Quote from: Old One Eye;520959Using blackjack or margin of success, the character with a skill 94 will have ~80% chance of victory and the character with a skill 33 will have ~20% chance of victory when the contest only requires one roll each.

QuoteFor that, that strikes me as completely insane and unrealistic.

Do you really think for example that John Doe's mom's (rather good, but unprofessional) home cooking has a 20% chance of beating an Iron Chief? Or that your typical, ride a bike a few miles each day John Doe has a 20% chance to defeat Lance Armstrong?

Quote from: gleichman;520963For that, that strikes me as completely insane and unrealistic.

Do you really think for example that John Doe's mom's (rather good, but unprofessional) home cooking has a 20% chance of beating an Iron Chief? Or that your typical, ride a bike a few miles each day John Doe has a 20% chance to defeat Lance Armstrong?

In the RPG context of John Doe's mom and the Iron Chef trying to make the perfect bunt cake while gunfire rattling about and their assistant baker is lying on the ground bleeding his guts out, yes I think giving John Doe's mom a 20% chance is appropriate.

In the safe and controlled environment of a typical kitchen, I already gave you the answer to amp up the Iron Chef's relative chance for victory - require multiple rolls.

Quote from: Old One Eye;520965In the RPG context of John Doe's mom and the Iron Chef trying to make the perfect bunt cake while gunfire rattling about and their assistant baker is lying on the ground bleeding his guts out, yes I think giving John Doe's mom a 20% chance is appropriate.

Are your skills used only in combat? Oddly, in my games combat skills are used in combat, few people are carrying out spitting and cooking contests.

And do you really think that the Iron Chef would lose more of his skiill in combat then John Doe's mom loses of hers? Or that Lance Armstrong wouldn't still beat Joe Doe on a 'get the hell out of line of fire' bike race?

The very concept is silly, first that you bring warfare into a cooking contest and second that you really think that lesser skilled individuals wouldn't suffer even worse if one was to do so.

Quote from: gleichman;520967Are your skills used only in combat? Oddly, in my games combat skills are used in combat, few people are carrying out spitting and cooking contests.

And do you really think that the Iron Chef would lose more of his skiill in combat then John Doe's mom loses of hers? Or that Lance Armstrong wouldn't still beat Joe Doe on a 'get the hell out of line of fire' bike race?

The very concept is silly, first that you bring warfare into a cooking contest and second that you really think that lesser skilled individuals wouldn't suffer even worse if one was to do so.

If you do not want John Doe's mom to have any chance of beating the Iron Chef, there ain't no damn reason to roll in the first place.  Just say the Iron Chef wins unless the player of John Doe's mom cheats or something to sway things.

Here's the situation:  John Doe's mom has 33 and the Iron Chef has 94 on a percentile roll under.  If you want John Doe's mom to have a chance of beating the Iron Chef, tell me what the hell your tolerance is and we can work from there.

Quote from: Old One Eye;520975If you do not want John Doe's mom to have any chance of beating the Iron Chef, there ain't no damn reason to roll in the first place.  Just say the Iron Chef wins unless the player of John Doe's mom cheats or something to sway things.

All other things being equal, I don't want John Doe's mom winning such a contest, it breaks suspension of disbelief (I would accept a less than 1% chance, i.e. the Iron Chef can fumble).

I also don't want to have to overule the game system for something this common to maintain suspension of disbelief. I should be able to play a game and not worry about stopping before every skill test to judge if the possible results are reasonable or not.


As an aside, I believe strange outcomes like these are an artifact of modern game design's rush toward a unified resolution mechanic. It also explains why warfare was brought into a cooking contest- the game mechanic was designed to handle the extremes of battle, and the unified mechanic forced that type of resolution and its assumed randomness everywhere, even into cooking contests.



Edit:
It should also be noted that the Age of Heroes opposed skill method (% = 50 + Player A - Player B) doesn't give John Doe's mom any chance of success. This method increases the importance of a character's skill level.

Quote from: gleichman;520976All other things being equal, I don't want John Doe's mom winning such a contest, it breaks suspension of disbelief (I would accept a less than 1% chance, i.e. the Iron Chef can fumble).

I also don't want to have to overule the game system for something this common to maintain suspension of disbelief. I should be able to play a game and not worry about stopping before every skill test to judge if the possible results are reasonable or not.

Ahhh....it mostly boils down to what appears to be your greater willingness to follow RAW and my greater willingness to make up a rule that matches the heat of the moment.  Cool deal!

