't learn anything from people that agree with me. I found out long ago, much to my disappointment, I'm not a know it all even though I may come across as one. It's a burden I have to bear being right all of the time. <grin, only kidding>

In the US we are assumed to be innocent until proven guilty so I'm willing to let Andy explain himself because I know he has a logical explanation. But if he wants to tell everyone to off so be it. Free speech etc. Remember, you can always stifle him.

I do have a few friends of mine, all wargamers, that did exactly what Andy said and are in Ukraine at this moment and were given weapons and are experiencing realism very few of us experience. If anyone else wants to go and join them let me know. I can make it happen.

Give Andy a break to explain himself.

Wolfhag

I'm curious where you think I denigrated others designs ushcha.

I would say Realism is perceived rather than absolute. List a load of factors as long as you like and players will likely ignore your list and just form an opinion.

Once you get all your factors together and you have a model with all it's sub models actions and actors. You test it against what? Someone's description of real world? Which description do you even pick?
When you try and compare to hard facts it's depressingly common to find the numbers have holes or are rather doubtful accuracy.

Factors like player knowledge can be quite a challenge. Players usually know faf too much about what's happening and can often be certain what they know is true.

If one produced a truly realistic set of rules then you'd probably be short on players.
The way staff wargames are run with at least three different rooms and decisions based on a warped view of the battle is arguably quite realistic. Very dull though.

My advice to anyone designing any system is to start top down and then validate bottom up.
By top down I mean first identify major factors, the big picture. Bottom up validation looks at things you think ought to be some sort of significance, see how they fit in.

OK guys. Andy O'Neal has been a great contributor for years…

Wolfhag:

Whoa, there. UshCha's outburst was uncalled for and certainly didn't add to the discussion.

Andy's comments were more subtle. His attitude is commonly held. I wouldn't have responded except for this:

It's not intended to be a real world model.
Head east for that experience. Ask nicely and someone may give you a rifle.

In two sentences Andy dismisses the entire discussion rather than engage, and disparages the whole issue of 'realism' by suggesting that if we are all that interested in realism, go grab a gun and head east… but only if you ask nicely.

This didn't add to the discussion either.

I would say Realism is perceived rather than absolute.

Andy:

This either/or dichotomy completely ignores the issues, relegating 'realism' to again, only personal opinion.

List a load of factors as long as you like and players will likely ignore your list and just form an opinion.

Is this surprising that gamers ignore your lists when you offer them meaningless game terms like "Kinda of like" and "Flavors"? That has become the normal response when fed such empty design descriptions for decades. I commented on how completely that ignoring the designer has become because of this list of empty terms and claims.

Once you get all your factors together and you have a model with all it's sub models actions and actors. You test it against what?

Obviously, you haven't read any of the thread.

Someone's description of real world? Which description do you even pick?

'Someone's? Just one person's? Is that what we have been talking about?

When you try and compare to hard facts it's depressingly common to find the numbers have holes or are rather doubtful accuracy.

Oh, gosh. No one has ever seen that before, and simulation and historians have never, ever figured out ways to deal with that problem. [Yes, sarcasm]

Factors like player knowledge can be quite a challenge. Players usually know far too much about what's happening and can often be certain what they know is true.

Of course we are talking about history and real combat when you say players know far too much.

That is why you:
1. Let them know what information you have used in the design and where.
2. What game experience you, as game designer have created for them in regards to history and 'realism.'

Game designers call themselves 'Experience Engineers.' Among of all the personal impressions, opinions and knowledge, a game designer means to have players experience specific things in playing his game. that is what makes the game successful.

If one produced a truly realistic set of rules then you'd probably be short on players. The way staff wargames are run with at least three different rooms and decisions based on a warped view of the battle is arguably quite realistic. Very dull though.

Really? That is your idea of 'truly realistic set of rules? Again, you haven't read the thread or OP's article. And I can go on line right now have find several wargames that were set up exactly the way you describe, so there are groups of gamers who don't think that arrangement is dull.

My advice to anyone designing any system is to start top down and then validate bottom up.

By top down I mean first identify major factors, the big picture. Bottom up validation looks at things you think ought to be some sort of significance, see how they fit in.

Uh, do you know what the word 'validate' means? What exactly are you validating? Against what? The game system itself? What major factors? What sort of significance? You don't mean history and things found in the real world do you?

Come on, can you get any more vague? Game systems are built on specifics because that is the end product: A very specific set of rules to achieve very specific objectives, whether some kind of history, realism or fun.

Andy, exactly what problems do you have with the definition of realism and how to achieve that in a game system? You know, the one I have delineated in the thread. So far, you have ignored it, even when it addresses all the problems you have raised in modeling the real world.

And again, it isn't my definition or methodology, but what has been developed by the simulation community, which includes game designers…unfortunately, outside this hobby. I know it works because I used it in my job as a trainer and educator. It had to work or I didn't get paid.

Long and interesting thread here. Many of the points raised about "realism" and the ability of rules to create realism seem to miss a fundamental point from the OP's original article. That point being that wargaming rules will never be truly realistic as there is no risk to life or limb in taking part in a game. In other words we start off with a limited level of realism being available to a rules writer.

