Quote from: Jam The MF on July 15, 2022, 07:13:11 PMWhy haven't Percentile-Based Systems won out in a big way, in the RPG market?

I would argue that they have won out in a big way in as much as any non-D&D system has. CoC and Runequest are among the few long-standing games in this industry.

Quote from: drayakir on July 18, 2022, 12:45:14 PM
Well, your counterpoint has several flaws. First off, you're just outright dismissing the fact that the d20 and d100 system in their most popular incarnations have an auto-pass and auto-fail number. So no matter how good you are, you have a 1% or 5% chance of just failing, regardless of how good you are. That's asinine. There's no reason that someone who is a grandmaster at riding horses is going to fail mounting a horse in ideal conditions. Just not going to happen. But the d100 and d20 system say that yes, you can have the most ideal conditions imaginable, but every 100 times you will fail every 100 times or every 20 times, depending on which system we're using.

The answer is yes. There are only disadvantages to using a single die over a dice pool.

Who is going to make a player roll just to mount a horse? Are you purposely this dense?

Quote from: drayakir on July 18, 2022, 12:45:14 PM
There's no reason that someone who is a grandmaster at riding horses is going to fail mounting a horse in ideal conditions. Just not going to happen. But the d100 and d20 system say that yes, you can have the most ideal conditions imaginable, but every 100 times you will fail every 100 times or every 20 times, depending on which system we're using.

Do you really roll for that kind of stuff though? I don't call for rolls unless there so some risk of failure and that the failure would have a meaningful consequence (injury, wasting limited time, etc.). Mounting a horse in ideal conditions does not require a roll from a grandmaster nor any other adventure for that matter.

5% is a pretty decent crit/fumble rate when circumstances provide for risk. Also, some d100 systems use a crit/fumble range that changes size depending on your skill. In those systems, a grandmaster would have a 1% chance of fumbling mounting a horse in situations where that action is risky, like the horse is moving, the grandmaster's jumping off a roof, etc.

Long ago I had a thief fumble their backstab roll. One a prone target, who was already wounded. It was stupid. Today I would declare that an automatic hit with no roll.

Reading the 4e BRP rulebook, the percentile system does have multiple difficulty levels and multiple degrees of success. Dice pool systems have no advantage in this regard.

Quote from: Jason Coplen on July 18, 2022, 01:59:37 PM

Quote from: drayakir on July 18, 2022, 12:45:14 PM
Well, your counterpoint has several flaws. First off, you're just outright dismissing the fact that the d20 and d100 system in their most popular incarnations have an auto-pass and auto-fail number. So no matter how good you are, you have a 1% or 5% chance of just failing, regardless of how good you are. That's asinine. There's no reason that someone who is a grandmaster at riding horses is going to fail mounting a horse in ideal conditions. Just not going to happen. But the d100 and d20 system say that yes, you can have the most ideal conditions imaginable, but every 100 times you will fail every 100 times or every 20 times, depending on which system we're using.

The answer is yes. There are only disadvantages to using a single die over a dice pool.

Who is going to make a player roll just to mount a horse? Are you purposely this dense?

It's what si known as an example. I could've come up with any sort of basic skill that a grandmaster would have. For instance, spelling their own name when rolling a Writing check. A d100/d20 system would say that on a nat 1, you fuck that up, somehow. A 5% or 1% chance. Whereas with a dice pool system, the odds aren't 0, but they're sufficiently close where the difference doesn't matter.

It's pointless to argue about the arbitrariness of a 1%/5% chance of failure for some routine task, because good GMs shouldn't be asking for rolls like that. Most games put in print that the expectation is that rolls are for situations where consequences have narrative significance.

Even if you did ask for rolls like that, a good GM has the power to define the scope of the failure, or ask for another roll, or just ignore the failure.