Quote from: gleichman;520976Edit:
It should also be noted that the Age of Heroes opposed skill method (% = 50 + Player A - Player B) doesn't give John Doe's mom any chance of success. This method increases the importance of a character's skill level.

There it is.  Problem solved.

Quote from: Old One Eye;520989Ahhh....it mostly boils down to what appears to be your greater willingness to follow RAW and my greater willingness to make up a rule that matches the heat of the moment.  Cool deal!

I wouldn't view it in quite those terms. Rather I'd view it as my requiring good game design, and you being willing to ignore poor game design for GM decisions of the moment.

Marks me as an out style dino in today's world I think.

Quote from: Old One Eye;520989There it is.  Problem solved.

For the few systems that use that method anyway.

Quote from: HombreLoboDomesticado;520946Chance-wise it's the same as comparing margin of success, but without having to perform subtraction. In other words, using margins of success is just making things more difficult than they are, intuitiveness aside.

Just to note, that's true but only with single-die rolls like d20 or d100. For something like GURPS [3d6], Blackjack and margin of success will yield completely different results.
This is just me being pedantic and doesn't really have any pertinence to any of the arguments currently going on, though. Carry on.
 
Blackjack also needs some sort of add-on for characters whose skills exceed (the equivalent of) 100%, to be equivalent.
 

Quote from: gleichman;520976As an aside, I believe strange outcomes like these are an artifact of modern game design's rush toward a unified resolution mechanic. It also explains why warfare was brought into a cooking contest- the game mechanic was designed to handle the extremes of battle, and the unified mechanic forced that type of resolution and its assumed randomness everywhere, even into cooking contests.

Amen.
 

Quote from: gleichman;520952It's not that easy. Here, let me you give you an example.
 
Let's see we have an expert spitting NPC, with a skill level of 94 against six PCs who think they are big shots each with a skill level of only 33. The odds for the NPC are not as good as you think. There's something like a one in three chance that he'll lose as I roughly eyeball the odds.
 
Let's use an extreme example of the possible outcomes to illustrate this.
 
The NPC rolls a rather mediocre roll of 63, making his skill by 31 points. Any of the foes now need an 01 or 02 to beat him.
 
Low chance? Not really. The odds here for all the PCs are .1- 98^6 or a 11.41% chance that one of them will beat him, even needing an 02 or less.
 
The closer the NPC rolls to this skill level, the greater this effect becomes.
 
That's the impact of rolling against so many others.
 
 
Now I don't know about you, but I think the idea of any number of skill 33 beating a skill 94 in a fair individual vs. individual contest an unacceptable simulation under any conditions.

Perhaps depends a bit on how you envisage "skill 94%".
Percentile systems have a tendency to limit skills at around 100% somewhere, but I don't think that's necessarily a good idea when tasks can have difficulty modifiers as well. An Iron Chef cookoff is a difficult affair even for a world-class chef- so wouldn't an Iron Chef be more like 150% ?

Quote from: Bloody Stupid Johnson;521005Perhaps depends a bit on how you envisage "skill 94%".
Percentile systems have a tendency to limit skills at around 100% somewhere, but I don't think that's necessarily a good idea when tasks can have difficulty modifiers as well. An Iron Chef cookoff is a difficult affair even for a world-class chef- so wouldn't an Iron Chef be more like 150% ?

You can indeed increase the allowed skill range as a means to reduce the impact of the die roll.

One needs to be aware of and perhaps adjust the degress of success (if the system includes them) or any other mechanic the examines the exact skill die roll in relation to the skill rating.

It's uncommon and a bit counter-intuitive (until the players get used to the idea that 94 is a good but not amazing level of skill) but workable I think.


As an aside, Age of Heroes goes above 100 in it's skill ratings to represent heroic (or non-human) ability that would put an Iron Chef to shame. But the non-heroic range is capped near 100.

Quote from: The Butcher;520956Wait a sec.

*does math on his head*

Fuck me, you're right. I feel like an idiot.

I'll probably be using it from now on.

I had exactly the same realisation when I first learned about the blackjack method. I'm now actually considering using it even for unopposed die rolls. IIRC Unknown Armies does this.

Quote from: Old One Eye;520960Yep.  Not having to do the substraction step is what I feel makes the blackjack method superior to margin of success.

Especially when calculating margin of success involves relatively high numbers, it's really a lot easier directly comparing the numbers on the dice.

