"least favorite dice mechanics " Topic

An earlier poll looks at favoured but I thought it might be instructive to look at the dice mechanisms that are not looked kindly on.

I personally, don't mind the 'bucket of dice' approach but I do tend to get confused by games that need a range of dice eg D6 for morale, D10 for firing etc.

I prefer the "curve" of 2D6, as opposed to the linear 1D10.

Years back I read an article by Jerry Pournelle, where he measured the exact surface area of each side of some D20s. There was up to a 30% difference among the sizes of each face. I'm sure that more modern manufacturing techniques have improved that, but I still don't trust D 20s.

Exploding dice. Really fiddly, usually pointless, sometimes really wonky results.

Percentile dice. You have to roll two dice for every roll and dose the difference between 50 and 55 percent really matter in most cases?

I'm not a big fan of opposed dice rolls, a la Force on Force and Pulp Alley.

Even as a huge fan of Full Thrust, I really don't like the re-roll mechanic.

Roll to hit, Roll Damage, Roll save. Just figure it out and roll once

Bucket o' Dice.

I had the opporrunity to play a game earlier this year that involved opposed rolls, odds or evens, and the difference between rolls all for one mechanic. Happily, I was not invited back (a bit out of my head wih PTSD that day,) because while I was excited to be gaming again, it was also too complicated for my twisted little bunny brain.

I'm not a fan of exploding dice either generally. Maybe an odd extra roll here or there but as a base mechanic I'm not keen on it. In miniature games I also don't like having different types of dice. I don't mind if we use d10, d20 or d6 or whatever, I'd just prefer they be the same.

I do like bucket o'dice to even out wonky rolling.

I had a hard time with Smoke on the Water's die conventions…..

Rolling off D12 or D10 or whatever for various types of guns and same idea for various types of armor. I did not like having to look at the type of gun, then figure out what die to use and then do the same thing for the armor I was shooting at.

I could be wrong, but I think this is what discouraged me from Stargrunt and Dirtside also.

Switching to different dice. D6 for this. D10 for that. Pick a die and stick with it. I do think that scaling dice are OK. Instead of +1, go up a die type.

Different dice. The more different types, the more I dislike it.

Years back I read an article by Jerry Pournelle, where he measured the exact surface area of each side of some D20s. There was up to a 30% difference among the sizes of each face. I'm sure that more modern manufacturing techniques have improved that, but I still don't trust D 20s.

What Pournelle failed to tell you was whose D20's he was measuring. There are two sources of error. First, bad molds. Second, and this is probably the one JP ran afoul of, those pre-inked dice are produced by coating the entire die with paint then throwing it into a rock tumbler until everything but the numbers is bare again. Of course, there is no way to control how much material gets removed from any given face, so even if the original casting was good, it's unlikely that the "finished" die is random.

I played a game once where instead of 2d6 dice we drew a card off the top of a deck of cards. 36 cards representing all the possible combinations of 2d6. The hitch was you never shuffled the deck after each draw. You gave the card to the GM and he held onto the card until all the cards had been drawn. I won't express my opinion on this dice mechanic except to say if I see game using this I walk away.

Terry

This is like asking me for my least favorite serving utensil. I would be unhappy if I had to serve soup with tongs, but that doesn't mean I dislike tongs. I like a tool that gets the job done efficiently and effectively, I don't like a tool that gets in the way or is unnecessarily clumsy. There are certainly dicing mechanics in games I dislike, but it's almost always because the results are whacky or the process is annoying.

In GQ3, two ships which get within 500 yds roll opposing d12s and collide if the rolls are within one. I understand the author was trying to encourage players to keep proper spacing, but the real result is multiple collisions in every game, way more than happened historically. I prefer to the lower roller swerves off course to avoid contact (typically going somewhere the player doesn't want it to and breaking up or diverting a formation). Punishment retained, result more believable.