Any set of rules has to be a balance between playability and realism/accuracy. The rules writer has to decide which aspects of the battlefield experience they want or can include in their rules. I don't want to get bogged down in definitions of simulation or modelling as descriptions of what the rule writer tries to achieve so let's call it modelling in the least specific sense. Part of the skill of a rules writer is to decide what kind of game experience they want to create and where they consider the important decision points will sit in that experience. The level of technical detail on weaponry effect, formations, manoeuvre rates, willingness to fight, ability to fight to factor in should be what is required to enhance the game experience BUT it should allow, no, require the period tactics to work. Therefore there may well need to be artificial rules to penalise unhistorical actions in ways that reflect the real world penalties.

That point being that wargaming rules will never be truly realistic as there is no risk to life or limb in taking part in a game. In other words we start off with a limited level of realism being available to a rules writer.

Elenderil:
No one missed that fundamental point. Because it is fundamental to simulation and game design, no one felt the need to mention it.

That is the principle benefit of simulations and wargames: The limited level of realism. I can simulate and learn to fly without the risk of crashing a 100,000 dollar plane. I can face *some* of the real world challenges of combat without the risk of life and limb. I can simulate galaxies colliding without having to go into space to push two galaxies together. I can test a bridge construction without building the bridge or a chemical reaction without blowing up the lab.

That is why simulations and wargames are so valuable… they have a limited, but very useful [very real] 1:1 relationships to the real world. If they didn't, folks would stop using them.

I trained people in a variety of skills and knowledge sets with simulations and game, for specific situations and jobs. They could take what they'd learned out into the real world and use them immediately, successfully, avoiding the failure rate of 'learning' in the real world. That 1:1 limited relationship to the real world with simulations and games has been tested, validated, proven and used successfully for decades. There is no question that simulation 'realism' is limited. No one would want it otherwise. That is their strength.

The questions are 1. how to establish/prove/test/validate that realism in a particular system or game, and 2. Know the limitations and strengths of simulation design to use the tool to its optimum.

This isn't rocket science, it isn't complicated. It is a proven methodology saving 'life and limb' daily as well as money. The result has often been fun, the relationship between commercial games and entertainment, the military, business, and research has always been two way.

Any set of rules has to be a balance between playability and realism/accuracy.

Of course. The question in playing the rules set remains "Where is the realism/accuracy in that game experience?"

I don't mean to be rude or dismissive.

I don't think the definition of reality is practical or achievable. If it was and you gave it platers I'm not sure what it would look like. From what I've seen gamers don't really seem to want realism.

Yes, I've read the thread. I don't agree with a lot that's posted though.

I wouldn't design systems in the ways described.
I don't really care about proving what I come up with to people I don't play with.
Sorry if you don't like that or would like to argue definitions.

That's just my opinion.

People differ.

I game with people routinely practice what would be called in software development continuous improvement retrospectives. I don't recall discussing statistical proofs ever.

Setting my own game designs aside.

For his AI in our pc game Ezra compares AI suggested tactics to expert recommendations and real world battles. AI recommends turn left flank (using it's model). What do subject experts think? What happened in the real battle?
Ezra designed UMS and UMS2 which won awards and sold many units.
He was consulted with by the us army for the Iraq invasion. His game ai software mate was originally commissioned for the US army to advise company commanders real time. Real world real time advice.

Myself, I work on software professionally and most recently betting software specifically. To arrive at a price for a bet we use models. They are refined against real world experience of outcomes which are very clearly defined. I don't build those models, that's very specialist stuff. I use them.
I would say sports book factors are far more easily recognisable than wargame factors. The way those models work is to run many simulations every minute or factor change. Outcomes vs criteria are counted success / total sims = base probability.
That's base because we have margin and odds are "snapped" to a ladder so we offer (UK format) 2:1 rather than 2.0132 : 1.

Sports fixtures in real world provide the data for comparison. There are absolute definitions of these criteria. Off side or not off side. Possession orbnot. Score goal or not. The game itself has absolute definitions defined.

Real world battles are far less clear cut. Tank A shoots at B first. Why was that? Real world the chances are simple as A spotted B first. Maybe not though. What proportion of time does turret rotation speed matter more than cupola design? What effect does commander and gunner quality have in that calculation? I can only guess.
How would you validate a wargame model?
There are no such real world rules, nowhere near the clarity of outcome.

In our period it's not that hard to get reasonable approximations for a simulation. Artillery sources give factors for setting damage, define the area of the target, use the factor for the target e.g. infantry moving the open, , vehicles, infantry in fighting positions to name a few. This gives the rounds required for the shooting calibre, which can then be referenced to see if the allocated resource is enough. Validated assumptions.

Tank limitations, set the visibility of the vehicle buttoned up, does this force the simulated tanks to use formations as defined in the manual, then you have a reasonable validation. Talk to ex tankers to confirm where on the spectrum you are in your assumptions and the implied limitations in the model.

Weapon ranges, text books and accounts to correlate if there is reason to say the text book/manufactures data is a bit optomistic for the real world. Validate again other data, say the frontage over which a platoon can be expected to defend, available from a number of sources.

Run test cases to see if the model holds up and no obvious discrepancies.

It's not thst hard without super precise data to get a credible simulation.

I don't mean to be rude or dismissive.

Andy: Okay, then please ignore what I wrote.

I don't think the definition of reality is practical or achievable. If it was and you gave it platers I'm not sure what it would look like.

