BRP-Like WHIFF Factor

Best Selling RPGs - Available Now @ DriveThruRPG.com
Mythras (simple version) has bonus/penalty mods of +/-20%, +/- 40%. +/- 80%, and so forth.

The version I am thinking of modifies task difficulties by a proportion of the skill as follows:

Screenshot 2024-04-11 at 18.48.05.png

Now, the way it works in JB007, Classified, and ForeSight is that you generate and record not the character's percentage chance for a standard task, but the equivalent of one fifth of that value— the "Primary Chance". Then a standard task has an Ease Factor of 5; difficult tasks have lower integers, 0.5 or 0.25; easy tasks have higher Ease Factors up to 10. You multiply the Primary Chance by the modified Ease Factor to get the percentage chance of success. So there are twelve degrees of difficulty, each delivering a chance of success in proportion to that of a standard task. And with no arithmetic more taxing than a one-digit multiplication.

So there are 12 levels of difficulty that are susceptible to straightforward arithmetic with addition and subtraction of integer modifiers. And each of them works out to a proportion propability of the probability of success on a basic task. 5%, 10%, 20%, 40%, 60%, 80%, 100%, 120%, 140%, 160%, 180%, and 200% of basic chance, allowing finer modifications than Mythras and simpler arithmetic.

Then a roll less than or equal to one tenth of the chance of success is a Quality Rating 1, less that or equal to two-tenths is QR 2 if not better, less than or equal to half is QR 3 if not better, and a roll of chance-of-success or less is QR 4 (if not better) except if it's an 00. Rolls of 00 are automatic failures. Rolls over chance-for-success are failures¹ (JB007 and Classified provide only a table of the results and conceal the rule that generates it. ForeSight provides both.)

So there are five (in JB007 and Classified) or six¹ (in ForeSight) degrees of success or failure rather than the three provided by Mythras. With correspondingly more arithmetic (or table look-ups), but providing a lively interest in rolling as low as may be even when chance-of-success is well over 100%.

Critical Success is equal to 10% of Skill Value, so I just handwave it by making it the '10s' number (eg: 48% has a crit value of 4%, whereas 72% has a crit value of 7%). Seems easy enough in play.
So long as the chance of success remains below 101 you can use a trick with the last digit: critical success or QR1 on a roll that ends in "0", QR2 on one that ends in "5", QR3 on one that ends in "3", "6", or "9", and QR4 otherwise. But, dealing often in chances that go well over 100%, I prefer to do the arithmetic.



¹ In ForeSight a roll greater than the success chance is QR 10 (a botch) if it ends in 0 and a QR 7 (ordinary failure) otherwise.
 
Last edited:
Well, both Delta Green and I think Raiders of R'lyeh suggest that the GM can easily assign a skill (or even trait) level and only those under it need to roll.

I'm not sure if this isn't in Mythras as well, but I suspect it should be.
CORPS did that - if the difficulty was under your level, you didn't need to roll to succeed.
 
The version I and thinking of modifies task difficulties by a proportion of the skill as follows:

View attachment 80629
I stopped using the default Skill Modifers a few years ago, they just seemed fiddily at the table for me, and I found the Simplified Skill Modifers so much easier.
They were always presented as an Optional Rule until the most recent edition of Mythras Imperative, wherein both ways of calculating modifers are presented equally valid,.
1712830572466.png
So I'm going with the simplier version, because I'm essientially a lazy bugga, heh heh :grin:
 
I think it's probably enough to dial back the overall rolling and be more generous with the "no roll needed". That approach has a downside in a system like the one in the picture above however. If your getting stiff penalties for harder checks and only rolling, generally, for harder checks, the full impact of your skill % isn't going to apply very often.

I don't mind the Mothership approach to this. Skills in MS tend to be pretty low but the game also has a pretty flexible only roll when it really matters approach. What the game also has is an advantage system that encourages the players to work together and make use of the environment and clever plans to get that advantage. So when the players are engaging with the system and diegetic environment they tend to succeed more. I think that's a great bit of design.
 