I'm a bit weary of the old naval gaming habit of rolling multiple times for every shot: roll to hit, roll for damage, roll for special hits, roll for special effects of those special hits, sometimes roll to save at one or more of those steps, etc. ad nauseum. I still play lots of games that follow this pattern, but honestly, it can be done with much less rolling and chart running if the author uses a little more imagination. Every possible result has a percentage chance of occurring; why can't we just roll once to find out which/how many occurred? So many games concentrate so much of the effort on gunnery that there's hardly any time for maneuver in a gaming session.

- Ix

A bad dice mechanic has less to do with the actual dice or combination of dice being rolled, but much more with how the result of that roll is interpreted.

The space combat game Silent Death rolled several different types of dice at once, but most people loved it because it was elegantly blended into the game as a whole.

For the acccuracy of D20: look up Lou Zocchi's video on youtube on this very topic.

I'm not familiar with:

Exploding dice.

If I were to guess it would be the sort of thing where you roll a d6 and on a 1, for example, roll another d6 for various things that could happen. Can someone confirm that for me?

Many different variations are possible, but usually (in its basic form), something like:

Roll D6 for damage, if your roll a 6, roll another D6 and add the result (ad infinitum).

Yes Tim, Exploding dice is something like that …

I've seen it with more high end rolls so …

Roll a number of d6, 1-3 miss, 5 = 1 damage, 6 = 2 damage and roll again. If on that re-roll you roll another 6, then you roll again … ad infinitum.

I don't mind exploding dice (in fact I actually like the mechanic). For me I dislike using different dice (although in some cases it works) and using several rolls to get to a conclusion i.e. Roll to Hit, Roll to Save, Roll to Damage etc.

Different types of dice in the same game.
Pick one type of dice and stick to it!

Bucket o' Dice. Pass.

Roll to hit, Roll Damage, Roll save. Is there another game we can play?

Dice mechanics is less wearying than poor game design, especially in historical games.

I will have to nominate a bunch of issues that I have seen during the years:

1.) Using a dice with high granularity of results, but then failing to utilise it in the game. For example using D100 for rolls, but all stats and modifiers are in multiples of 5 – the same range of results could have been achieved using D20. The same if you use D20 but all stats and modifiers are in multiples of 2 – you could as well halve them all and use D10 instead.

2.) Dice mechanisms that do not scale nicely with the size of the unit. For example, Medieval Warfare has missile fire resolved by rolling a bunch of D10 to hit, with the number of dice directly proportional to the number of stands in the unit. However, if any of those dice rolls 1, the whole unit will then end up being low on the missiles. This can lead to some gamey tactics, such as organising your archers into tiny units in order to minimise their chances of running low on missiles.

3.) Rules using very unique dice – only available from the manufacturer of the rules. Even worse if it is not available in a single package with the rules. Even worse if the dice are so popular that the maufacturer frequently runs out of stock.

4.) Choosing to use D6 just because it is "easily available", while the added granularity of polyhedral dice (which are not exactly super-difficult to find) might have made for a better game. Especially problematic argument if players are also expected to somehow locate 18 mm Slobbovian Horse Grenadiers (and 32 x 47 mm bases to mount them) actually to play the game. (Notice: I do not have anything against use of D6 as such.)

5.) Endless opposed die roll fests. Close combat between large units in TSATF comes to my mind – I think you may well end up resolving 30+ opposed die rolls in a single round of combat, and in most cases the rolls will be for identical cases (e.g. another generic redcoat versus another generic Zulu warrior with assegai).

I don't really have any problems with any type of dice mechanics. Sometimes I have problems with the way a dice mechanic is embedded into the overall game mechanics. But that is highly contextual; not liking opposed die rolls here doesn't mean I won't like them there.

Griefbringer for the win! All excellent examples of poor dicing mechanic engineering.

Well griped, sir! Well griped indeed! :-)

- Ix

Yes what Griefbringer said.

Don't care much for being penalized for rolling the best possible fire combat result by going "low on ammo" such as Fire & Fury does, for example.

