d20 vs. 3d6 in HeroQuest?

[buserian]

Mutual fumbles in group contests -- I had forgotten about this. (And the post spoke of them generally, leading me to assume he meant all contests.)

So, yeah, 10 results, depending on the type of contest.

buserian

[Nicolas Crost]

Nick, you're not playing right. Re-read the rules.

Sorry, but I think I am playing it by the book. Ok, my former post was jumbled, I agree.

But just look at those numbers

Code Select Expand


                            10w:10w    19:19     20:15     5w:20    10w:5w
         Complete Victories  0.3%      0.3%      0.3%      1.3%      0.3%
         Major Victories     4.5%      4.5%      5.5%      8.0%      5.8%
         Minor Victories     24.8%     4.5%      21.5%     18.5%     34.8%
         Marginal Victories  18.0%     38.3%     22.8%     22.8%     10.5%
         Ties                5.0%      5.0%      4.0%      3.5%      3.8%
         Marginal Defeats    18.0%     38.3%     36.8%     37.8%     26.8%
         Minor Defeats       24.8%     4.5%      5.5%      8.0%      14.8%
         Major Defeats       4.5%      4.5%      3.5%      0.3%      3.3%
         Complete Defeats    0.3%      0.3%      0.3%      0.0%      0.3%

(I hope those number are correct, I got them from Paul Watson. But by my short double-check they seemed allright)

The first two columns are equal-skill opponents, the last three show an opponent with a skill value 5 points higher than his opponent.
So, what you can see, is that the probabilities change depending on where the skill value is in regard to the range of 1 to 20.

Eg: Looking at two opponets with equal skill you have two different disitributions when the skill is 19 or 10W. You get a lot more Marginal Victories with 19 vs 19 than you get with 10W vs 10W.

Same thing with a person being 5 point better that his opponent. The three examples for a person being 5 points better than his opponent result in three different distributions.
With 5W vs 20 you get a lot more Marginal Victories and less Minor Victories while looking at 10W vs 5W there are half as many Marginal Victories and double the chance of a Minor Victory.

All in all with the HQ Distribution the chances for Minor and Marginal Victories change depending on whether the skill value is in the middle of the 1-20 range or at the borders. This also means that raising a skill by one point is not the same if I do it from 10w to 11w vs raising it from 17w to 18w.

Well, I hope I made it clear that while it is of course curved, there is some inherent problem with the HQ distribution. You might say that it doesn´t matter that the chances for Marginal and Minor Victories and Failure double and halve over the range of 5 skill points (and i would tend to agree, given that HQ is great otherwise), but the problem is still there and I can see people being uncomfortable with it.

Nicolas

[Valamir]

The wonkiness comes from how the masteries cancel out.

Someone who is skill 20 vs 15 is 33% better than their opponent.
Someone who is skill 10 vs 5 is 50% better than their opponent.

The chart shows what I would expect.  That the curve for the 50% better guy is skewed more towards minor victories, while the curve for the 33% better guy tend more towards marginal victories.  Makes perfect sense.

Until you put Masteries into it.

A 10W guy vs a 5W guy is really 30 vs 25 which is only 20% better.
But because the Masteries cancel before the roll, the roll is made 10 vs 5 so you get the statistical effect of being 50% better.

This gets really screwy at the super high masteries.
19W6 vs 1W6 is really 139 vs 121.  The 19W6 guy is only 15% better.
But because the Masteries cancel before the roll, the roll is made 19 vs 1
which is 1800% better.

Whether this is a bad thing or not...but it definitely is a wierd artifact of the mechanic.

Pendragon didn't have this problem because anything above a 20 was just added to the die roll.

[Mike Holmes]

Someone who is skill 20 vs 15 is 33% better than their opponent.
Someone who is skill 10 vs 5 is 50% better than their opponent.

Well, by direct mathematical comparison. But few are the games where such a comparison means anything. You can say in Sorcerer that a 6 is 100% better than a 3, but that doesn't say anything about the probabilities involved.

Are you saying that one should be able to decipher the odds from a glance at the stats? I'm pretty sure that's not your stance because you've come out on the "opaque" side of the "transparency" debate before. So I'm not sure what it is that you're finding problematic here.

A 10W guy vs a 5W guy is really 30 vs 25 which is only 20% better.
But because the Masteries cancel before the roll, the roll is made 10 vs 5 so you get the statistical effect of being 50% better.