?? Really? I gave a very simple wargame "Blockbuster" as an example of what they 'would look like.' Wolfhag has spend a good deal of time illustrating what he has done. Why can't you see that? It's not practical? What would make it practical… if those examples don't meet that criteria, let alone 'achievable?'

From what I've seen gamers don't really seem to want realism.

Well,
1. If you don't know what realism in a game looks like, why should they. How can you want what you don't recognize.

2. That doesn't address all the designers who claim that realism for their games: Like Black Powder:
"Naturally, we wish our game to be a tolerably convincing representation of real battle; however no pretense is made to simulate every nuance or detail of weaponry, drill or the psychology of warfare."

The designer of Bolt Action claims the same thing for that game. Why continue repeating such things if gamers aren't interested?

3. And how can they recognize 'realism' when the history and real world models are described in such terms as 'kind of like' and 'Flavors?' Do gamers what that? Perhaps that is what they take as a description for 'realism' with nothing better offered.

Yes, I've read the thread. I don't agree with a lot that's posted though.

Yes, I got that.

I wouldn't design systems in the ways described.
I don't really care about proving what I come up with to people I don't play with.

Actually, I haven't described 'how' a wargame needs to be designed. Only how it's realism is determined. Lots of ways to create wargames for lots of varies purposes. No one says you have to do it one way. Wolfhag's 'way' and Phil Sabin's way with 'Blockbuster' are quite different.

I have been talking about how to determine 'realism' in whatever way you design your system. Do you prove what you came up with to those you do play with? How?

Sorry if you don't like that or would like to argue definitions.

We aren't arguing definitions. You haven't given ones and I have said that "Kind of Like" and "Flavors" don't have any meaningful/useful definitions.

That's just my opinion.

Yes, but I haven't been discussing opinions. I have been discussing technical how-to's, evidence and proofs. I have been discussing objective measures. It isn't opinion that Union troops wore blue coats in 1863. It isn't opinion that the British used linear tactics in 1812. It isn't opinion that the Americans used specific tank tactics in WWII.

People differ.

Over what they like, but that doesn't change the color of Union coats.

I game with people routinely practice what would be called in software development continuous improvement retrospectives. I don't recall discussing statistical proofs ever.

I am not surprised, but I am certain you discussed the odds of that cavalry charge breaking a square or a Sherman knocking out a Tiger or whether that MG42 was better than the MG43 etc. etc. Statistics. It's just that no one bothered to dig up any evidence and actually build a model around those questions.

Setting my own game designs aside.

For his AI in our pc game Ezra compares AI suggested tactics to expert recommendations and real world battles. AI recommends turn left flank (using it's model). What do subject experts think? What happened in the real battle? [Are you saying those questions were or weren't answered?]
Ezra designed UMS and UMS2 which won awards and sold many units.
He was consulted with by the us army for the Iraq invasion. His game ai software mate was originally commissioned for the US army to advise company commanders real time. Real world real time advice.

The point being? You seem to note that his I.A. software was also used in commercial 'fun' games. As I said, the 'realism' desired by the military many times ends up being sold as entertainment…and vice versa. But wargamers aren't interested in 'realism?'

Myself, I work on software professionally and most recently betting software specifically. To arrive at a price for a bet we use models. They are refined against real world experience of outcomes which are very clearly defined. I don't build those models, that's very specialist stuff. I use them.

But you do recognize how they are refined against the real world experiences to be useful, to be 'realistic.'

I would say sports book factors are far more easily recognisable than wargame factors.

Sports fixtures in real world provide the data for comparison. There are absolute definitions of these criteria. Off side or not off side. Possession or not. Score goal or not. The game itself has absolute definitions defined.

And so? There isn't real world data, historical and current available? You know the stuff that historical wargames are supposedly illustrating. They can be tested against them too.

It all depends on what the designer is trying to do. So, there are few 'absolute definitions' in wargame design? That makes it too hard? Absolutes seems to be a theme here. As I said, that lack of absolutes isn't some new or insurmountable problem, from the military to astronomy. Simulation designers deal with that lack every day and still succeed in creating functional simulations, real world models, tested designs.

It is just a matter of knowing how it is done. Again, look at the game "Blockbuster". Two pages of rules, 20 counters played on a 6 X 9 grid board. Yet, veterans of urban warfare say it is a real world model. It accurately portrays the tactical issue in BUA combat. The 'realism' is confirmed. RPG, APCs, artillery and airstrikes are not part of the rules. There are no "absolutes." In its obviously limited fashion, it has been proven to 'realistically' portrays the dilemmas of urban combat and tactics.

Real world battles are far less clear cut. Tank A shoots at B first. Why was that? Real world the chances are simple as A spotted B first. Maybe not though. What proportion of time does turret rotation speed matter more than cupola design? What effect does commander and gunner quality have in that calculation? I can only guess.

Uh, those questions are exactly what Wolfhag has been addressing and what the U.S. Army spent time and money answering for WWII and after. Did you miss that???

You are great at coming up with the questions, but seem to see them all as insurmountable problems without answers, which is probably why you have not seen how Wolfhag addressed them, let alone bothered to find out how simulation designers tackle this lack of absolutes.

How would you validate a wargame model?
There are no such real world rules, nowhere near the clarity of outcome.

By clarity, you mean absolutes? I have spent some time with that question of validation on this and other threads, given Wolfhag examples, again which you seem to have missed or dismissed.