Also don't like stuff like "6s always hit, 1s always miss" regardless of modifiers.

And then there are averaging dice …

None that really bother me, but opposed rolls "feel" more skirmish oriented.

If a game requires special dice to play, I won't buy it, though I've played a few and enjoyed them.

Griefbringer for the win! All excellent examples of poor dicing mechanic engineering.

I forgot some other interesting ones:

6.) Game designer being inconsistent on whether low or high rolls are required to beat target values. I do not particularly care which option is chosen, but I would appreciate if it was applied consistently.

7.) Designs where you first roll a die to determine target value, and then you roll another identical die to beat this value. This is luckily not very common design, but I have seen it doe even with D100 (giving around 50.5 % overall chance of success – flipping a coin would give almost the same probability distribution).

Using only one type of dice with tables of modifiers when they just could have used an assortment of polyhedrals and minimal, if any, modifiers.

Those two mechanisms (changing die size vs. DRMs) don't necessarily give the same distribution of results. Changing only die sizes means there are no guaranteed/prevented results, e.g. it's always possible for a d4 to beat a d20. That might be fine for some simulations, but there are plenty of situations in wargaming where certain results should be impossible, and DRMs can guarantee that.

Having said that, I personally prefer rules that use a very small number of modifiers. It can really bog a game down to have too many modifiers to run through, and long lists of modifiers make arithmetic errors very likely.

- Ix

I'm surprised more hasn't been posted about the superiority of diceless games.

1. Long lists of modifiers.

These usually appear where a system tries to do a lot with a single die roll.

For example, if a single die roll is used for a "Roll to Hit" taking into account both shooter and target status then you very quickly get stuck with a long list of mods:

- shooter skill
- target skill
- range modifier
- shooter status (e.g. pinned or suppressed)
- target status for concealment
- target status for activity (e.g. staying low or going loud)
- target type (eg dispersed infantry vs giant tank)

Meanwhile the opposing player sits on his hands while you run through your grocery list of modifiers making sure that you haven't forgotten anything.

2. Single Die for Most Combat Resolution

I've played games using d10s in which you generally roll just a die or two for each attack and most times the difference in your "hit number" is a 5+ or 6+. In other words, it's pretty much a coin toss and so you end up with wild swings in results.

Significant advantages of rolling more dice and breaking a single roll into a couple of rolls include a more even distribution of results, BOTH players are involved in the action, and any given roll is subject to just a couple of mods making it easier to remember (it allows you to follow the "rule of 3" as they say in cognitive science…it's easier to remember things in chunks of 2 to 4 items rather than 6 or more). Examples include Flames of War, 40K, and Gates of Antares. In each case two or three die rolls are used to resolve an attack, each roll has only a few mods, and each roll uses a number of dice instead of just 1.

In theory I find a single die roll to resolve an entire process attractive but it just doesn't work that well, at least without maybe crossing indexing to a combat results table of some kind (showing my age with an oblique reference to early Avalon Hill games!)

3. Multiple Die Types in a Single Game

I strongly dislike hunting around for a d4 or d8 or whatever. Keep it simple and use one type of die. It gets worse when a system mods a roll by changing die type (eg mod from d6 to d8). And some go further by modding die type AND presenting you with different target numbers (mod from d6 to d8 AND shift from 4+ to 3+…what the heck?!?!).

4. Multiple Ways of Reading Dice and Modding Them

I dislike systems in which rolling high is better in some cases but not others or in which mods are sometimes made to a base number and other times to a die roll result. It's crazy inconsistent and confusing. Build your system to work in "one direction" and stick to it.

Significant advantages of rolling more dice and breaking a single roll into a couple of rolls include a more even distribution of results

As long as the dice are independent of each other (the outcome on one does not physically affect the outcome of the other), it does not make a more even distribution of results. Any algebraic combination of multiple die rolls can be collapsed into a single uniform die.