A failed skill-roll doesn't mean you'll fail miserably - at least not every time . It means you didn't succeed well enough at a most likely non-routine task.
Unfortunately BRP and most derivatives assume that trick to come in the BGG (Bag of Generic Gamemaster-techniques, sold separately, if at all) and don't bother too much with pondering on things like difficulty and consequences.
The attack/parry ping-pong (or Naked Gun-style shootouts) of BRP can go on to the point where a fight is determined by fatigue rather than wounds, but it's not like you're supposed to seek out fair fights after all. Mythras alleviates some of those issues, but at the cost of added crunch. And it can still be a slog if the dice refuse to play ball.
 
Sorry, man, it is what it is...

Also, different legal standards change your odds of a successful legal defense.
Different dice don't. Whether you're rolling head-or-tails, 4 dice in ORE, or 1d100 with 50% TN, it's the same thing, assuming a binary success-fail mechanic.
It's opposed checks and the other mechanics that make it different, but the type of dice? No, not really. What would matter is if the 60-to-80% roll is opposed by a 40-to-50% defense.
Then your odds drop down to 36-to-40%.
While I did address that in other parts of this thread. That wasn't what I was talking about in the post you quoted. I don't care if we are talking about ORE, d100 or anything else. If "professional" skill level is considered to be 50% and the average difficulty is considered to be the default skill level then I have a problem with that from a design standpoint. My solution is not the change how character skill levels are determined with the system. But to rethink what is considered to be the various levels of difficulty.

As for my replies on 50% with 3d6 versus 50% with d100 versus 50% with a dice pool mechanic. My answer remains the same. That 50% odds of success will show up over repeated rolls. For betting on a single roll to see if the player's plans succeed or not at that moment, then a bell curve roll of any type is easier to evaluate as you know most of the results will fall within a narrow range at the top of the curve. Or if we are splitting hairs, the peak of the roof when we are talking about 2DX probabilities.

Yes, over time, 50% is 50%. But players and referees are not looking at the long term but rather making short-term decisions in the moment.
 
While I did address that in other parts of this thread. That wasn't what I was talking about in the post you quoted. I don't care if we are talking about ORE, d100 or anything else. If "professional" skill level is considered to be 50% and the average difficulty is considered to be the default skill level then I have a problem with that from a design standpoint. My solution is not the change how character skill levels are determined with the system. But to rethink what is considered to be the various levels of difficulty.

As for my replies on 50% with 3d6 versus 50% with d100 versus 50% with a dice pool mechanic. My answer remains the same. That 50% odds of success will show up over repeated rolls. For betting on a single roll to see if the player's plans succeed or not at that moment, then a bell curve roll of any type is easier to evaluate as you know most of the results will fall within a narrow range at the top of the curve. Or if we are splitting hairs, the peak of the roof when we are talking about 2DX probabilities.

Yes, over time, 50% is 50%. But players and referees are not looking at the long term but rather making short-term decisions in the moment.
Are you suggesting that it is easier for people to calculate the odds of getting a particular result on 3d6 than it is on d%?
 
No. If you fail the roll, you fail. BRP is happily binary since 1978. From the SRD:
View attachment 80634

Everyone is obviously free to modify that as they see fit, but it is a houserule, not official rule.
Fail can mean many things, and it doesn't say "fail miserably" anywhere in that screen capture.

If the roll is to see if you pass the exam with an A+ and you fail, perhaps that means you merely pass with a B. You have failed to gain an A+, but that doesn't necessarily mean the effort was nothing abject failure.

A failed attack doesn't mean you didn't even manage to swing your sword, it means you have no opportunity to land a telling blow.
 
In other words a simple failure shouldn't be enough for the GM to level significant consequences on your PC in addition to failing at the task.
 
If the roll is to see if you pass the exam with an A+ and you fail, perhaps that means you merely pass with a B. You have failed to gain an A+, but that doesn't necessarily mean the effort was nothing abject failure.
In my 25+ years of experience no GM ever asked for a roll to "pass the exam with A+", while instead just to "pass the exam". But that is anedoctical at best.
A failed attack doesn't mean you didn't even manage to swing your sword, it means you have no opportunity to land a telling blow.
Which is totally equivalent regarding the outcome. You don t inflict damege. Binary result equal to zero. Full stop.
 