Quote from: drayakir on July 18, 2022, 07:08:24 PM

Quote from: Jason Coplen on July 18, 2022, 01:59:37 PM

Quote from: drayakir on July 18, 2022, 12:45:14 PM
Well, your counterpoint has several flaws. First off, you're just outright dismissing the fact that the d20 and d100 system in their most popular incarnations have an auto-pass and auto-fail number. So no matter how good you are, you have a 1% or 5% chance of just failing, regardless of how good you are. That's asinine. There's no reason that someone who is a grandmaster at riding horses is going to fail mounting a horse in ideal conditions. Just not going to happen. But the d100 and d20 system say that yes, you can have the most ideal conditions imaginable, but every 100 times you will fail every 100 times or every 20 times, depending on which system we're using.

The answer is yes. There are only disadvantages to using a single die over a dice pool.

Who is going to make a player roll just to mount a horse? Are you purposely this dense?

It's what si known as an example. I could've come up with any sort of basic skill that a grandmaster would have. For instance, spelling their own name when rolling a Writing check. A d100/d20 system would say that on a nat 1, you fuck that up, somehow. A 5% or 1% chance. Whereas with a dice pool system, the odds aren't 0, but they're sufficiently close where the difference doesn't matter.

Dude, I don't even entirely disagree with the original point you were trying to make (I don't really care very much and think it's pointless nitpicking, but I wouldn't say you're wrong on principle per se), but this example is BAD, and it isn't even true for ANY system, cuz no system makes you roll for routine pointless tasks. If anything they do the complete opposite and explicitly tell you not to roll unless it matters.

And climbing on top of a horse under completely normal circumstances (when the horse isn't even moving and you're not in the middle of combat) or trying to write your name doesn't matter. And in circumstances where it does matter (you're trying to jump onto a moving mount in the middle of combat), having a 1% or even a 5% chance of automatic failure isn't such a big deal. Things have a way of going south when you're working under pressure IRL all the time.

Quote from: drayakir on July 18, 2022, 07:08:24 PM

Quote from: Jason Coplen on July 18, 2022, 01:59:37 PM

Quote from: drayakir on July 18, 2022, 12:45:14 PM
Well, your counterpoint has several flaws. First off, you're just outright dismissing the fact that the d20 and d100 system in their most popular incarnations have an auto-pass and auto-fail number. So no matter how good you are, you have a 1% or 5% chance of just failing, regardless of how good you are. That's asinine. There's no reason that someone who is a grandmaster at riding horses is going to fail mounting a horse in ideal conditions. Just not going to happen. But the d100 and d20 system say that yes, you can have the most ideal conditions imaginable, but every 100 times you will fail every 100 times or every 20 times, depending on which system we're using.

The answer is yes. There are only disadvantages to using a single die over a dice pool.

Who is going to make a player roll just to mount a horse? Are you purposely this dense?

It's what si known as an example. I could've come up with any sort of basic skill that a grandmaster would have. For instance, spelling their own name when rolling a Writing check. A d100/d20 system would say that on a nat 1, you fuck that up, somehow. A 5% or 1% chance. Whereas with a dice pool system, the odds aren't 0, but they're sufficiently close where the difference doesn't matter.

Fair enough. It seems you're talking basic odds with no DM intervention of sorts. Am I correct in thinking this?

Quote from: Jason Coplen on July 18, 2022, 01:59:37 PM

Quote from: drayakir on July 18, 2022, 12:45:14 PM
Well, your counterpoint has several flaws. First off, you're just outright dismissing the fact that the d20 and d100 system in their most popular incarnations have an auto-pass and auto-fail number. So no matter how good you are, you have a 1% or 5% chance of just failing, regardless of how good you are. That's asinine. There's no reason that someone who is a grandmaster at riding horses is going to fail mounting a horse in ideal conditions. Just not going to happen. But the d100 and d20 system say that yes, you can have the most ideal conditions imaginable, but every 100 times you will fail every 100 times or every 20 times, depending on which system we're using.