In fact, if you've ever made a table of outcomes from a two die roll (d6+d6, max(d6, d6), whatever…) you've done exactly that. Each cell in the table has an equal chance of being the aggregate result.

You might need some funky dice to do any give roll, but using multiple dice gives you a fixed, discrete outcome.

mod from d6 to d8 AND shift from 4+ to 3+…what the heck?!?

The four combos of the two dice and two hit numbers give you four different probabilities:

   4+  3+
d6 50% 66%
d8 62% 75%

Dear Ochoin

That's because diceless games are not superior, but inferior to ones with dice. First of all without chance there is no real game. Granted Chess is certainly a game, and it does not have dice (Well that form of it originally known as Chaturanj does,) and it can be exciting, but only from the standpoint of seeing or not seeing a fortuitous move, or a disastersous one. In Such cases however the excitement comes from omission or commission and not unexpected circumstances.

It's like predestination in religion. There can be no dramatic enlightenment and conversion, no emotional apogees and nadirs, and diceless games are unrealistic simply because chance is so much a part of day to day life. In a diceless game the person who can forsee more and make provision for form than his opponent always wins, and it is just (within that game) that he do so. But real life, which war games are always compared to, the person who uses good tactics and due diligence can nevertheless be frustrated and defeated by a person who uses bad tactics and makes stupid decisions simply by the dictates of happenstance and coincidence. We have seen this countless times in history and countless times in our own lives.

As long as the dice are independent of each other (the outcome on one does not physically affect the outcome of the other), it does not make a more even distribution of results. Any algebraic combination of multiple die rolls can be collapsed into a single uniform die.

Let's say that I score a "hit" on a 4+ using six-siders.

If rolling 1d6 the result is binary: hit or miss. Even if I change the target number from 4+ to 2+ or 5+ the result is still the same: binary, hit or miss.

If rolling, say, 8d6, then the result is most likely to be somewhere in the middle (e.g. 3-5 hits) and less likely to be among the exceptional results (zero hits or 8 hits). And if I change the target number from 4+ to one of the extremes (ie 2+ or 6) then I also shift the expected result to more or fewer hits.

Thus, with rolling handfuls of dice one can shade results across a spectrum while still getting the chance for extreme results (OMG!!! 2+ on 8d6 and just 2 hits!!!). Makes for a more exciting game than (2+, oh look, I rolled a 1).

There's also a visceral or tangible connection between number of dice rolled and the in-game action. Sniper or ATGM shot? One or two dice on a 2+. Quad Gatling Pulse Laser Cannon? Twelve dice barrage at a 5+. The player has a tactile experience that corresponds to the action being represented on the tabletop which is very different from rolling 1d6 whether firing your .50 cal sniper rifle or your Specter Gunship's Autocannon of Death.

As for the 4d6, 8d6, 2+, 4+ thing, yeah, I know the odds, but it requires the player to track multiple means of modification: number of dice up and down, type of die up and down, and target number up and down. It's simply easier to tell a player "roll Xd10 looking for 3 or less, +2 for short range".

If a game requires special dice to play, I won't buy it, though I've played a few and enjoyed them.

This is a serious consideration for publishers. Every time you add a unique requirement such as anything other than d6, a particular basing system, or a gridded playing surface you add another barrier to acceptance of your product.

We try to make things as easy as possible for players: d6s, loose basing requirements, small collection, etc.

Question: how far can die type requirements go before being considered too inconvenient?

IMO:

d6 – no brainer, find your Risk and Monopoly games.

d10 – even some mainstream games use these but still less common than d6…including with some games at BN iirc such as Axis and Allies.

d20 – a little harder to find than d10 but not by much…comes with some D&D games sold at BN.

d4, d8, d12 – faugetaboudit…especially if you need several of each…you're talking special order or trip to game store to spend some serious money.

I've tried them "all." I decided I dislike reaching for various types of dice to complete a combat round. And I really, really dislike "buckets o' dice": 2D6 for combat resolutions, and 1D100 for tests or assigning morale is it for me….