I would really appreciate you explaining what you mean by
"How would you validate a wargame model? There are no such real world rules, nowhere near the clarity of outcome."

Many functional/useful simulations are not betting software nor do they require numbers and absolutes to be successful.

How would you validate a wargame model?
There are no such real world rules, nowhere near the clarity of outcome.

Andy:

I tried to respond to everything you said. 'Long-winded.' But to make it simpler. Tell me what you mean by the above and give me a historical or current example that you feel couldn't be 'validated.' I'll describe how its done.

Bird and Livingston's book "WWII Armor & Ballistics" has some excellent data if you want to implement more detail on armor and armor-piercing rounds. The book was specially written for wargamers:

it models:
armor slope effects
high hardness modifiers (more chance for spalling and internal damage)
cast armor deficiency and armor flaws (armor performs as 15% less)
APCR shatter gap (APCR rounds are not the "Silver Bullet" like they are in other games)
spaced and layered armor
edge effects
over penetration and damage level

I think it is out of print but you can find digital copies.

Good info and charts for penetration, hit chance, armor layouts for German, British, and US AFVs, firing tests, compound angles, and much more.

I've used it for modifying armor values, and implementing APCR shatter, and damage without impacting playability.

Wolfhag

Andy and McLaddie,
Real world battles are far less clear cut. Tank A shoots at B first. Why was that? Real world the chances are simple as A spotted B first. Maybe not though. What proportion of time does turret rotation speed matter more than cupola design?

I think cupola design has more to do with Situational Awareness. Turret's traverse speed has to do with the speed of engaging a target.

What effect does commander and gunner quality have in that calculation? I can only guess.

Based on training standards and AARs better crews are quicker in performing their drills, executing their orders, and reacting to threats. They are more accurate because their gunnery skills are better, estimate the range better, and are more apt to keep their gun bore-sighted. They are quicker because of training and teamwork.

Below are the results for the three crew training level differences from M60s:

Another example:

The timing included reaction time (Situational Awareness) probably under ideal conditions (unbuttoned, no suppression), engagement time (getting the gun on the target, traverse speed), and finally, the fine adjustments the gunner makes to get the crosshairs on the target (gunner skill, estimates the range, laser rangefinder, etc).

I've attempted to model those three steps using the historic timing with better crews a little quicker and poor crews a little slower. Some factors are pretty clear, some are my interpretation. I admit it's not an exact science.

Step 1: What I've modeled is a Situational Awareness Check which covers reactions and spotting. If you are overwatching where the enemy appears and you are unbuttoned and not suppressed you should notice him the second he appears. If you are buttoned up, suppressed, or flanked it may take 5-15 seconds of an Engagement Delay to notice the threat. By then it may be too late. The player rolls a D6 with several modifiers. If the result is <=0 he can go into action right away. If the result is > 0 that is the number of seconds of an Engagement Delay before the crew goes into action. Delays give your opponent several seconds of an advantage to seize the initiative and shoot first. This may force the player to use a less accurate Snap Shot to shoot first.

Step 2: Next is target engagement. If your target is 45 degrees off and you have a turret traverse rate of 20 degrees/second it will take 3 seconds (round up) to get the gun on target. If you are overwatching and the target shows up within 20 degrees the gunner will already have it engaged.

Step 3: Next are the fine adjustments and range estimation. This is taken from the US M60 tank manual:

Using the Battlesight tactic allows the gunner to shoot a few seconds sooner but with slightly decreased accuracy.

A unit concealed in ambush can engage a target at long range and then hold fire & track the target waiting for it to get closer, present a side shot, or come to a range marker.

This is my attempt at modeling the three steps from the manuals above:

1: Situational Awareness Check: The player rolls a D6 with several modifiers to see if he spots and reacts immediately or if there is an Engagement Delay of 1+ seconds. Units that are buttoned up, flanked, suppressed, no gunner periscope, and have 1 or 2-man turrets will take longer.

2: Engage the target by maneuvering or traversing the turret. The player determines how long it takes to maneuver or traverse. If the target appears where you are overwatching it is already considered engaged. You can angle your hull to the target to increase armor protection and ricochet chance.

3: Ranging Fire Action Timing (first shot to estimate the range, aim, and fire) the player rolls a D6 and modifies it by the crew type (no modifier for Veteran crews), and if the target is moving (T Move).

Snap Shot: The player now has the option to subtract up to the entire amount of Action Timing in step 3 for a Ranging shot (not for Reloading) just as real crews did to shoot first. However, there is an increasing accuracy penalty for each second subtracted (Risk-Reward Tactic). Even a point-blank shot can miss.

Recreating the OODA Decision Loop: Just as in real combat, there is no orders phase or waiting for your turn to shoot or move. When the player shoots he observes the results. If he wants to fire at the same target he rolls a D6 for his Reload Action Timing and modifies it for the crew type (no Snap Shot). If he wants to engage a new target he goes back to step 1. If he wants to Shoot & Scoot he places a movement marker. The player does just what a real crew would do.

Known Range: If the new target to engage is within 100m of the last target you hit the gun is considered engaged and ranged in. Perform a Situational Awareness Check and then roll a D6 for the Reload Action Timing and use the more Ranged In row on the Gun Chart rather than the Ranging row. This forces players to keep their units historically dispersed and not bunch them up.

Each unit has a customized data card with all of the info the player needs.