In other words a simple failure shouldn't be enough for the GM to level significant consequences on your PC in addition to failing at the task.
Correct, as I already said there is no 'fail forward' in BRP. You simply fail. Binary result equal to zero.
Having the GM adding other elements to a failed roll is houseruling BRP.
 
Having the GM adding other elements to a failed roll is houseruling BRP.
That kind of depends on how that GM is framing the action declarations. A good example is can you pick the lock before the guard rounds the corner. A failure on that action might still result in an open lock because the failure state is based on speed rather than unlocking. Nuanced action declaration and adjudication can add a lot to any game, but especially BRP and other binary games.
 
Are you suggesting that it is easier for people to calculate the odds of getting a particular result on 3d6 than it is on d%?
Sense of the odds, not calculate the odds. You know on 2d6 that most of your rolls will be 6, 7, or 8. You know on 3d6 that most of your rolls will be in the 9 to 12 range. And most of the time the rolls will be in that range due to the nature of the bell curve.

So if you know that you have to roll 8 or less on 3d6 it is probably not good odds. But if you have to roll 13 or less you will be likely to succeed.
 
Are you suggesting that it is easier for people to calculate the odds of getting a particular result on 3d6 than it is on d%?
As for the WHIFF factor, having a sense of the odds combined with the fact that most of the time, the result of the bell curve rolls lies along the peak of the curve (or roof for 2DX) means the result is perceived as less swingy than a 1DX roll. You may still fail 50% of the time (using the example given earlier). But the dice are rolling more as you expected instead of when there is an equal probability of any number appearing.
 
I stopped using the default Skill Modifers a few years ago, they just seemed fiddily at the table for me, and I found the Simplified Skill Modifers so much easier.
They were always presented as an Optional Rule until the most recent edition of Mythras Imperative, wherein both ways of calculating modifers are presented equally valid,.
View attachment 80630
So I'm going with the simplier version, because I'm essientially a lazy bugga, heh heh :grin:
That simpler version seems to assume a skill of 40% when assigning easier tasks' bonuses, and is brutal to anyone under 60, 80, or 100% when it comes to harder tasks.
 
Last edited:
Sense of the odds, not calculate the odds. You know on 2d6 that most of your rolls will be 6, 7, or 8. You know on 3d6 that most of your rolls will be in the 9 to 12 range. And most of the time the rolls will be in that range due to the nature of the bell curve.

So if you know that you have to roll 8 or less on 3d6 it is probably not good odds. But if you have to roll 13 or less you will be likely to succeed.
I think you will find that a lot of people do not really understand how bell curves work in practice, and that the vast majority of people find it much easier to understand what it means to have a 75% chance of success when you need to roll 75 or less on 1d100 than they they do the chance of rolling 9 - 12 on 3d6. While it's true I know the results cluster strongly around the mean, I couldn't quickly give you the actual odds on the 3d6 range (and while I cand find the 2d6 odds fairly easy, many people would be completely lost).

I haven't played a 3d6 game in quite a while, and was surprised when I saw (when I just looked it up) that the chance of rolling 14 or less is around 91%. I was thinking it was more likely to be in the 75-80% range.
 
As for the WHIFF factor, having a sense of the odds combined with the fact that most of the time, the result of the bell curve rolls lies along the peak of the curve (or roof for 2DX) means the result is perceived as less swingy than a 1DX roll. You may still fail 50% of the time (using the example given earlier). But the dice are rolling more as you expected instead of when there is an equal probability of any number appearing.
Oh, I absolutely agree that people perceive skill systems that use 2d6 and especially 3d6 as "less swingy" than those that use a d20 or d%. It's just that I also think the people saying those things typically have no idea what they're actually talking about, and think the perception they have means things that it doesn't.

For the record, I'm guessing you probably have a better grasp of what all these odds mean than I do -- but I think you're underestimating the degree to which many others let their perceptions lead them to utterly false assumptions about what's really going on.

Unless you're simply saying that 3d6 is great for reducing the sense of whiff because people are easily manipulated, in which case perhaps you're right.
 
