Last year, Chris Doary debuted some ASL themed precision dice; since then, he has added a few more to the mix. Although I have never been one to be attracted to the precision dice lure, I don’t have much willpower where ASL themed items are concerned. Since he started offering them, I have acquired a number of them and last October I started to use them as my regular ASL dice, replacing the standard MMP dice I had been using forever.
However, I will say that from the very first ASLOK when the first dice appeared, I was concerned that the BattleSchool dice, despite ostensibly being precision dice (i.e., very finely balanced dice), were rolling low an awful lot.
It’s hard to get a true measure of that just from observing the die results of normal ASL play, because the main feature of the Battleshool dice is that the 1-pip face is replaced by an image. As a result, the eye is immediately drawn it, more so than would be the case for an ordinary 1-pip face. And, since you’ll basically be seeing at least one of those dice 11/36 of the time, psychologically, it could be easy to come away with the idea that they were rolling low a lot, especially when some of those 1s will be paired with 2s or 3s, when in fact, the dice were really rolling normally.
I have always considered myself a crappy luck sort of guy when it came to ASL, with a stormcloud hanging over me, but last ASLOK, my dice were overall pretty good (I did get really diced on one scenario). Between ASLOK and Winter Offensive, I only played a few times, but my dice were good in every one of those times. And this year at Winter Offensive, my dice were good a lot, and I really really diced one of my opponents.
All of this cumulatively had me quite worried about my dice. Were the Battleschool dice actually not as balanced as advertised? Could I be inadvertently using dice that are unbalanced in my favor? This really dug into my conscience on the drive home from Winter Offensive yesterday and I decided to test some Battleschool dice.
I pulled out two different Battleschool dice to test, and picked a standard MMP die as well, sort of as a control. I was originally going to do a 4th die, from an altogether different source, but my arm got really tired.
I utilized a chi-square test, with 600 rolls for each die. With that many rolls, you can get a reasonable idea if the die is balanced (obviously, given the nature of dice, you can never be 100% sure).
I was very relieved to discover that both of the Battleschool dice easily passed the test.
Interestingly, the one MMP die I tested badly failed the test and seemed to be unbalanced (and unbalanced the unfavorable way, to boot!). I would not want to be using that die!
Anyway, I have uploaded the results of my testing, if anybody wants to take a look. For those of us who have been wondering about the BattleSchool dice (and it seems that I have not been the only one), this should be somewhat reassuring.
Possibly the reason you noticed in the first place was that your original dice were the flawed ones and now you have a fair playing field. I recenly got some Battleschool dice and they sit alongside me normal precision dice, I’d say they are fine in all honesty.
Since you have used the bad dice for 100 yrs and then moved to balanced dice for a short period of time, it’s going to show a higher incidence of lower dice….meaning the percentages of the lower rolls FINALLY came around..to be normal.
So, the higher incidence is actually “normal”
Good information! Here are additional rolls, divisable by 6, and the SSE, Sum of the Squared Error. If a person wanted to roll 300 times (50 expected results per die face) and if the SSE was below 553.50 you have a 95% expectation your die is fair. I like to run the calculation at 30, rolls, then 90 rolls, 180 rolls and more to see how the SSE is progressing. If it is not getting closer to the maximum SSE then it is likely that the die is far! Haven’t found one die that was fair at 300 rolls get unfair at 600 rolls… just makes one feel better.
ROLLS / Maximum SSE; 030 / 55.35, 060 / 110.70, 090 / 166.05,
120 / 221.40,150 / 276.75, 180 / 332.10, 210 / 387.45, 240 / 442.80,
270 / 498.15,300 / 553.50, 330 / 608.85, 360 / 664.20, 390 / 719.55,
420 / 774.90, 450 / 830.25, 480 / 885.60, 510 / 940.95, 600 / 1107
In regard to your dice test I was wondering how you rolled the dice. Was it in a tower (how many baffles change of direction), in a glass, on a table or in a cardboard box? One player I talked with had even better results (lower SSE) when he used his tower. He noted that how one puts the dice in the top of the tower can have an impact on the results.
Mark Pitcavage says
Yikes, John. I rolled the dice in a dice tower. I have no idea how many baffles it has. I also used my right hand, if that helps. 🙂