Sherman data card:

Panther data card:

‌

You can see by comparing the timing factors (the ones highlighted in tan) that the Sherman has a great advantage to get the shot off if they both spot each other at the same time because of a quicker turret traverse and easier-to-use gun. He is a little quicker engaging a moving target because the gunner has a roof-mounted periscope. His rate of fire is faster because of a quicker reload time. These are historical factors. If a Sherman shows up on a Panther's flank he may get off 1-3 shots before the Panther can get off his first one. Firing on the move is quicker because it has a gun stabilizer. Without the stabilizer, it is an additional 6 seconds for moving fire.

The Panther has an advantage for first-shot accuracy if he has a higher velocity gun and a range finder but it takes 6 additional seconds to use it. If the Panther is in a concealed ambush position, as it was much of the time in NWE, it will almost always shoot first which is the reason the Americans used recon by fire whenever they could. It may also get off more than one shot before being spotted because it has smokeless and flashless powder. Tactics matter.

How would you validate a wargame model?

For me, it is researching the timing for crews to perform actions and weapons platform performance and combat footage. Mine is designed for the cause to show what the relationships are and differences in performance. The results are mostly historical because I designed it that way. There are reasons it is "realistic" and also reasons it is not.

In a design-for-effect game using non-historic traditional IGYG and activation rules if the end result is a historic outcome you could claim the design is validated without going into all of the detail for design for cause. I think this is where the players are looking for the right "feel" using special rules and die roll modifiers to tweak the outcome to what they feel it should be.

In my system, there is no tweaking or abstractions for "balance" either. Real combat is rarely balanced.

There are no such real world rules, nowhere near the clarity of outcome.

I would have to agree with that. However, what I've done is not write up rules that I've created or borrowed from other games from the past. All I did was document the real-world timing factors for crew actions crews and weapons platforms perform with a minimum of abstractions and special rules.

It is very playable because a manually operated game clock parses the action second to second. When the clock matches players Act Time to execute an order he pauses the game, executes his order, and immediately issues his next move or shoot order. The initiative goes to the quickest.

I would think that military-grade simulators would use the same method that I've used because a simulator is Time Competitive, as is my system.

Wolfhag

I'm not doubting the effort put into research.

To pick one aspect of WW2 tank combat simulation.
Crew quality seems to be important.
How important though?
Exactly what were the significant effects and how significant compared to other factors?

To what extent did it mean tank crews did stupid things?
How do you then incorporate that – make players do stupid things?

To what extent did it mean tank crews did stupid things?
How do you then incorporate that – make players do stupid things?

I also think that the most important factor is the human element, for which there is no "hard" data. Just anecdotes and extrapolation. The lads in 4. Armoured performed superbly against those remnants of 9th SS Panzer on 25 July, which tells us what, exactly, about how they'd perform at any other time under any other circumstances against any other opponent?

I was also wondering: how much extrapolating can you do before you're no longer "simulating?" So we've got all this great data on Panthers vs. Shermans, OK. I assume we don't have that on every possible match-up of tanks in the war, much less on AT guns, ATRs, artillery used in an AT role, Flak used in an AT role, etc. I assume you don't have ready data on an Italian M11/39 versus a Soviet BT-7 or somesuch? But a game needs to have that data. Where does it come from, then?

And – I think this came up last year with the guy who claimed very grandly to be creating the perfect Napoleonic simulation – what happens when the players want to do something that didn't happen in history, ie. for which there is no data? For example, I want to fight my US army against my friend's British army because nobody else showed up at the club. What are we "simulating" then?

Getting back to the human element, wouldn't you have to compile some sort of database of all possible human reactions in a given situation, i.e. a "range" of reactions, just like you do for weapons, armor, and so on, in order to calculate the odds of Result X vs. Result Y happening?

i.e. our lads have had enough of this and they're bugging out… Or, No, our lads are full of fight today and they're toughing it out. What are the odds of X happening instead of Y, and how did you arrive at those figures?

I was also wondering: how much extrapolating can you do before you're no longer "simulating?"

Garth: That is why you test and validate. There are times when you don't know 'how much extrapolating is too much' until you test your system.

Remember the Scientist wanting to know what happened when two galaxies collided? The first scientist to ever attempt a simulation of it and many after him had to do a boat-load of extrapolating complete unknowns. The proof was in the testing it against reality. It looked something like this, but with only 1,000 pixels:

svs.gsfc.nasa.gov/10687

Or in our case, test it against historical evidence.

I also think that the most important factor is the human element, for which there is no "hard" data. Just anecdotes and extrapolation.

Uh, no. There is 'hard data.' And when there isn't, it can be created. In fact, professional simulation designers in a variety of venues have found that determining human behaviors becomes fairly easy to predict based on statistical evidence, particularly when they are groups, the larger, the easier to predict. Remember, the process for creating 'hard data.'

From tank crews to freeway drivers. Stupid actions as individual events can't be predicted, but the odds of where and how often can with some precision. There are several books out about human behavior in combat, a number based on that 'hard data.' They are easy to find on Amazon. And no, I am not talking about Trevor Nevitt Dupuy's work. Brains and Bullets by Leo Murray is one such book. One thing he notes in modern warfare is that troops that are outflanked retreat, troops attacked frontally stand. We are seeing that in the Ukraine. 100% true? No, but largely and could be made into hard data.