"Statistical effect"? What statistical effect? There is none. The actual odds are quire complicated to calculate. There are only two important things, what the actual stat says, and what the odds produce. Which is the problem here?

This gets really screwy at the super high masteries.
19W6 vs 1W6 is really 139 vs 121.  The 19W6 guy is only 15% better.
But because the Masteries cancel before the roll, the roll is made 19 vs 1
which is 1800% better.

It's not 1800% (whatever that means) it's an actual difference of 18 points. This is actually one of the advantages of the system. Instead of keeping odds proportional, HQ makes odds relative. So, yes, if you're almost a mastery below that other character, you have little chance to defeat him. This is precisely the difference between a starting demigod, and an "advanced" demigod (what the 6 mastery level implies).

1W4 Heroes are almost always defeated by 19W4 Heroes. 1W2 Masters are almost always defeated by 19W2 Masters. And the 1W journeyman is almost always defeated by the 19W journeyman. It's just totally appropriate. I love the fact that, unlike many sim games where there's always that ridiculously small chance to succeed, that after you get up by 4 masteries that there's no chance of success. This is the narrativist engine telling you to look for another contest, one that's actually interesting. It means that punks can't beat gods on a good die roll (but if you get the right augments you have a shot).

Basically the masteries are "levels" of a narrativist sort.

Whether this is a bad thing or not...but it definitely is a wierd artifact of the mechanic.

I'm not sure if it's intentional or not, but though it may seem wierd, it's really beneficial in play.

Another advantage of this is that drama will dictate whether or not it makes sense to look for more augments or not. If you're a 1w6 vs a 19W6, then as the higher, you don't have a lot of incentive to augment, but the lower guy can get a lot of milage out of just another +4 in augments. So that incentivizes him to remember that little charm that he picked up in palookastan from the cute girl. In a sim game, the small bonus would mean next to nothing. In HQ, it might mean everything. The system enforces drama. Which is neat.

Mike

[Mike Holmes]

At the risk of following immediately on the post directly above this one...

Well, I hope I made it clear that while it is of course curved, there is some inherent problem with the HQ distribution. You might say that it doesn´t matter that the chances for Marginal and Minor Victories and Failure double and halve over the range of 5 skill points (and i would tend to agree, given that HQ is great otherwise), but the problem is still there and I can see people being uncomfortable with it.

You've made a point, but you haven't said why the effect you mention is bad or a problem. So spending a point here or there is more or less effective - so what? I haven't seen it affect anyone's decisions in play. Nobody bothers min-maxing the distribution, nor is there much to be gained from doing so. Consider that none of this takes into account HP spending, which becomes important at different times precisely because of this variation in the curves.

Most importantly, in HQ play (which I've been doing a lot of lately) the effects of the system are so beneficial in terms of getting players to just go with their guts, that I really wouldn't trade it for anything else that I've seen so far.

Even if the methods mentioned were more "realistic" or something, that just points to the fact that HQ is a dramatic system, not a realistic one. Which is a plus in my book.

Mike

[Valamir]

I didn't say it was bad, I was pointing out where the source of the wonkiness that Nicolas identified came from.

And it is wonky...where wonky here means "doesn't intuitively follow"

It doesn't intuitively follow that someone who is just a hair worse than someone else has an enormous difference in the odds.  You may well be right that its actually a beneficial feature of the system, but it isn't at all intuitive.

consider...2 characters whose skill starts at 13.  Over the course of many many sessions character A buys that skill up to 10W4 and character B buys that skill up to 5W4.

I don't have the actual hero point cost for what it would have cost to get there handy, but it will be pretty substantial.

I would bet however, that character A did not spend twice as many hero points as character B.  Yet in the end he wound up twice as effective.

Character B likely spent very nearly as many points as character A did.  Yet because of the manner in which Masteries cancel, A winds up twice as good...statistically 10W4 vs 5W4 is identical to 10 vs 5 (this is the statistical effect I noted and you queried).


To give an extreme illustration...what this means is that someone who increases their character's ability by 10,000 points, is going to win 95% of the time against someone who increases their characters ability by 9981 points.

No matter how you cut it...that's wonky.

May be good wonky.
But its wonky.

[Mike Holmes]

Well, I've said that from the start. It is different from other systems out there, sure. But as soon as you start to look at the difference in abilities as, well, the difference (subtraction) instead of the proportion (division), you'll see that it's actually quite intuitive.