The answer is yes. There are only disadvantages to using a single die over a dice pool.

Who is going to make a player roll just to mount a horse? Are you purposely this dense?

Maybe if the PC is playing a city boy, who's never been on a horse before.  It is possible for such a person to fall and bust their tail.

Quote from: Jason Coplen on July 18, 2022, 08:42:30 PM

Quote from: drayakir on July 18, 2022, 07:08:24 PM

Quote from: Jason Coplen on July 18, 2022, 01:59:37 PM

Quote from: drayakir on July 18, 2022, 12:45:14 PM
Well, your counterpoint has several flaws. First off, you're just outright dismissing the fact that the d20 and d100 system in their most popular incarnations have an auto-pass and auto-fail number. So no matter how good you are, you have a 1% or 5% chance of just failing, regardless of how good you are. That's asinine. There's no reason that someone who is a grandmaster at riding horses is going to fail mounting a horse in ideal conditions. Just not going to happen. But the d100 and d20 system say that yes, you can have the most ideal conditions imaginable, but every 100 times you will fail every 100 times or every 20 times, depending on which system we're using.

The answer is yes. There are only disadvantages to using a single die over a dice pool.

Who is going to make a player roll just to mount a horse? Are you purposely this dense?

It's what si known as an example. I could've come up with any sort of basic skill that a grandmaster would have. For instance, spelling their own name when rolling a Writing check. A d100/d20 system would say that on a nat 1, you fuck that up, somehow. A 5% or 1% chance. Whereas with a dice pool system, the odds aren't 0, but they're sufficiently close where the difference doesn't matter.

Fair enough. It seems you're talking basic odds with no DM intervention of sorts. Am I correct in thinking this?

Correct. I am talking probability trends in the resolution mechanic system. I obviously agree that a good GM is going to say "Okay, the person who was literally raised on horse milk, blessed by the Horse Goddess Epona, has been in the saddle since she was three, has tamed hundreds of stallions? She doesn't need to make a roll to mount a horse."

Quote from: drayakir on July 19, 2022, 01:02:43 PM
Correct. I am talking probability trends in the resolution mechanic system. I obviously agree that a good GM is going to say "Okay, the person who was literally raised on horse milk, blessed by the Horse Goddess Epona, has been in the saddle since she was three, has tamed hundreds of stallions? She doesn't need to make a roll to mount a horse."

I think if you present a realistic example, you will find that a 1% or 5% chance of a blunder is reasonable. Also, as Zelen said the scope of the blunder is up to the GM. Suppose it is raining and muddy, so the grandmaster must roll to mount the horse. Oops, it's a 1. So he slips and muddies himself. No big deal. A complete novice rider, on the other hand, might break his arm.

Statistical differences in two reasonably designed systems don't matter in play. Other factors, such as GM discretion and die quality, overshadow statistical differences. Besides, comparing raw percentages is one thing, but a chi-squared or similar measure of significance is another. How many rolls would need to be made, with real dice on a real table, before the difference between two systems is statistically significant? Give me a dice pool setup that models a typical PC in a d20 system, and I will run that test myself.

Quote from: drayakir on July 19, 2022, 01:02:43 PM
Correct. I am talking probability trends in the resolution mechanic system. I obviously agree that a good GM is going to say "Okay, the person who was literally raised on horse milk, blessed by the Horse Goddess Epona, has been in the saddle since she was three, has tamed hundreds of stallions? She doesn't need to make a roll to mount a horse."

What matters for how "swingy" a system rates is a comparison of:

(1) The attribute scale of the system
(2) The variance of the die roll used

I commented on this recently in the thread on low-crunch superhero systems. Olivier Legrand's Crusaders system uses percentile dice, while Steve Kenson's ICONS uses a bell curve 1d6-1d6. However, I think attribute matters more in the Crusaders system. As I put it,

QuoteIn Crusaders, average human is 10, while max attribute can go up to 24. In ICONS, average attribute is 3 while max human is 6 and max superhuman is 10. Even though the die roll is smaller in ICONS, those numbers are very close together. An average human has a fair shot (8%) of beating max human, which is well beyond impossible in Crusaders (average human has only a 5% vs a 19, and 0% against 20 or higher).