Here is some 'hard data' I created surrounding one question for a set of rules I am developing:

How often did units [multiple battalions and/or brigades] retreat from a firefight?

Pretty simple, right? I pretty quickly found 70 from various battles over the 20 years of Revolutionary and Napoleonic Wars.

Randomly selecting 50 [why fifty I'll explain later…statistical magic], I found that the number was
0. The Spanish at Talavera wasn't really a firefight, but a single French volley.

For units in a firefight to retreat, one of three things had to happen: The enemy was reinforced, the enemy charged or outflanked the friendly troops. Otherwise, you had the firefight simply continue.

So, I create a combat system that models that. How do I test it? ONE way is to look at the 20 examples I didn't use in my data development and see if my model predicts the actions in those other twenty? If so, I have some hard data.

That is basically how 'hard data' is developed. The amount of precision varies. Because of the zero exceptions, I can trust that data conclusion.

Garth:
What I provided as data development above is rougher than say Andy's example of the gambling software, but the same method applies.

The reason I can trust my firefight result is because of what researchers have proven over and over again. If you have 50+ data points of an unknown total base, you can trust the results with a +/-15% error.

It works like this: Pascal first developed this odds conclusion:

He told the story of sailors landing on an island. They had no idea how many natives lived on the island. The first five people they met were redhaired. Were all the natives redhaired? When is it reasonable to assume all the natives are redhaired? If met randomly, by the time you've met 50 and all are redhaired, Pascal deduced mathematically, that there was only a 15% chance that the sailors would meet a native with another color hair. After the 100th, it is lower than 5%.

Does that mean 51st native would be redhaired? No, just that the odds were against it 95-5%. That quality in statistical determination has been proven a thousand-fold and a basic in statistical work. I have just mentioned one of eight different ways a simulation can be tested against history/reality to see if it works successfully.

Professionals have all sorts of methods and tactics for statistically analyzing events and human behavior, but for our purposes in modeling real history, such basic methods are 'good enough' and far and away, better, than current wargames 'modeling' history and the real world.

That means I can create a combat process that has a 85% chance of the discovered firefight results, with 15% chance of retreat. Actually, with zero retreat results, the odds are well below 5%. That isn't Absolute, it isn't perfect, but as a model of what happened in Napoleonic combat, it is far, far closer to the history being modeled.

If you are thinking, wait a minute, what about all those chance events. With Pascal's example, the native might be old with white hair, the black-haired natives all ran away because the sailors had black hair, or it was black-haired native swim day and they were all on the other side of the island swimming. All those things and more could happen.

Here is the thing about statistical analysis. If with my example of firefights, I found that only some behaved the way I described, others ran away, and still others surrendered, with a smattering of strange events like the troops in the firefight deciding to attack someone else, what you do have? A statistical range of events, the odds of each happening. I would look to increase my data base to 75 to 100, but then I have basically a combat results table, don't I?

To what extent did it mean tank crews did stupid things?
How do you then incorporate that – make players do stupid things?

Andy: I have mentioned how to find out to what extent tank crews did anything, including going stupid.

How do you incorporate that? It isn't by making players be stupid. A participatory simulation, where players make the decisions, you don't make them do stupid things. You simply create the environment they operate in and leave them to it. The environment can of course, include those stupid non-player people. That is a dynamic simulation.

If you make players behave in certain whys with rules requirements, like those infamous McClellan rules for Antietam, you skew history badly, giving players choses that have nothing to do with the history portrayed. Or you control all their decisions and have a Static Simulation: A movie where events are decided while the audience watches, the outcome always the same. We all have played games like that, which weren't really games at all.

How often did units [multiple battalions and/or brigades] retreat from a firefight?

Pretty simple, right? I pretty quickly found 70 from various battles

Weren't there about 10,000 actions fought during the period 1792-1815? Seventy examples would represent… a 0.007% sampling, right?

If you have 50+ data points of an unknown total base, you can trust the results with a +/-15% error.

Pascal's example of people with red hair is easy to validate. Granted, what you call "red" and what I might call "red" could be different, but still… If you have red hair, then you have red hair no matter what the weather is doing, no matter what other people are doing, whether or not you're hungry or afraid, etc, and we can be pretty sure that you also had red hair the day before that, and before that.

IF we could agree on what "red hair" means, then I'd call that "hard" data.

But: Whether or not British tankers will withdraw from an engagement against enemy tankers is dependent upon a thousand factors, most of which we can't know, either because they're based on individual perceptions that were momentary and varied from person to person, or because 99% of them weren't recorded, or because the actors themselves might not have known what they were doing or why. I can think of hundreds of variables that would be involved in a decision to withdraw, and some of them would only apply in circumstances A, B, and C, while others would apply in circumstances B, C, and D, and then sometimes E but in bad weather, depending upon what you think "bad weather" is, and so on.

Randomly selecting 50

Randomly? Really? You put all 10,000 actions into a big bowl, shook it, and drew 50 at random?

Or was it the case that you chose fifty? The fifty that came to your mind for whatever reason, because you were looking for something? Maybe it was just as simple as: those were the 50 for which you had enough information. For the other 9,950 there isn't enough information, meaning: You don't really know what happened most of the time.