Most importantly, like Ron says about Sorcerer, more is better, and that's all the player needs to know, really. I think that Ron would agree with me that it's good precisely because it's odd.

Further, again, the things that you point out are irrellevant outside of play. For example, nobody uses abilities unaugmented. Your actual level in an ability is almost irrellevant. In one contest recently Brand had a 5W augment to 5W2 (actually it was more, wasn't it)? So the HP that you spend on an ability is kinda irrellevant to the outcome. What really matters is, overall, how you've managed to keep your character aimed at goals that are important to him and in which he can shine (very TROS SA that).

In the end it all works out great. Again, I think they may have lucked into it. But the more I play the system, the more I notice it doing all manner of things that I wouldn't have thought before playing it a lot.

I'm not saying it can't be outdone or that there aren't ways to improve the system. I'm just pointing out that the things that some people think are strange or damaged in the system are actually of benefit in play.

Mike

[Valamir]

In the end it all works out great. Again, I think they may have lucked into it. But the more I play the system, the more I notice it doing all manner of things that I wouldn't have thought before playing it a lot.

I agree.

If it were necessary to determine, I'd support the "they lucked into it" idea myself.  I suspect that the primary purpose of going to a Masteries-cancel-out system rather than a straight up add-the-difference system like Pendragon was for no deeper reason than to cut down on the need to perform math with larger numbers.

But I may just not be giving two titans of the industry enough credit there.

[RaconteurX]

If it were necessary to determine, I'd support the "they lucked into it" idea myself.  I suspect that the primary purpose of going to a Masteries-cancel-out system rather than a straight up add-the-difference system like Pendragon was for no deeper reason than to cut down on the need to perform math with larger numbers.

If you had a roll-and-add system a la Pendragon, you would have to define a number of success levels beyond critical due to the open-ended nature of contest resolution in HeroQuest (more terms to learn, thus increasing the learning curve) or just tote up abstract "success levels" per twenty points of ability rating and subtract those out at the end anyway ("Six successes to your four means I have an advantage of two... that's a major victory, so your troll is Injured").

I don't think Robin and Greg lucked upon it; the effects were pretty plain from the start. An unstated-but-hinted-at design goal was to discourage the tendency of many people to "play it safe" by calculating their chances in advance and avoiding high-risk activities... decidedly unheroic, and not the least bit mythic. Given the cinematic idiom to which they aspired, the weird probability distribution gave Robin and Greg exactly the effect they desired. Players started to throw caution to Orlanth, take Storm Bull by the horns, dive headfirst into Magasta's Pool, etc. and genuinely embrace the epic possibilities.

One of the first things my playtest group recognized and enjoyed about HQ was that indeed, as Virgil said, fortune favors the bold and abandons the timid.

[newsalor]

The reason HQ might seem illogical is that people tend to think 20W as 40, which it isn't! 20W is 20W is 20W.

Each point in the skill that is a mastery ahead of you is kinda twice as effective. In my campaign it can be seen in the characters Gownar and Turak. Both players have attended an equal amount of gaming sessions. Turak has Sword and Shield Fighting 17W, but Gownars highest ability is Dragon Staff 3W. You might say that Gownar is a more multi-faced or fleshed-out character, but the fact is that Turak is a thane and Gownar is a cottar nowadays. (He has been acting strangely. )

There are some problems that arise from the fact that 1W is't a very good ability if you are going against someone with more masteries than you, but that's what augments are for. Still 1W is better than 20 against a fellow with 15W. Getting into the same mastery level radically lowers the chance of a major defeat.

All in all, I think that masteries are a good bargain. What you lose in minor wonkyness, you gain by making the contests simple even when "high level" characters are going against each other.

BTW, I too wrote a program that calculates all the possibilities in simple contests. ;)

[buserian]

An unstated-but-hinted-at design goal was to discourage the tendency of many people to "play it safe" by calculating their chances in advance and avoiding high-risk activities... decidedly unheroic, and not the least bit mythic. Given the cinematic idiom to which they aspired, the weird probability distribution gave Robin and Greg exactly the effect they desired. Players started to throw caution to Orlanth, take Storm Bull by the horns, dive headfirst into Magasta's Pool, etc. and genuinely embrace the epic possibilities.