Quote from: Bloody Stupid Johnson;521005Just to note, that's true but only with single-die rolls like d20 or d100. For something like GURPS [3d6], Blackjack and margin of success will yield completely different results.
This is just me being pedantic and doesn't really have any pertinence to any of the arguments currently going on, though. Carry on.
 
Blackjack also needs some sort of add-on for characters whose skills exceed (the equivalent of) 100%, to be equivalent.

All true, and worth mentioning. I simply don't allow skills to exceed 100% in my Cthulhu games, but I realize that's an artificial limit. It's not so uncommon for games to do this, though, and I'm basically lazy and not looking for particularly realistic task resolution. The latter is also the reason why for me it's fine to just compare dice results and get on with it.

Hm, in a 2d6 roll-under-stat + skill system, where stat + skill cannot exceed 13, in what way would you have to modify Blackjack to substitute it for margin-of-success?

Quote from: HombreLoboDomesticado;521157All true, and worth mentioning. I simply don't allow skills to exceed 100% in my Cthulhu games, but I realize that's an artificial limit. It's not so uncommon for games to do this, though, and I'm basically lazy and not looking for particularly realistic task resolution. The latter is also the reason why for me it's fine to just compare dice results and get on with it.

I guess I don't really have a problem with having ratings over 100% since I see it as a good thing if skills are over 100% for some tasks. IMHO that gives a mechanism for defining what is a routine/easy task that doesn't need a roll, without it being entirely GM fiat (or kludged in somehow, like d20s Take-10 rules).
 

QuoteHm, in a 2d6 roll-under-stat + skill system, where stat + skill cannot exceed 13, in what way would you have to modify Blackjack to substitute it for margin-of-success?

Ouch, nasty. You could use blackjack, but the results will be very different. I can't imagine a way of making the blackjack result equivalent in this setup.
 
Calculating actual numbers is very tricky so I’m not sure how to do it mathematically, but if you look at it graphically, the actual result distribution is quite different.
 
 
 
So here we have the probabilities for various rolls on 2d6...
 

 
Basically to see what the blackjack result is for a skill number, cut off the part of the chart above ending at that number. The margin-of-success result is the mirror image of that, but is also (I think) moved up 1 each skill level.
In other words at skill 7 with roll-under, likely results are going to be between 0 and 5, with 0 (minimum success) the most likely; for skill 7 with blackjack, the highest result (7) is the most likely.

It seems safe to say that in such cases blackjack isn't a very practical option. As long as it's only 2d6, margins of success aren't that difficult to calculate quickly anyway.

Quote from: Old One Eye;520959Using blackjack or margin of success, the character with a skill 94 will have ~80% chance of victory and the character with a skill 33 will have ~20% chance of victory when the contest only requires one roll each.

Quote from: gleichman;520963For that, that strikes me as completely insane and unrealistic.

Do you really think for example that John Doe's mom's (rather good, but unprofessional) home cooking has a 20% chance of beating an Iron Chief? Or that your typical, ride a bike a few miles each day John Doe has a 20% chance to defeat Lance Armstrong?