I recently read a study of voting patterns in elections since the 1970s, written by a pair of political scientists. It had something like 1,700 footnotes. Each footnote represents the confirmation of one data point. Study X showed that female urban voters (by "urban" they meant: living within the city limits of the following eight cities…), between the ages of 34-49, with university degrees, who regularly got their groceries at Sainsburys, etc, etc, had a 55% likelihood of voting Labour…. in that one election.

Great. I love knowing facts, and knowing how we know that they're facts. Of course, Study X was nationwide; it didn't tell me whether the 42-year-old female voter in Lincolnshire was more likely to vote Labour than the 42-year-old female voter in Sheffield.

Maybe if I were doing a game about the elections of 1979 – Maybe! – I'd break down the information in such a way that such facts would inform my game outcomes.

But we don't have anything like that sort of data for the outcomes of battles, especially not battles from long ago and far away. Claiming "accuracy" for such things just makes me laugh.

And in any event you seem to be assuming that there's a consensus regarding the desirability of such claims for accuracy. As if, at the conclusion of a game, players will think: "Well, that was great; I really enjoyed the game so much more knowing that the author claimed that his data was validated. I have no idea whether he's right or not, but it was awesome knowing that HE thought he was!"

"But we don't have anything like that sort of data for the outcomes of battles, especially not battles from long ago and far away. Claiming "accuracy" for such things just makes me laugh."

+1 Garth

"And in any event you seem to be assuming that there's a consensus regarding the desirability of such claims for accuracy. As if, at the conclusion of a game, players will think: "Well, that was great; I really enjoyed the game so much more knowing that the author claimed that his data was validated. I have no idea whether he's right or not, but it was awesome knowing that HE thought he was!"

+2 Garth

To pick one aspect of WW2 tank combat simulation.
Crew quality seems to be important.
How important though?
Exactly what were the significant effects and how significant compared to other factors?

Overall my opinion, based on what I've run across, is the crew is the most important factor in the whole process. Years ago I was at a WWII fighter aces symposium with Jim Swett and Jeff DeBlanc and others. Someone asked the group what was the best plane in WWII. They all discussed it for about 30 seconds and their answer was "It's the pilot, not the plane."

With all other factors being equal the better crew will win. Poor crews need to use ambush tactics, maneuver to the flanks, and overwhelming numbers to win or a faster weapons platform.

I think Arracourt is a good example. The Geman Panzer Brigade had new Panther tanks and new crews. The Americans had inferior tanks but experienced crews. The outcome was pretty lopsided for a variety of other factors than the crew.

Why? With all things being equal the faster crews "get inside" their opponent's OODA Decision Loop to seize the initiative to shoot first and shoot quicker. However, an Ace crew in an IS-2 with a reload time of 20-25 seconds will always be slower than the worst German crew in a Panther, StuG, Tiger, etc.

If you look at the data card when getting off the first Ranging shot the Ace crew has a -2 modifier and the poor crew +4 giving the Ace crew a 6-second advantage. When performing a Situational Awareness Check (reaction and spotting) the Ace crew also has a 6-second advantage. That's an average 12-second advantage. Against a veteran crew, the Ace crew has a 2-second advantage for SA Checks and a 2-second advantage in a Ranging shot for a total of 4 seconds. I think this compares favorably to the crew training level differences from the M60s I posted above.

I was also wondering: how much extrapolating can you do before you're no longer "simulating?" So we've got all this great data on Panthers vs. Shermans, OK. I assume we don't have that on every possible match-up of tanks in the war, much less on AT guns, ATRs, artillery used in an AT role, Flak used in an AT role, etc. I assume you don't have ready data on an Italian M11/39 versus a Soviet BT-7 or somesuch? But a game needs to have that data. Where does it come from, then?

Great question. I mention many of my sources below. Lacking any real data or evidence one way is to modify the gun's ideal rate of fire based on the turret ergonomics and if there is an ammo bin within easy reach like the T-34/85 which is easy to find.

Here is the trial of a T-34/76 in 1941: link

Now tell me what you would use to determine the reload time and rate of fire. You'd have to somewhat extrapolate it.

The T-34 had an unusual problem in that the ammo was stored in three round bins that the crew stood on that was the floor, there was no turret basket. To get to a bin the loader had to remove the rubber mat covering, pull a can out, open it up, get the rounds out, and load them in a cramped two-man turret. How long did that take in combat? Who knows.

Using customized data cards for each unit and gun type makes it easier to customize the data without looking up multiple modifiers to a base value. I hate doing that.

How historically accurate is this method in my game? It's hard to say because there is not a lot of hard data and the results in combat can be so variable. There is no scientific formula to use that I'm aware of. The key thing is to make the game intuitive and playable.

To determine reload time you can take the ideal rate of fire and modify it based on the ergonomics of the turret, if it had a dedicated loader and ammo rack. So a field gun with an ideal rate of fire of 15 rpm in a tank might only fire at 70% of that rate or 10 rpm with a three-man turret or at 50% with a two-man turret. Ideally, you get inside the tank and try it yourself. Something I'm hoping to do at the Collings Foundation Tank Museum later this year.

The only "scientific" time and motion study I know of is the British doing trials of how long it takes to reload a Panther or Tiger II. They did it with a short, medium, and tall loader and attempted to reload from the different bins, neither had an ammo rack. The Tiger II was from 6.5 to 21.9 seconds. The Panther was from 4.0 to 12.5 seconds. The maximum times being the radioman having to pass a round from the forward bin back to the loader. The Tiger II loader took 2 additional seconds to toss the shell out of the turret.