I absolutely agree that people perceive skill systems that use 2d6 and especially 3d6 as "less swingy" than those that use a d20 or d%.
I might have missed something as I skimmed the above thread, but are we actually trying to argue that 2d6, 3d6, etc don't operate on a bell curve and/or that d20 / d% isn't linear? The so-called swingyness of a system has everything to do with its mathematical distribution. So "perception" isn't remotely what's being referred to in those conversations.
 
I agree with everyone that a curve doesn't cure swinginess by itself, it's also the system designed around it. My opinion is that curves make it easier to design around swinginess because the limited number of options and how they scale quickly up from the lower values makes it easier to work with and that the historical examples back this up. If you make 5- (~28%) or 6- (~42%) on a 2d6 mean "professional" and then make people who don't know the system farm it for bonuses it's still going to feel swingy. Most games I know of that use 2d6 make 7- or 8+ (depending on roll under or high) which is very close to 60% as a baseline and any bonus quickly makes it significantly better. Ditto similar but expanded ranges for 3d6.

No one else mentioned it since we seem to be picking on percentile systems, but I happen to love Savage Worlds. It relies on odd curves and bonuses, but in play most people like the increasing dice and exploding is fun. Savage Worlds is nonetheless swingy in it's own way, but it tends to swing from success to super-success rather than starting at failure. Anyone who has seen a boss or even PC go down to a single REALLY good hit in Savage Worlds knows what I mean (cue discussion of "not using bennies enough or correctly...") I see this as a case of system design pushing the swinginess in favor of success not against it, which tends to annoy players less in my experience.

I'm not an expert but my understanding is that hitting someone with a handgun, particular in a short range firefight that is most common situation, is a lot harder in RL than one may think. I have some relatives who served in their country's required military service and they noted this as well.

Most of the time they'll have a rather low hit rate, because a lot of their fire won't be with the express intention of hitting someone, and when they do they'll be firing bursts or at the very least 'until the target falls over'. I'd suggest SWAT officers or the like, but they seem to go for 'aimed shots at people who are not firing back' or 'massed unaimed fire'. A professional hunter might be your best bet, though they generally work to have lots of favourable circumstance bonuses, and very seldom are 'in combat' when they're shooting.

They probably are "pros" in a setting where Hollywoodisms are the baseline, but I was mostly just noting that cops aren't professional shooters in a realistic setting. The vast majority spend a tiny amount of time training to shoot and almost none in actual retraining. And I agree that a professional hunter is probably a better baseline, but even in a system like Sine Nomine's various ones, I'd have them use 2d6 for shooting animals but 1d20 for shooting in a firefight (and they may have Assassinate, so to with Snipers). Target shooting and hunting are sports not combat skills. I agree or at least think that SWAT and regular military up to Special Forces are where you find the professional combat skills. Also combat is adversarial and I think some amount of adversarial swinginess is a feature not a bug.

CORPS did that - if the difficulty was under your level, you didn't need to roll to succeed.

Corps/CORPS is one of my favorite games. I even ran it a couple times way back when.
 
I might have missed something as I skimmed the above thread, but are we actually trying to argue that 2d6, 3d6, etc don't operate on a bell curve and/or that d20 / d% isn't linear? The so-called swingyness of a system has everything to do with its mathematical distribution. So "perception" isn't remotely what's being referred to in those conversations.
Whether or not a system is swingy depends on a lot more than just the dice used.

Which is swingier?
  • A d100 roll under system where the typical trained skill is around 80%, and modifiers typically range from +10 to -10.
  • A 2d6 system where 8+ is required for success, skill bonuses typically range from 0 to +2, and difficulty modifiers are not generally applied.

Alternatively, how would you compare the following systems?

3d6 roll under:
Very hard: 6-
Hard: 9-
Average: 11-
Easy: 14-

1d100 roll under
Very hard: 10%
Hard: 40%
Average: 60%
Easy: 90%
 
The thing is, GURPS (and any number of other games) often cares about margin of success, and 3d6's less swingy nature matters quite a bit when it comes to MoS (or failure). GURPS' MoS is far, far more predictable than that of Bushido, for example, which uses a d20.
 