If die rolls are over the same range, then a bell curve has lower variance than a linear roll. So, for example, 2d6 has a lower variance than 1d12. However, there are other number ranges at play in a system.

I think C.J. Carella's Cinematic Unisystem and Greg Porter's CORPS are two of the least swingy dice systems, and they both use linear dice (1d10).

Quote from: BoxCrayonTales on July 18, 2022, 04:16:45 PM
Reading the 4e BRP rulebook, the percentile system does have multiple difficulty levels and multiple degrees of success. Dice pool systems have no advantage in this regard.

I don't think BRP went far enough in this area. Instead of just having a skill value and a single special success, they could have had a range of success value.

For example, if a skill value is 110, then it would be written on the character sheet as 110/55/27/13 and a skill value of 45 would be 45/22/11/5

So if you roll the first number or less, it is one success, roll the second number or less for two successes, etc. Now a difficult roll might require a level 2 success to succeed (which is similar to how BRP does it) and if you had a Task, such as picking a lock, that required 3 total successes, you could complete the task with three rolls of 1 success, or one roll of 3 successes.

A contest between two characters would just compare level of success versus each other, so 2 successes would beat 1 success, for instance.

This not only gives value to skills above 100 but it allows more skilled characters to complete tasks not only more often but faster than less skilled characters, again giving value to skills above 100.

That's the way I'd do it.

Quote from: hedgehobbit on July 19, 2022, 04:02:09 PM

Quote from: BoxCrayonTales on July 18, 2022, 04:16:45 PM
Reading the 4e BRP rulebook, the percentile system does have multiple difficulty levels and multiple degrees of success. Dice pool systems have no advantage in this regard.

I don't think BRP went far enough in this area. Instead of just having a skill value and a single special success, they could have had a range of success value.

For example, if a skill value is 110, then it would be written on the character sheet as 110/55/27/13 and a skill value of 45 would be 45/22/11/5

So if you roll the first number or less, it is one success, roll the second number or less for two successes, etc. Now a difficult roll might require a level 2 success to succeed (which is similar to how BRP does it) and if you had a Task, such as picking a lock, that required 3 total successes, you could complete the task with three rolls of 1 success, or one roll of 3 successes.

A contest between two characters would just compare level of success versus each other, so 2 successes would beat 1 success, for instance.

This not only gives value to skills above 100 but it allows more skilled characters to complete tasks not only more often but faster than less skilled characters, again giving value to skills above 100.

That's the way I'd do it.

They have three levels of success: Success, Special and Critical. The SRD omits Critical. You don't need to achieve a particular level of success to achieve an action. Special and Critical just add additional benefits to your success.

Difficulty is applied by modifying the target %. Such as by halving it to represent difficult actions.

Results are already compared in opposed rolls.

How 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.

Quote from: hedgehobbit on July 19, 2022, 04:02:09 PM

Quote from: BoxCrayonTales on July 18, 2022, 04:16:45 PM
Reading the 4e BRP rulebook, the percentile system does have multiple difficulty levels and multiple degrees of success. Dice pool systems have no advantage in this regard.

I don't think BRP went far enough in this area. Instead of just having a skill value and a single special success, they could have had a range of success value.

For example, if a skill value is 110, then it would be written on the character sheet as 110/55/27/13 and a skill value of 45 would be 45/22/11/5

So if you roll the first number or less, it is one success, roll the second number or less for two successes, etc.

hedgehobbit, are you familiar with the James Bond 007 RPG? That is percentile and has four levels of success very similar to what you suggest here. It has a varying multiplier for difficulty, though, which varies with the task.