| ▲ | yellowapple a day ago |
| Is soaking the dice even necessary? A naïvely constructed die - i.e. a perfect cube, but with pips dug out for each face - will already bias in favor of 6 rolls and away from 1 rolls simply because six pips require removing more material (and therefore mass) than one pip. Likewise with 5/2 and 4/3. The "precision" dice used in e.g. casinos address this by filling in the pips with material exactly as dense as the die's base material; the injection-molded dice in most board games (let alone wooden dice) obviously ain't constructed with that level of care. This is also part of the reason why some dice games - particularly those typically played with cheap dice - deem 1 to be more valuable than 6 (example: Farkle) or require at least one 1 roll to win (example: 1-4-24). Or they'll require some number of high dice to make the game ever-so-slightly less brutal (example: Ship-Captain-Crew). |
|
| ▲ | burch45 a day ago | parent | next [-] |
| Yes. Very funny for the author to spend a lot of time talking about null hypothesis testing, but not actually running a control experiment to test the null hypothesis that his dice are actually different from the stock dice. |
| |
| ▲ | MathMonkeyMan 18 hours ago | parent | next [-] | | The author's wife could only tolerate so much. | |
| ▲ | brookst 20 hours ago | parent | prev [-] | | I like to believe that Klaus Teuber secretly specified dice biased towards 1-2-3, strictly for personal benefit in games, and the author’s approach has obscured a much greater effect in that direction. | | |
| ▲ | rollcat 9 hours ago | parent [-] | | Catan is played widely enough that many groups tend to introduce their own house rules[1]. E.g. one of my rules is: rolling 2 is the same as rolling 12, so these hexes are slightly less undesirable. I guess if this really matters, you could come up with some set-up rules to swap some 3s with 11s, 4s with 10s, etc, so that biases towards 1-2-3 are less impactful. But that would be painstaking and annoying. I absolutely wouldn't care to do that, unless we track ELO and have an evenly matched group, which is a bit absurd in a casual game. And as in any strategy game FFA, everyone will still gang up against the current leader (even with imperfect information). [1]: https://www.rollc.at/posts/2024-07-01-house-rules/ |
|
|
|
| ▲ | hinkley a day ago | parent | prev | next [-] |
| I remember a retired engineer was selling perfectly balanced dice intended for RPG players. RPG players are going to gravitate toward the most unfair dice they perceive in their set. I appreciate his enthusiasm but he’s only going to sell those dice to competitions and maybe GMs. I think that’s one of the reasons GMs sometimes make a high roll from the player into a punishment. Especially by asking for the roll first and telling what they were looking for after. It’s a way to balance out the consequences of unintentionally loaded dice. |
| |
| ▲ | ultimafan a day ago | parent | next [-] | | About the last point- I remember hearing somewhere, though it could be an urban legend, that that's precisely the reason early Dungeons & Dragons (OD&D/AD&D era) had so many variations within different dice roll mechanics for whether or not a high (or low) roll was good or bad (ie high rolls for your attack was good, high rolls for initiative were bad). If the player used the same dice for all rolls, a balance check against biased or loaded die was therefore built directly into the game, with the perk of making it very obvious if a player was using specific dice for specific rolls | |
| ▲ | vkou a day ago | parent | prev | next [-] | | "20? Your axe cleaves straight through through the orc, decapitating him, and reducing the pillar behind him to rubble!" "You should probably know that this was a load-bearing pillar." | |
| ▲ | gigaflop a day ago | parent | prev | next [-] | | You reminded me, and I think I have some, or at least some variety of these as a set of 10d6. Metal, and the pips are precisely machined to such depths where they're perfectly balanced. Nice bronzed finish, with black pips. Also, if someone is obviously cheating with a loaded die at an RPG game, they're not the kind of player that should be invited back. Most characters have ways of increasing their modifiers to rolls that matter most to them (My current ranger is 1d20 +16 for Perception), and having high-enough base numbers can mean that anything other than a natural 1 is usually some kind of success. | | |
| ▲ | hinkley a day ago | parent [-] | | I would like to have Laura Bailey’s dice checked by an independent party for instance. Her substantial superstitions about good vs bad dice are an example of what I’m talking about above. Lucky dice don’t have to be intentional cheating, but people who have lucky dice are likely cheating in plain sight. |
| |
| ▲ | yellowapple 17 hours ago | parent | prev [-] | | Seems like the solution to that (from a cheater's perspective) would be to bias the dice toward the mid-range numbers, but I don't know enough about D20 layouts to know if that's feasible. |
|
|
| ▲ | geoffpado a day ago | parent | prev | next [-] |
| It looks in the photo at the bottom like that the pips are painted on, not dug out. While that might bias things slightly, I'd expect that the amount of paint used is minimal. |
| |
|
| ▲ | ajsnigrutin 19 hours ago | parent | prev | next [-] |
| I just checked a cheap set of dice in my 100-games-in-1 box, and the 1 dot is both wider and deeper than the 6 dots on the 6, the dots in the 2 also seem a bit deeper than the ones on the 6 side. I don't have any precise instrument to measure them, but it could be that the mold designer accounted for the removed (well.. not cast) material for the dice. |
| |
| ▲ | yellowapple 17 hours ago | parent [-] | | That's indeed a possibility, especially with modern manufacturing techniques allowing sufficiently-precise molds. |
|
|
| ▲ | GavinMcG a day ago | parent | prev [-] |
| Is the bias from having more or fewer pips just as strong as the bias introduced by the water? |
| |
| ▲ | yellowapple a day ago | parent [-] | | Depends on a lot of variables, I'd think: - How dense is the wood? - How much wood does each pip remove? - How much water does the wood absorb per unit of volume? - Are any capillary effects at play transferring absorbed water into the rest of the die? - Is it better to soak the 6 side to take advantage of more surface area? Or the 1 side to take advantage of more soakable volume? - Is the wood even uniformly dense to begin with? |
|