That simpler version seems to assume a skill of 40% when assigning easier tasks' bonuses, and is brutal to anyone under 60, 80, or 100% when it comes to harder tasks.
That feels about right to me.
Generally I allocate -20% penalties, so it's a pretty big deal if I allocate a -40% penalty.
I find that Mythras isn't suited for Rollicking Adventure or Heroic Fantasy, but its really good for gritty genres.
The modifers probably reflect this.
 
Last edited:
I understand (though am not personally bothered by) the 'whiff factor', but has to be emphasized that, as was noted above, odds of success in BRP (RQ, etc.) for starting characters are broadly similar to those in D&D and many other games.

I think the more problematic issue in BRP type games is that at higher levels of character competency the attack/parry resolution rules mean you can get into endless 'do loops' where both sides always succeed at attack and defense. Two combatants who are both 95% at their weapon skills have a much lower chance of hurting each other in a given turn than two goofs with 25% skill.
 
Whether or not a system is swingy depends on a lot more than just the dice used.

Which is swingier?
  • A d100 roll under system where the typical trained skill is around 80%, and modifiers typically range from +10 to -10.
  • A 2d6 system where 8+ is required for success, skill bonuses typically range from 0 to +2, and difficulty modifiers are not generally applied.

Alternatively, how would you compare the following systems?

3d6 roll under:
Very hard: 6-
Hard: 9-
Average: 11-
Easy: 14-

1d100 roll under
Very hard: 10%
Hard: 40%
Average: 60%
Easy: 90%
Ok I get what you're going for now. I definitely missed that upthread. That's what happens when you try to quickly skim the thread instead of actually read it.
 
I think the more problematic issue in BRP type games is that at higher levels of character competency the attack/parry resolution rules mean you can get into endless 'do loops' where both sides always succeed at attack and defense. Two combatants who are both 95% at their weapon skills have a much lower chance of hurting each other in a given turn than two goofs with 25% skill.
I think that depends on the exact rules in use for the BRP game, because of the way special and critical successes interact with parries in the various different rules. If critical and special hits bypass parries that aren't at least as high quality(or at least get normal hits), two guys with 95% skill will hit each other almost as often as the two goofs will.
 
At least Bushido has most characters' main skills start in the 10-12 range, and many tasks will be 'long tasks', so you're building up successes, and not just looking at pass/fail.
Hmm, we have yet to do a long task...

Another one that is frustrating in Bushido, there are places you can use one skill to boost another, but you add the margin of success/failure to the other skill. Using a skill with a BCS <= 10 is worse than not using it, and you really want a BCS of like 15 with the supporting skill to be worth it - OR - the situation is so desperate that you need the possibility of a bonus to BCS to have a chance of succeeding.
 
As for the WHIFF factor, having a sense of the odds combined with the fact that most of the time, the result of the bell curve rolls lies along the peak of the curve (or roof for 2DX) means the result is perceived as less swingy than a 1DX roll. You may still fail 50% of the time (using the example given earlier). But the dice are rolling more as you expected instead of when there is an equal probability of any number appearing.
Yea, that's where the feel of 2d6 or 3d6 as being more predictable comes in. It's not the actual percentages, it's that the middle results show up more. Like you say, 50% is 50%. Even on a single roll, it's still 50%.

Hmm, I wonder though if rolling for a 10 or less on 3d6 would actually feel more swingy because you are going to frequently just miss with an 11 (1 in 8 times).

Margin of failure can also play in, most of your failures for 10 or less on 3d6 will just have a small margin of failure. So on the one hand, just missing by one 1 in 8 times may be perceived as bad, the fact that you rarely feel by more than 3 may feel better.

There is so much at play due to the probabilities of various points on the curve, how well the player understands probability, what the meaning of any given roll is, how modifiers work out, etc. that predicting how something will feel to a player may be a lost cause.

I prefer d100 for the simplicity, and Cold Iron's normal distribution for what it does (and the fact that it is simple to determine odds), but 1d20 and 3d6 also work really nicely, and even the dice pool systems I've used (L5R, Burning Wheel, and Dogs in the Vinyard) all have their own charm (and probability tables are available for L5R and I think for Burning Wheel - I could have sworn I saw something in the books).
 
