Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

Hey guys i just brought a new set of 18 dice today from the pound shop and thought i should probably test them out to see if they were going to be rolling acccurately. Anyway i was quite surpirised when i saw how much more likely i was to roll a 1. After seeing these results i thought i should test my old dice aswell, and the results also seemed odd, although this time making it better for me. Anyway what i was getting on to was your opinions on this. Are these results much different to other dice (I always assumed they were weighted almost perfectly) or am i just over reacting. Here are the results.

New dice

1.61
21.11%
2.46
15.91%
3.46
15.91%
4.42
14.53%
5.42
14.53%
6.52
17.99%
289

Old dice

1.24
15.58%
2.27
17.53%
3.31
20.13%
4.27
17.53%
5.20
12.98%
6.25
16.23%

154

Unbalanced dice can be predisposed to any side due to internal flaws from the molding process. It is always possible to get dice which roll 1s or dice that roll 6s. As long as you roll them in batches most times the averages even out to no significant impact. People have dicerolling machines which roll dice by the millions to test them.

Just don't buy the hype that somehow they are always predisposed to 1s as that is a myth. For dice to be biased for 29% like the myth they would have to be visible rectangular.

nkelsch wrote:Unbalanced dice can be predisposed to any side

I wish more people recognized this. It's just common sense. I've also seen a study that points out that the most critical factor is a certain dimension being too long, as nkelsch also mentions. I.e., a die that is 1 x 1 x 1.01 or the like. Basically, biasing a dice towards a certain "axis" (1 and 6, 2 and 5, or 3 and 4) rather than a particular single number.

Your "new" dice look to be biased towards the 1 and 6 axis, as those are the two highest percentages. Funny how people only notice that they're rolling more 1's, though, not also more 6's

Human nature: people naturally store more negative memories than positive. That's why most people complain about "always rolling 6's" when rolling for moral tests and then complain about "always rolling 1's and 2's" when shooting.

Yes, dice can be 'messed up' to produce a single side/axis over another, but it is mostly our perception that dictates how "well" some dice roll. This is why I laugh so hard when I see people complain about their dice and trow them across the store (yes, happens many times, just 3 days ago was the last one).


My Honest opinion is to just play with the dice and try to never remember how they roll past the last turn. Less problems for you, less problems for your opponent, and better sportsmanship scores

Waffles

However it is scientifically proven that if the dice colors match your army's color scheme they will roll better

That and painting your ork vehicles red will make them go faster.....

Wait..... That one is true....

thecapn226 wrote:However it is scientifically proven that if the dice colors match your army's color scheme they will roll better

This. Your army won't respect the dice if they aren't in the same color scheme.

You rolled a handful of dice 300 times and think that proves something?
I wrote a quick script that generates 289 random numbers between 1-6 (equivalent of rolling dice). This should approximate an even distribution. These are the results I got:
1: 31 (10.7)
2: 82 (28.3)
3: 61 (21.1)
4: 45 (15.6)
5: 37 (12.8)
6: 33 (11.4)

In short? 289 is not a big enough sample. 2890 is not a big enough sample. As others have said, most of the round-cornered dice made in a tumbler will come out randomly biased towards a certain axis, but over the set of all dice this will bias will also be evenly distributed and thus cancel out.
Injection moulded dice (such as Game Science dice) or Casino dice will roll much more evenly. This might be important when you're rolling only one-two dice in RPG's, but for a game like 40k you really need the ability to roll big fistfulls of dice at once and round corners are much better at this.

Hückleberry wrote:

thecapn226 wrote:However it is scientifically proven that if the dice colors match your army's color scheme they will roll better

This. Your army won't respect the dice if they aren't in the same color scheme.

How true this is. I use black and white for my BTs, red/bronze for khorne models, pink/black for Slaanesh, blue/purple/gold for tzeentch, and green/bronze for nurgle. It pisses my friends off, but it pleases the chaos gods. Guess who I'd rather have on my good side...

timetowaste85 wrote:

Hückleberry wrote:

thecapn226 wrote:However it is scientifically proven that if the dice colors match your army's color scheme they will roll better

This. Your army won't respect the dice if they aren't in the same color scheme.

How true this is. I use black and white for my BTs, red/bronze for khorne models, pink/black for Slaanesh, blue/purple/gold for tzeentch, and green/bronze for nurgle. It pisses my friends off, but it pleases the chaos gods. Guess who I'd rather have on my good side...

Exact opposite for me.

