Quote from: Steven Mitchell on July 21, 2022, 05:43:54 PM
If the checks are roll under and the effects are roll high (e.g. damage dice), then it discourages players using biased dice.  Works better in systems where it is all the same dice type

Ha, only if the players typically own only one die of the given type. IME it's a pretty rare player that both (1) only has a single d20 and (2) brings their own dice.

Quote from: rytrasmi on July 21, 2022, 02:39:38 PM
Interesting. (I think you were quoting me and not deadDMwalking.)

Yeah, sorry about that. I'm usually squeezing posts in during a short break. I might have even been known to misspell reciprocal.

QuoteI'm struggling with the idea of Case Probability. I recall that a popular statistician was lambasted for predicting Clinton to win over Trump. Of course, he was wrong. His reply was that he only gave Clinton a 70% chance, so he wasn't actually wrong. This kind of thing is impossible to test. In other words, had he said 90% instead of 70% what difference would it have made?

The issue is more of whether or not what we're talking about is an element of a class. If instead of an election it were a lottery with 5 winners drawn and 100 tickets, if I were to hold up an individual ticket and claim it's 5% likely to win, is that an unreasonable claim? If that ticket turns out to be one of the winners, will there be any lambasting? Will anyone be laughing, "Haha, read it and weep, asshole. The ticket you said was 95% likely to lose just won. So suck it!"? If I had said it was 90% likely to win rather than 5%, would I have been more right or less? Would any of it be falsifiable by experience?

When the intellectual heavyweights who studied this drilled down to the heart of the matter, the difference came down to whether you had deterministic causality or non-deterministic causality. Going back to the Prisoners Dilemma, backwards induction is assuming deterministic causality. Because game theory sez defect on the last step, therefore when you're on the second to last step you can rule out future cooperation, therefore no good reason to cooperate now, so on and so forth all the way back to step 1. Non-deterministic causality is the insight that humans can choose to do other than what game theory tells them. Thus the non-zero probability, which in turn changes future expected payoffs and therefore influences the earlier states.

Elections fall under case probability because people choose who to vote for. Even though there's a ton of science behind polling and campaigning, Spend X dollars in Y market with message ABC and net Z votes. That sort of thing. But just because a pattern has held historically does not mean voters are obliged to keep turning out in the same pattern. There will always be some small deviation from the pattern. And some probability p > 0 of a large enough regularity to overturn the results of an election.

Doors from old school style dungeon crawls, unless otherwise noted, are always stuck, never locked, and won't stay open without being wedged or spiked. Those doors are elements of a class. So much so that in AD&D, your strength score actually told you what your probability of opening a door was. Like no one even asked the door how difficult it felt like being. Doors were that standardized. They are definitely elements of a class.

QuoteIt would seem that if you're building a unique Case Probability out of bits and pieces of things that have well-understood probabilities, then the ultimate Case Probability is certainty (i.e., you have total information). Or maybe I'm misunderstanding the concept.

Even though we normally don't think of it this way, probability by definition means our knowledge concerning the content of a proposition is deficient. We do not know everything for which would be required for a definite decision between true and not true. But on the other hand we do know something about it.

Class probability means we know or assume to know, with regard to the problem concerned, everything about the behavior of a whole class of events of phenomena, but about the actual singular events or phenomena we know nothing but that they are elements of this class.

Case probability means we know, with regard to a particular event, some of the factors which determine its outcome, but there are other determining factors about which we know nothing. So in case probability you actually have a greater deficiency in knowledge than in class probability.

Quote from: Mishihari on July 21, 2022, 05:17:56 PM
(a) I don't think dice preferences are entirely irrational.  Some preferences make sense.  Quicker is better than slower.  Less work is better than more.  There are solid reasons to prefer obscuring rather than revealing probabilities or vice versa to support a preferred style of play.

(1)   It's not a big difference, but manipulating two digit numbers is verifiably more time, work, etc than 1 digit numbers.  It doesn't matter to everyone, but for those who do care it's a pretty logical reason

I have a bigger point in all this, but my response to this, perhaps our disagreement, supports my bigger point.

As soon as got to the part where you say "quicker is better than slower" red flags already started going up. And it didn't take long for gut feeling to be vindicated. If you want to say manipulating two digit numbers in general is verifiably more time, that's one thing. But it's not applicable to when you're only manipulating one digit--the 10's digit--because modifiers are multiples of 10. That's not a verifiable truth. Whether you agree or disagree, the point is it's disputed and that there is no objective quicker vs slower that's being compared here. It sounds like a nice, indisputable statement. But it falls apart just like all the other reasons under scrutiny.

Quote(2)   In a percentile pass/fail roll, a 1% modifier will change the result from fail to pass or the other way around once in every 100 rolls, on average.  That's not often enough for me to want to make the slight effort to add it in.  Admittedly, not all percentile systems use modifiers this small, but if, frex, your smallest modifier is 5%, you're better off using a d20.

Here again, you claim to be responding to my points but are not. I had two separate points on this, and one of those points had nothing to do with probability at all. So it's non responsive to come back with a probability argument. The other one had to do with shifts in probabilities that already had a low base chance, like criticals, and the possibility of zeroing them out in some cases. The incidence of criticals are rarely ever statistically significant. But their magnitude when they do come up makes them meaningful. That's kind of the whole point of them. And what I had pointed out is that eliminating them is obviously meaningful.