The general result of a failed climb check should be "didn't successfully climb," not "fell to their doom."
The two issues a game must address are when to roll and how to adjucate the results. This is critical. More important than the actual dice convention and probabilities. It can be done implicitly; rules that strongly define success & failure, probability design that doesn't fuck the game if you like to roll dice, etc. It can be done explicitly; guidelines & examples on what to roll for, what is too much or too little rolling, advice & examples on adjucating results. But the game has to do one of those. No fucking off with "if the GM is unsure of automatic success or failure and there are possible consequenves for failure then you should roll" and leaving it at that with a game where an above average athletic trained adult has barely a 50% chance to make the "average" roll for climbing a tree or riding a horse.

Bulletproof your dice system, math, and rolling rules OR give lots of really good strong examples & guidelines. Make sure you define the failure states such that you don't turn the game's Conan the Barbarian, Gandalf the Wise, and James Bond archtypes into members of the Three Stooges.

Assume prople GMing your game don't know how to ride a horse, swim a raging river, free climb a cliff, decipher a coded message, disarm a bomb, pick a lock, perform a play, or give a speech to a hostile crowd. Then set up your system so that you don't make everyone hate playing when the words "um, I don't know, roll for it" come out of the GM's mouth.
 
Oh, I absolutely agree that people perceive skill systems that use 2d6 and especially 3d6 as "less swingy" than those that use a d20 or d%. It's just that I also think the people saying those things typically have no idea what they're actually talking about, and think the perception they have means things that it doesn't.
It's about the feel of the dice roll, so of course, people come up with various reasons why they feel the way they do. And will often take it a step further and try to rationalize it.
For the record, I'm guessing you probably have a better grasp of what all these odds mean than I do -- but I think you're underestimating the degree to which many others let their perceptions lead them to utterly false assumptions about what's really going on.
Since I started self-publishing from 2008 onwards, I have made a concerted effort to run things at game stores and conventions, then later running online campaigns using VTTs to playtest the stuff I write. Prior to that, I ran GURPS for folks whose only experience in RPGs was with D&D. A commonplace situation due to where I live, rural NW Pennsylvania. After 2008, I also ran Fantasy AGE using 3d6, playtested a AGE based sci-fi RPG, and playtested a Fudge-based system.

The difference between d20 and 3d6 is not as pronounced, but it is there and is remarked on from time to time. But even the most math-illiterate players noticed the difference between a d20 and the extreme bell curve of 4DF. They may not have been able to articulate it correctly, but they noticed how many -1 to +1 results they rolled compared to the extremes of -4 to +4, along with how much of a boost even a +1 bonus gives.

How much of an authority does that make me? ¯\_(ツ)_/¯ However, I am not operating from instinct or guesses either.


Unless you're simply saying that 3d6 is great for reducing the sense of whiff because people are easily manipulated, in which case perhaps you're right.
The design of the system counts far more than the dice roll mechanic. Unless the effect is quite pronounced, like with 4DF. In most cases only when all else is equal that you start noticing the differences between 1DX, bell curve rolls like 3d6, and dice pools.

I happen to have noticed it because atypically, I use the same setting for a given genre regardless of the system I am using: GURPS in the Majestic Wilderlands, AGE in the Majestic Wilderlands, Swords & Wizardry in the Majestic Wilderlands, and so on. I have also used a science fiction setting and a superhero setting over the years.
 
Whether or not a system is swingy depends on a lot more than just the dice used.

Which is swingier?
  • A d100 roll under system where the typical trained skill is around 80%, and modifiers typically range from +10 to -10.
  • A 2d6 system where 8+ is required for success, skill bonuses typically range from 0 to +2, and difficulty modifiers are not generally applied.

Alternatively, how would you compare the following systems?

3d6 roll under:
Very hard: 6-
Hard: 9-
Average: 11-
Easy: 14-

1d100 roll under
Very hard: 10%
Hard: 40%
Average: 60%
Easy: 90%
I've rarely seen brp like systems where characters, even after playing for a while, have lots of 80%
 
Banner: The best cosmic horror & Cthulhu Mythos @ DriveThruRPG.com
Back
Top