Green for Lizardmen = Slaan blowing himself up every time I used him in 7th edition, apart from the one time where I wanted him to blow himself up (he was in B2B with a KoS) and on 6D6, he successfully cast Flaming Sword of Rhuin on a 29 by rolling one 6, three 5s and two 4s.

Blue for my Eldar = Good rolls sometimes, a lot of horrible ones to go with it. Last game I played I had all but 3 units (WS, FP, Fragons) on the table on turn 2, and the amount of 1s I rolled for hitting and wounding, and the high numbers I rolled for scattering was insane; the only thing I did that turn was sacrifice an empty wave serpent to immobilise a Furioso. In another game, everything possible went wrong.

I seem to have worse lucks with colours that match my army than I do the other way around.

I bought casino quality dice. They're milled not cast and the numbers are filled in with the same density material. They roll perfectly.

Why? Because I'm a Space Marine player. I don't roll a whole lot of dice so they don't really balance out or anything. I wanted good rolls, so I got them.

Trasvi wrote:Injection moulded dice (such as Game Science dice) or Casino dice will roll much more evenly. This might be important when you're rolling only one-two dice in RPG's, but for a game like 40k you really need the ability to roll big fistfulls of dice at once and round corners are much better at this.

Additionally, the casino style dice's hard corners and edges will eventually chip and wear, changing the balance and rolling qualities; this is something that doesn't tend to happen to dice with rounded corners and edges. This wear and tear will affect the results of rolling casino dice unpredictably. This is why casinos replace the dice used in craps tables several times a day; the games are all skewed towards the house with perfect dice, so the last thing they want are dice that are out of true. This wear and tear on the dice will only be increased if/when you roll them with large bunches like you need in GW games.

Considering the increased wear and tear and the fact that casino dice can easily cost more than a dollar apiece (used), you can see how it can be impractical to use casino dice for a game like 40K or WHFB.

Lotus wrote:I bought casino quality dice. They're milled not cast and the numbers are filled in with the same density material. They roll perfectly.

Why? Because I'm a Space Marine player. I don't roll a whole lot of dice so they don't really balance out or anything. I wanted good rolls, so I got them.

Casino Dice don't give you 'better' roles, just average rolls with less deviations. It is very possible for rounded dice to be predisposed to 5s and 6s equally as they would be to 1s. But the casino dice will become quickly damaged by the nature of wargaming and end up taking on random personal bias per dice just like the rounded dice which defeats the purpose of the casino dice.

As long as you are prepared to share your dice with opponents there is no issue. Buying better dice believing you have an advantage then refusing to share the dice means you are wanting an in-game advantage which is not balanced by points.

@sonyca:

Did you roll one die or multiple dice for these studies? If multiple dice, did you record each separate die's results?

Did you bounce each 1 time, 2 times, with or without a backboard? What was the technique employed?

154 and 289 rolls is likely not near enough to draw any conclusions, especially if you were rolling multiple dice. In fact, if you're rolling multiple dice and not recording the results individually, these numbers are meaningless. You could have rolled 10 dice and have 8 dice roll perfectly uniform and have 1 die favor a result of 1 and 1 die favor a result of 3 and never know it.

To me, more balanced is better. I like to play fair. I'm also aware of the risk of edges wearing which is why I only roll them on softer surfaces like plastic and such.

I would love to see all these dice philosophies tested on the Dice-o-matic:

http://www.youtube.com/watch?v=7n8LNxGbZbs

You need to record the rolls of individual dice within the sample. To aggregate the totals takes a lot of the meaning out of the data. For a start, a single die could be skewing the average. You can't determine this from these data and certainly can't defend making a claim about a trend of the whole pool of dice used.

Make sure to use proper rolling techniques. Plenty of people will just pick up dice, shake their hands for a moment, and release them straight down. This is a horrible technique that will result in rolling the same numbers over and over.

You want your dice to have plenty of room to tumble while they're being shaken, and you want them to tumble a few times while they're rolling. If you do that, your randomization should be much better.

And, if you really think a die is badly weighted, there's two techniques you can try. The first is to drop it, gently, into a glass of water. If one side comes up predominantly more than the others, it may be poorly balanced.

The other technique is a bit more complicated, but works much much better. Get a clamp of some sort to hold opposite corners of the die. You want to die to be able to freely spin, but you don't want it to fall out. Now, spin the die. It should come to rest very smoothly without reversing direction. If it wobbles back and forth (as in stops, reverses direction, stops, reverses) as it comes to a rest, you've got a die without good balance.