But now that I'm repeating it, I think criticals by themselves disprove your point. The same gamers who will agree "Less than 5% is insignificant, Just use d20" also like criticals. 3E called for confirming die rolls just to make sure probabilities for getting criticals could shift in units smaller than 5% increments. Something here just doesn't add up. Because it's fake rationale.

Quote(3)   I find roll over systems to be more intuitive because in so many things, more is better.  I'd rather have more money, more cars, more square footage in my house.  Higher is better on a roll is just what I expect.

Non-responsive. I said you have to roll higher than the other guy's fuck shit up skill if you want to dodge his attack. Higher is better. And we preserve his skill rating as being a nice, intuitive percent chance to hit.

So here's my bigger point. If I were to ask for a show of hands, how many gamers out there feel like the d100 system is not your cup of tea. Now of those of you with your hand up, you can put your hand down if you have some reason outside the standard 4 that keep getting repeated. Keep your hand up any one of the four, any combination of the four, or if all four are really what bothers you about d100 systems.

Now suppose I propose we play an RPG where every one of those four concerns are addressed, sort of like I've been explaining. You will never have to perform two-digit addition or subtraction. There will be modifiers that are not multiples of 5 or 10. Small percentages will have a significant impact on the game. You will never know for sure your exact odds of success--that information will never be there for you to break your immersion. And when you roll, higher is better. But it's still going to be a percentile system.

I'd be shocked if 1 out of every dozen gamers who still had their hands up at the beginning of the paragraph would take me up on such a game. Because these reasons are all bullshit. Some people just don't like d100 systems for no good reason. And that's perfectly fine. You yourself addressed my comments point for point but ignored 3 out of 4 of my bullet points entirely. Why? Why would you do that? Because none of this discussion really has anything at all to do with reason.

Just goes to show it really is all about the numbers.

Quote from: hedgehobbit on July 19, 2022, 04:45:05 PM

Quote from: BoxCrayonTales on July 19, 2022, 04:11:30 PMHow familiar are you with BRP? It already addresses the stuff you're bringing up. Or at least the 4e rulebook does. The SRD is crap.

I'm fairly familiar with older versions. Stormbringer was the 2nd campaign I ever ran and my Runequest game was my longest campaign before D&D 3e came out. I understand the BRP already does similar things to what I was talking about, but I find that converting success values into numbered successes, rather than named values, is much more flexible (as you can see in the James Bond 007 RPG). So instead of having a Attack and Defense Matrix like on page 193 of BRP 4e, you just subtract the target's successes from the attacker's. Plus, by having successes listed in numerical values, you can have a list of things to spend those successes on in resolution. For example, 1 success would do damage, but with 2 successes, you could do damage twice, or change the hit location and do damage, or trade both in for a Disarm effect, etc. I prefer systems where you spend successful rolls over system where you have to declare your special action (Called Shots, Disarm, etc) in advance as declaring in advance slows the game down.

For me it is all about speed of resolution and flexibility.

The Pow vs Pow table can work for that too. I forget what page it's on and don't have the book with me at this moment, but it lets you compare the skills of the attacker and defender and reduce it to just one roll.

Honestly, the problem with a lot of these systems is that they always make combat more complex than it needs to be and devote much of their rules to combat. To be entirely honest, I think you can heavily simplify the combat similar to what's been done with True20.

Quote from: BoxCrayonTales on July 22, 2022, 09:21:45 AM

Quote from: hedgehobbit on July 19, 2022, 04:45:05 PM

Quote from: BoxCrayonTales on July 19, 2022, 04:11:30 PMHow familiar are you with BRP? It already addresses the stuff you're bringing up. Or at least the 4e rulebook does. The SRD is crap.

I'm fairly familiar with older versions. Stormbringer was the 2nd campaign I ever ran and my Runequest game was my longest campaign before D&D 3e came out. I understand the BRP already does similar things to what I was talking about, but I find that converting success values into numbered successes, rather than named values, is much more flexible (as you can see in the James Bond 007 RPG). So instead of having a Attack and Defense Matrix like on page 193 of BRP 4e, you just subtract the target's successes from the attacker's. Plus, by having successes listed in numerical values, you can have a list of things to spend those successes on in resolution. For example, 1 success would do damage, but with 2 successes, you could do damage twice, or change the hit location and do damage, or trade both in for a Disarm effect, etc. I prefer systems where you spend successful rolls over system where you have to declare your special action (Called Shots, Disarm, etc) in advance as declaring in advance slows the game down.

For me it is all about speed of resolution and flexibility.

The Pow vs Pow table can work for that too. I forget what page it's on and don't have the book with me at this moment, but it lets you compare the skills of the attacker and defender and reduce it to just one roll.

Honestly, the problem with a lot of these systems is that they always make combat more complex than it needs to be and devote much of their rules to combat. To be entirely honest, I think you can heavily simplify the combat similar to what's been done with True20.

I actually think it was a good move for later editions of CoC to dispense with the contest table. It was definitely a legacy feature from a time when RPGs were frequently needlessly complicated. The benefits of hindsight, eh?