I get what you're saying, but I don't think your examples are adequate to demonstrate your point. The Iron Chef vs. JD's mom, for example, leaves out several factors that prevent it from being a valid example. First, the skill percentages, IMO, do not directly represent one competitor beating another in this particular contest. The chef's 94% and Mom's 33% represent their chances of creating a meal, given average ingredients and no time limit, that will taste "good" to most people. It is perfectly reasonable to believe that both the IC and mom could produce palatable meals at the same time, especially given average ingredients and plenty of time (which is part of the point OOE was making about being under pressure). The problem is, an Iron Chef competition does not use "regular" ingredients, nor do the competitors have plenty of time to cook their entries. So, penalties need to be assigned to each participant's chances due to unusual ingredients and a strict time limit. Where the IC is going to have a near overwhelming edge is in the fact the IC is far more used to working under these conditions than mom. Even though mom is a pretty dang good cook (IMO 33% is fairly good for a non-professional), she is going to be no-where near as used to cooking under these conditions, so her penalties are going to be far more severe than the IC's. This will give the IC a pretty overwhelming margin, but it's still not completely unreasonable that both competitors could produce palatable meals. So their successes (if they both achieve successes) need to be further filtered through each judge's personal biases and tastes if the contest is that close at that point. Also, you list the spitting contest with a fairly good spitter and a bunch of moderately mediocre (but not horrible) spitters. Of course the good spitter has a higher than you'd think chance of losing, he's up against a crew of spitters who, while not great, aren't horrible either. Yet the good spitter is still more likely to win than to lose. I see nothing wrong with those odds. I have also personally known individuals, and once upon a time would have included myself in this list, who would have a not completely unrealistic chance of beating Lance Armstrong in a bike race if the "amateur" was having a great day and Lance was having a bad one. I used to ride hundreds of miles a week and have done centuries myself more than once. The trouble is, Lance (in his heyday), IMO had a better than 94% chance of winning a given bike race, and most cyclists, despite being good "normal" cyclists, would have far less than 33% chance of winning an actual bike race (which isn't anything like just cycling for fun and fitness). Also, Lance tailored his entire training year around the singular goal of winning the Tour every year. The vast majority of time spent every day, all year long, was meticulously designed around training to win one particular race. Unlike Eddy Mercyx, Lance rarely ever won any other races during the year, he only participated to use the race as training for the Tour. So, if racing Lance in Paris-Robaix, a given cyclist would have a much higher chance of beating Lance than they would racing up the Alp d'Huez during the Tour. Especially if the cyclist had trained for it. If the cyclist had not trained to race at all, I would give them only a very small chance of even finishing a bike race let alone coming anywhere near winning it against any pro cyclist. If we're sticking with the 33% vs 94% we're talking about a cyclist who has some racing experience and a Lance that is really nowhere near his peak, so yeah, he could lose.

I don't have any trouble with the BJ system so far, but I am open to being proven wrong. However, that it's bad game design is yet to be demonstrated IMO.

Quote from: Sigmund;521365I get what you're saying, but I don't think your examples are adequate to demonstrate your point.

First, I didn't say it was a formal Iron Chef style event (I just used Iron Cheft to note a world class chef), not that I think it makes any differenw to be honest. Just who made the best meal.

Second opposed rolls are not 'a decent meal', they are by definition who does 'best'. They need no further details or exceptions such as those you're looking for.

Third, I think you're reach to defend a system that can't be defended. You're doing so by reaching for fuzzy defintions and the like. Rather disheartening from my point of view and why I stay away from Internet conversations like these. There are effectively dishonest.

Quote from: gleichman;521369First, I didn't say it was a formal Iron Chef style event (I just used Iron Cheft to note a world class chef), not that I think it makes any differenw to be honest. Just who made the best meal.

Well to be fair, you didn't say it wasn't either. To be honest, with no time limit and no chance of "winning" or "losing", I as a GM would not even have anyone roll for this. While mom's cooking might appeal to her family, and be pretty good for what it is, the superior experience, knowledge, and skill of an Iron Chef would mean he or she would create a masterpiece mom could not match without even having to roll.

QuoteSecond opposed rolls are not 'a decent meal', they are by definition who does 'best'. They need no further details or exceptions such as those you're looking for.

Indeed. Why do you think this specific contest would call for opposed rolls? That's part of my point. The creation of a meal I would rule to be more complex than a single roll would allow, unless the outcome was not that important, in which case who cares? It is not a very good example to use to make your case.

QuoteThird, I think you're reach to defend a system that can't be defended. You're doing so by reaching for fuzzy defintions and the like. Rather disheartening from my point of view and why I stay away from Internet conversations like these. There are effectively dishonest.

You are certainly entitled to think whatever you wish. However, many people seem to use the system and enjoy the experience, and seem to have no trouble with it's outcomes. Your assertion that it is indefensible is what actually seems indefensible to me, because I'm here at this very moment defending the system... apparently successfully at that since you have not yet provided an actual valid example with which to make your case. On top of that, in my post  I allowed that you could be correct and that I remain open to being convinced. I simply took issue with the examples you used to try and "prove" your point, and stated why. I encourage you to point out what you consider to be dishonest about my points. I can assure you, for what it's worth, no dishonesty was intended. Can you say the same? I'm starting to wonder, because you're characterizing my points as dishonest with no apparent evidence to support that claim, and my posting record here demonstrates that I am willing and able to be convinced, will admit when I'm wrong or don't know something, and have never before deliberately argued dishonestly or in bad faith. If the system is actually bad design as you assert, show us. To be honest, it's probably my least favorite subsystem of the system in question, so I am very much open to being convinced. I don't like it and never have, but I have used it and have never found it not to work, I just find it kinda "clunky". It's part of what motivated me to think about my "flipped-over" percentile system to be honest.