If rolling 1d6 the result is binary: hit or miss. Even if I change the target number from 4+ to 2+ or 5+ the result is still the same: binary, hit or miss.

If rolling, say, 8d6, then the result is most likely to be somewhere in the middle (e.g. 3-5 hits) and less likely to be among the exceptional results (zero hits or 8 hits). And if I change the target number from 4+ to one of the extremes (ie 2+ or 6) then I also shift the expected result to more or fewer hits.

You are right that the outcome you describe first is a binary output space. And the second one is not. So you are not comparing a dice mechanic, you are comparing a fundamental approach; the two systems have different input and output spaces as well as different mechanics.

A comparative example with one die would be use a dX and score one hit above Y, and incremental hits at Z, A, B, and however far you want. By picking the right variables (which are interdependent), you can make the odds for different hits exactly equal to the 8d6 probabilities.

Makes for a more exciting game than (2+, oh look, I rolled a 1).

There's also a visceral or tangible connection between number of dice rolled and the in-game action. Sniper or ATGM shot? One or two dice on a 2+. Quad Gatling Pulse Laser Cannon? Twelve dice barrage at a 5+. The player has a tactile experience that corresponds to the action being represented on the tabletop which is very different from rolling 1d6 whether firing your .50 cal sniper rifle or your Specter Gunship's Autocannon of Death.

Didn't say anything about any of that. In fact, I'm not a fan of the single die roll mechanic. But none of that affects the fact that any distribution of outcomes you can get with multiple die rolls can be represented with exactly the same odds with a single die roll mechanic.

I know the odds, but it requires the player to track multiple means of modification: number of dice up and down, type of die up and down, and target number up and down.

The PITA factor is a different thing than the why would you do this. If you meant that in your original post, sorry I misinterpreted what "what the heck" meant.

I don't mind the different dice mechanic, but it is best suited to games with focused roll spaces, like Shadowrun Duels.

.A comparative example with one die would be use a dX and score one hit above Y, and incremental hits at Z, A, B, and however far you want. By picking the right variables (which are interdependent), you can make the odds for different hits exactly equal to the 8d6 probabilities.

Which illustrates why single die resolution is so limiting. You can only tweak a single variable and you're locked into a range of results based on the dX.

Some games implement a version of your suggestion, even those using multi-die resolution (e.g. "critical success or critical failure" results on any given die).

You're also asking more from the players wrt post-roll math processing which makes for an anticlimactic playing experience. There's a visceral experience "rolling x dice for 5+" and seeing the immediate result versus, say, rolling 3d6 and then adding or subtracting from a given target number and then translating that calculated result to a combat result. The former is instant gratification while the latter feels more like school work.

To put it another, we've all seen cheers go up around a game table when a guy needs to score 3 hits on 3d6 for 6s to finish off that stubborn unit and nails it…with 3x 6s. Can't quite replicate that experience when asking a guy to roll a d20, subtracting the roll from a target number, and then translating the result to some scale or range of values to derive a result.

Which illustrates why single die resolution is so limiting. You can only tweak a single variable

There is no limit to the number of set points you can have and you can put them anywhere you want.

you're locked into a range of results based on the dX.

With multiple dice, unless you allow the count to increase infinitely, have the same type of limit.

You're also asking more from the players wrt post-roll math processing which makes for an anticlimactic playing experience.

There are no modifiers in the system I described.

There's a visceral experience

I didn't say anything about the visceral nature of rolling a fistful of dice. The visceral experience is a different thing than the ability to create a probability distribution, and doesn't affect it.

I don't like D6s with all those 1's 2's and 3's on them.

Yellow dice. I really hate yellow dice.

Thanks to etothepi and Visceral Impact Studios for some great postings, has got me thinking

There is some good discussion about probability curves (distribution) for different dice systems. I have a feeling that some game designers don't understand this, and are more interested in a "fun" mechanism than targeting a specific probability curve.