British Report: link

For the Firefly: link

Based on these you can make an educated guess for other vehicles.

To what extent did it mean tank crews did stupid things?
How do you then incorporate that – make players do stupid things?

LOL. I don't have to force players to do stupid things, they do that on their own. The game system forces players to think and act like a real tank commander which is a real challenge for most new players. His units are always active and observing to react to enemy threats, including changing orders and issuing new ones. He needs to pay attention at all times. There are no IGYG turn, orders phase, spotting phase, etc. They use the same tactics in the same way real crews did except they have time to think about it. So as far as decision making all new players are "poor crews" and make mistakes as a learning process.

From my experience, the older players are pretty clueless about what to do and take about 30 minutes of gameplay to catch on to a game that is Time Competitive with OODA Loop timing for each unit. I attribute this to playing IGYG games that tell you when you can move or shoot, activate units, etc. as it interrupts the player's natural use of the OODA Loop – when you execute an order immediately issue your next move or shoot order, there is no orders phase. The action is parsed based on how quickly units execute their orders, not arbitrary or abstracted rules. It's a different concept.

With young players, I explain it like a tank video game they have played and they get the timing concept right away. The plus side of the design is that if you do have a wealth of knowledge of historical AFV combat and tactics you can use it to your advantage in the game.

I will help them by telling them their options and the advantages and disadvantages but ultimately they make the decision, not the dice. The dice will only tell them how long it will take to execute their order.

Each time a unit fires there is a 5% chance of a SNAFU and the SNAFU Chart includes some of the historical mistakes and stupid things crews did in WWII and mechanical malfunctions. Players can make up their own if they like.

Getting back to the human element, wouldn't you have to compile some sort of database of all possible human reactions in a given situation, i.e. a "range" of reactions, just like you do for weapons, armor, and so on, in order to calculate the odds of Result X vs. Result Y happening?

I guess you could but that would be pretty difficult to do. I think the real player reactions are good enough. A player all of a sudden facing 3-1 odds will go "Oh s---t" and immediately put his vehicles at full speed and jinking to get out of the enemy LOS. If in a tight spot, he may decide on a less accurate Snap Shot to shoot first.

i.e. our lads have had enough of this and they're bugging out… Or, No, our lads are full of fight today and they're toughing it out. What are the odds of X happening instead of Y, and how did you arrive at those figures?

I think that could be something to define as part of the pre-game scenario for morale or a modifier for a morale check. Historically, small engagements of company-size units lasted from minutes to a few hours. There are no odds of X or Y happening. It all depends on the player's decisions and how quickly their units execute their orders. There is a 5% chance of a SNAFU each time a player shoots.

I'm no expert in this stuff and my short undistinguished military experience was as a grunt. What I've attempted to do is use original documents and accounts. That includes various British War Office Reports, Ballistics Research Lab, training manuals, after-action reports, trials, and combat footage to gather the data.

I've had assistance from many former and current tank crewmen, a few of which have had experience on various WWII tanks while working at museums like the Collings Foundation, Bovington (visited there) and I've had email correspondence with driver/mechanics at the Russian tank museum in Kubinka. Unfortunately, as we all know, not all of these sources agree so I had to make a decision on what to use and will not please everyone.

Just to be clear – I'm not striving for the perfect historical simulation of AFV combat or making that claim. I'm not attempting to take into account all of the human factors involved, both are impossible. My opinion is that 1:1 AFV combat is a game of seconds, and Otto Carius agrees. A Time Competitive OODA Loop approach is what works for me.

All historic weapons and vehicle platform performance can be measured in seconds so that was my approach. I tried other ways that did not work. For me, this is the level of "realism" I like and is playable enough for beginners too.

I've found the same Time Competitive and OODA Loop approach works for air strikes, artillery/indirect fire, naval battles, low-level infantry combat, wild west, modern warfare, air combat, and science fiction games too.

Wolfhag

I also do not think the hard data is really available or quite so hard.

I interviewed lots of WW2 veterans way back. One of them was involved in analysing cause of tank destruction.
He said much of his work was done in drinking establishments.
When he started they looked at a number of wrecked German tanks as thoroughly as they could. It was pretty much impossible to define root cause of battle loss. Crews blew up abandoned tanks, passing tanks shot them up after they were KO'd for whatever reason, etc etc.
I didn't attach him to a lie detector or anything but his contention – that the task was almost impossible so why bother trying?

On stupid actions.
You would need a set of forced actions or abstractions model such actions regardless of player action.

Eg some inexperienced panther crews drove along roads rather than trying to keep frontal armour presented to enemy. Consequently they were easier to KO.
Sure if you had reliable hard stats ( which imo do not exist) you could abstract out bad crews effects and give Shermans a side armour hit kill advantage. Where players move their panthers is then ignored. I doubt you'd get many players in games where player actions are effectively ignored, but it'd maybe model that aspect better than free will.

"I also do not think the hard data is really available or quite so hard."

+1 Andy

"It was pretty much impossible to define root cause of battle loss. Crews blew up abandoned tanks, passing tanks shot them up after they were KO'd for whatever reason, etc etc.

I didn't attach him to a lie detector or anything but his contention – that the task was almost impossible so why bother trying?"

+2 Andy