If you want some justification for the wonkiness, you can always say that the "laws of physic are exceeding strange (as they are in our world), or that the highest gods have myserious goals. Or maybe just a quirky sense of humor. (Ratslaff: "Look, Acos, I've skewed the probabilities so that Kargan Tor has a 20% chance of a Major Victory this time around, but only a 5% chance of a Minor or Marginal one. Isn't that hilarious?" Acos: "That shall be the rule, er, law." Larnste: "Grumble." Orenoar: "Illogical!" Kargan Tor: "As long as I get any level of victory, I don't care." Harana Ilor: "Can't we all just get along?")

buserian

[Alai]

I have heard HeroQuest variants using 1d12 and 1d10 suggested (each of which leads progressively to greater cinematic play, and even faster hero development)

As the infractor responsible for the original of the former, those weren't really the intent, or indeed the effect as eventually massaged.  Variants on that variant, mind...

To recap, what we play is "on an 11, roll again and bump up", (mm, 12, down) with 1 and 10 being "ordinary" successes /failures, as appropriate.  This doesn't actually make either fumbles or criticals more common, though it does make them slightly more progressive, and it does of course allow for "multiple crits", etc.  Means you always have a finite chance of winning _any_ contest.  Exponentially diminishing, perhaps, but never zero. ;-)

If you give out as many HPs, yes, you'd get more rapid advancement.  What I did instead at first was to give out fewer, but I disliked that as it rather made it "lumpier" instead.  I disliked this, so I introduced yet another brutal hack to retain broader ability progression, but that's another tale...

[Alai]

The wonkiness comes from how the masteries cancel out.

Someone who is skill 20 vs 15 is 33% better than their opponent.
Someone who is skill 10 vs 5 is 50% better than their opponent.

[...]

This gets really screwy at the super high masteries.
19W6 vs 1W6 is really 139 vs 121.  The 19W6 guy is only 15% better.
But because the Masteries cancel before the roll, the roll is made 19 vs 1
which is 1800% better.

I don't think is wonky, and I reckon it's not merely explicable, but entirely intended.  Everything I've heard Greg and Robin say on it indicates that your 10W6 guy (errr, god) is _as much better_ than a 10W5 being, as a 10W chap is better than a 10.  Which makes a great deal of sense, of you want to be able to represent the abilities of Third Idiot on the Left, clan weaponthanes, tribal champions, minor deities, and Forces of the Universe all on the one scale.  (Roughly speaking, having it be logarithmic, if that's not unspeakably mathematical and s*m*l*t**nist a comment to make in these fora.)

Now, there *Is* a small difference in effect from difference of 5 at 10, and at 10, under this account.  Back to crying Emerson!/Narrativism!, as one is wont. ;-)  But seriously, it's pretty minor, and somewhat swings and roundabouts, as one in effect trades off a smaller chance of winning bigger, for the reverse.  Not that I can recall which is which, much less care...

[RaconteurX]

As the infractor responsible for the original of the former, those weren't really the intent...

Actually, Alex, I was referring to a variant which a friend of mine came up with which used the existing rules with no change other than in the type of die rolled. It was motivated by a desire to finally use all the twelve-sided dice that came with the twenty-odd dice sets he purchased over the years. :)

Welcome to the Forge, by the way.

[Alai]

Actually, Alex, I was referring to a variant which a friend of mine came up with which used the existing rules with no change other than in the type of die rolled. It was motivated by a desire to finally use all the twelve-sided dice that came with the twenty-odd dice sets he purchased over the years. :)

Bah, I'll fight him for precedence!  I'll take ye all on!  *self-medicates heavily*  OK, maybe I was just leaping on the chance of re-expounding My Briilliant Idea.  (Official unofficial Stafford reaction -- "You know, that has a certain charm.  Shame they don't make d22s." D'oh!)  I'll admit part of my motivation was pretty similar -- d12s are such nice dice, and most game systems avoid them like they had a rash.  (Anyone remember when the _one_ weapon in RQ2 was "errata'd" from 1d12 to 2d6?  I was aghast, but consoled myself I'd just gotten another 0.5 points of damage on average -- confessions of an adolescent Gamist.)

Don't you get a rather odd-looking ability progression using d12s and no other mods, though?  As in, you'd go from 12 to 1W, etc?  (Or were you rolling d12s against TNs ranging from 1 to 20?)

Welcome to the Forge, by the way.

Thanks.  (According to the software, we seemed to have joined on the same day -- spook!)

Cheers,
Alex.