Quote from: jibbajibba;390496Well in an opposed roll you want the guy with twice the strength to win all the time becuase they are twice as strong. Unless you think that you could outlift an olympic weightlifter 1 time in 3 ?

I believe the idea proposed by the OP would give a different meaning to the numbers. In his system, having a stat twice as high as your opponents doesn't mean you are 'twice as strong' (or whatever the stat in question is), but means literally that you win 2 out of 3 times.

The amount that can be lifted by a character with a Strength of 2 versus a character with a Strength of 1 is NOT represented or described by the OP's system, or at least not by the part of it in question. The only thing represented by the two numbers are their relative odds of success.

So, a Olympic weight lifter should have a far higher strength score (more than double) if he's supposed to beat the average person more than 2 times out of 3.

That means that there needs to be some other mechanic in place determining how much a given character can 'lift' or whatnot.

Well, the OP didn't quite quantify what he meant by "proportional".
Yes, using a mechanic based on the ratio I pulled out, someone 1/99th as strong as you are beats you 1% of the time.
However, the formula may very well be a misrepresentation of his position...it could use further clarification from World.


Proportionality may not go far enough but note that's still better than any number of standard systems. In the d20 Str 1 vs. Str 3 example say. something with triple the lifting capacity only gets a +1 (+5%) to their roll, thanks to how the encumbrance system shifts from exponential to irregular at the lower end.

Of course if you're really only worried about very low stats and you don't want exact proportionality, you could just rejigger the bonus table in d20. I know someone who dumped the standard modifier scale and just dramatically increased the penalties from about 6-down, on the grounds that these were "sub-functional" attributes.

Quote from: Cranewings;390523If a blue belt in brazillian jujitsu wrestles another guy of equal skill who can bench press twice as much, like 200 vs. 400, he will almost never win.

In this system though, stats do not translate linearly into a particular measurement, such as weight lifted: Someone with a strength of 10 doesn't necessarily lift twice as much as someone with a strength of 5.

Quote from: Cranewings;390523Personally, I thing they don't win enough.

If a blue belt in brazillian jujitsu wrestles another guy of equal skill who can bench press twice as much, like 200 vs. 400, he will almost never win.

Someone the can run a 10 minute mile will never out run someone with a 6 minute mile.

This is the entirity of the matter.
I use exactly this theory in my Amber diceless skill system. If I am substantially better than you I will always win. If we are closely matched then other variables come into play.

Sorry for the delayed response. To clarify: what I am searching for is a system in which scores progress in a linear fashion, and two attributes set in opposition to one another will have success ratios which remain constant as long as they are proportionally the same (so a 5 vs. 10 should have the same chance of success as 1 vs. 2).

So, going with the example of Strength (because it's easy to quantify), one point of it refers to a discrete amount of force one can exert. Someone with 10 Strength should, in this scheme, be able to lift twice as much as someone with 5 Strength. Moreover, the chance of the 10 Strength guy beating the 5 Strength guy at a relevant task should be the same as if a 2 Strength guy went against a 1 Strength guy. What that chance is, exactly, isn't what I'm interested in here.

Quote from: FrankTrollman;390280The reason that a strength modifier is not proportional in the way you are thinking it is because there is a base chance of success. No modifier succeeds on a DC 11 50% of the time. If you set Strength to 1 (mod -5), you succeed 25% of the time. If you set Strength to 20 (mod +5), you succeed 75% of the time.

On the other hand, with percentages, there is no base chance to succeed. Set your skill to 1 and you succeed 1% of the time. Set it to 20 and you succeed 20% of the time - twenty times more often if your value is 20 times higher. A percent system, or any roll-under system, is going to give you a linearly proportional increase in success chances as long as you don't throw modifiers into the mix.

I think I follow your reasoning. Is there a way to do that that doesn't screw up when you add modifiers, or possibly a way around that issue?

Quote from: FrankTrollman;390280Dice pools increase in average hits in a linearly proportional manner. But the chances of achieving success at all do not rise in that manner. One die averages 1/3 of a hit and has a 1/3 chance of succeeding at a basic task. Three dice averages 1 hit, but it only has a 21/27 chance of succeeding at a basic task.
-Frank

Perhaps one could sidestep the problem by setting up the system to always be against static numbers, like how some d20 games calculate "Defenses" as 10 plus the relevant modifiers, and roll dice pools against that.

Quote from: The Worid;390799Sorry for the delayed response. To clarify: what I am searching for is a system in which scores progress in a linear fashion, and two attributes set in opposition to one another will have success ratios which remain constant as long as they are proportionally the same (so a 5 vs. 10 should have the same chance of success as 1 vs. 2).

The simplest way to do this is probably to have a roll thats [any dice you want] x your ability score. For example, having both characters lift 3d6 pounds x STR score; compare 'lift numbers' to see who wins in an arm wrestle.

I once used a similar principle trying to get a Marvel Super Heroes variant to work without a chart; characters got a base number equal to 1/5th their stat, then multiplied by d10 and compared to the opponent's attribute as a target number (slightly favours attacker due to average on d10 being 5.5, of course).

Quote from: The Worid;390799I think I follow your reasoning. Is there a way to do that that doesn't screw up when you add modifiers, or possibly a way around that issue?

As long as your modifiers are multipliers or divisors (i.e. on an easy task your roll is doubled, on a hard task your roll is halved) then proportionality is preserved in a percentile system (barring cases where one party's chance of success reaches 100%).  But does maintaining proportionality for easier and harder tasks make sense?  I think that is the better question.

Quote from: Bloody Stupid Johnson;391000The simplest way to do this is probably to have a roll thats [any dice you want] x your ability score. For example, having both characters lift 3d6 pounds x STR score; compare 'lift numbers' to see who wins in an arm wrestle.

I once used a similar principle trying to get a Marvel Super Heroes variant to work without a chart; characters got a base number equal to 1/5th their stat, then multiplied by d10 and compared to the opponent's attribute as a target number (slightly favours attacker due to average on d10 being 5.5, of course).

That may be the best method (by virtue of being the simplest); I've been considering it. Unfortunately, the only published game that I know of that uses it (so that I can see how to fit it into a system) is Maid, which is bad in both the sense that there is only one game to draw information from, and in the sense that it's Maid. :nono:

Quote from: The Worid;391597That may be the best method (by virtue of being the simplest); I've been considering it. Unfortunately, the only published game that I know of that uses it (so that I can see how to fit it into a system) is Maid, which is bad in both the sense that there is only one game to draw information from, and in the sense that it's Maid. :nono:

Bah. Even a stopped clock is right twice a day (Or once)

Quote from: Narf the Mouse;392363Bah. Even a stopped clock is right twice a day (Or once)

Its sometimes surprising where useful ideas turn up. I actually can't think of any other multiplicative systems either (and now I have to try to check out Maid - thanks  )

Another related thought would be, I can think of a couple of systems that work by taking larger numbers and then dividing both by a common factor until it fits on the standard resolution table - Forgotten Futures and JAGS (high-end armour vs. weapon damage). It does get messy in that you have large, weird, probability shifts whenever the divisor changes.


EDIT: not going to Necro the thread for this, but if anyone else likes digging through the archives see also
http://www.therpgsite.com/showthread.php?t=19822
for a 'floating dice' mechanic that would also work to give proportionality.

Quote from: Narf the Mouse;392363Bah. Even a stopped clock is right twice a day (Or once)

True. My general rule is that all games of any significant length (so, 7 page free online RPGs don't count) will have at least one good idea, regardless of how bad the game is.

Quote(and now I have to try to check out Maid - thanks  )

I just found another game based on multiplication: BASH!.