I do think it is important to make games fun, in particular the non-phasing player should have stuff to do, which in itself justifies things like opposed dice or saving throws even if the same outcome could have come through a simpler mechanism, or it is hard to see what a saving throw might physically represent. I know some people disagree with this.

I don't have a problem with the buckets of dice approach. It really simplifies things to add more dice when charging for instance. However, there is a claim that it is superior because the probability distribution is more center weighted (or lower sigma if you like). I am not convinced that this represents the outcomes of fighting. However, really crazy stuff should be very rare. If you abstract combat (or shooting or what ever) and say that it reduces the effectiveness of the unit (rather than counting bodies) then it would not be uncommon to have a unit rendered useless in the first encounter, or to be encouraged by their success and do even better in the next encounter. So long as I feel the game designer understood how the probabilities work it's all ok to me.

I have played a game where the different types of dice was very well handled. The different types of dice were different colours so it was easy to tell them apart and each unit used the same type of dice for all rolls. I think that if there were different dice for different effects (shooting, combat, morale for instance) then it would be easy to make mistakes

Things I don't like (some already mentioned)

1. Special dice for that game. I'm not cheap, I'm frugal. Ask my ex, she will tell you. To defend saga (I really like that game) they provide downloads so you can make your own.
   
2. Sometimes having to roll high and sometimes low. This confuses me for some reason
   
3. Messed up probabilities or sloppy design. I am a nerd so details bug me. A lot of the time I have difficulty explaining why something annoys me. A flashy fancy system that gives strange results would only get played once ….
   
4. Card driven activation where it could be replaced by simple dice throw. If the card activation adds to the mechanism that is good. For instance you can hold more than one card and play multiple cards when you need a little extra. Cards can help manage if there are many activities that the cards could represent and replace a complex lookup table.
   
5. Mechanisms that make claims that aren't true. No charts for instance …..
   
6. No random effects. To be fair, I've never actually played a wargame like that, but ….
   

Incidentally if you use colour to signify something (entries in a table or different coloured outcome dice for instance), then it fails if you are colour blind. I'm not, but my Dad is, he doesn't go shopping for clothes on his own.

John

Incidentally if you use colour to signify something (entries in a table or different coloured outcome dice for instance), then it fails if you are colour blind.

Good point. I felt very smug about streamlining game play by using color matched dice, charts, range finding sticks, etc)… and then I found out that about a quarter of the gamers I play with have some sort of color blindness. Doh!

- Ix

I recently submitted two articles to Miniature Wargames, discussing various dice procedures. One is about opposed die rolls, the other about buckets of dice. I delve a little bit in the statistical behaviour of these method, and what effect this should have on games design. If they are accepted for publication, you'll find them in one of the next issues.

Anyway, I concur that dice rolling is an integral part of the game and should be treated accordingly. These days, people can order dice of different sizes, colours, etc. For a bucket of dice game, I typically use a collection of dice all of the same color and size. My polyhedral dice are all colored according to type etc. if wargamers spend a fortune on their figures and scenery, spending some money on dice is no big deal, and it adds to the visual appearance.

And please don't give the argument that you prefer D6s because you can find them in your Monopoly set. That argument went out of fashion ever since D&D was produced in the 70s.

And please don't give the argument that you prefer D6s because you can find them in your Monopoly set. That argument went out of fashion ever since D&D was produced in the 70s.

This was also on my earlier grief list. I am not sure how easily available polyhedral dice were in the 70's, but nowadays I can easily stroll to my FLGS and buy D4, D8, D10, D12, D20 and even D30 (not to mention copious variants of D6) in a choice of various colours. Apparently they can also provide D60 as a special order (what game uses those?).

So those are for me pretty trivial to obtain locally, as are variety of paints and other basic supplies. However, for many rules, books and models I am reliant on mail orders from UK.

DoctorG,where are you?

Oh,and,uh,nobody's mentioned D2s yet.

Oh,and,uh,nobody's mentioned D2s yet.

Nothing wrong with flipping a coin.

John