So what's left?

[GbajiTheDeceiver]

It just might be "too good" a fix-and push the curve too far the other way.

I actually thing it likely does. It's probable that there's no real "fix" that's both statistically correct and mathematically simple enough to be usable, so folks will end up picking whichever method seems to them to be the least evil.

Interesting question time:

I note that in the clarification pdf, it's stated that both fail = reroll. Does this change things?

 

[Rurik]

I note that in the clarification pdf, it's stated that both fail = reroll. Does this change things?

I think it is talking only about ties.

 

[GbajiTheDeceiver]

I note that in the clarification pdf, it's stated that both fail = reroll. Does this change things?

I think it is talking only about ties.

Could be, admittedly. The exact quote is:

Q: In an Opposed Skill Test, if both characters fail or if they both succeed and they both roll the same number, what happens?
A: Roll again - the only permissible results are the three listed on page 20.

I don't have page 20 for reference (what do I want? the PDF! when do I want it? now!) so I can't say more on that bit, but the computer programmer in me says that and has precedence over or.

Do we need a clarification of the clarification? :lol:

 

[atgxtg]

It just might be "too good" a fix-and push the curve too far the other way.

I actually thing it likely does.

I fear it does too, except...

THat is how Hero Points work, so we will see the same effect there.

Perhaps I should just scale back the bonus a bit. Instead of getting a automatic second chance, I could hand out a set number of points (say 2 per 100% to cover A&P). If the high skilled character can't win with thosem it goes back to the reguarl halving.

It's probable that there's no real "fix" that's both statistically correct and mathematically simple enough to be usable, so folks will end up picking whichever method seems to them to be the least evil.

I can think of one, but I don't think people will like it. Keep the high skills and put in another skill range category or two to keep all skills over 100% from being "just" a success. Something like 1/2 skill and 1/4 skill would be easy.

This was how RQ could have 101%+ skills before. But is is a deliberate step back to the "old ways"

Heck, I'd actually be happier with just rolling skills as is and let the chips (and inexperienced characters) fall where they might. I really have little problem accepting that a guy with Chess 20% is going to need to roll a series of critical to have a chance against a Grandmaster with 240%.

 

[iamtim]

I really have little problem accepting that a guy with Chess 20% is going to need to roll a series of critical to have a chance against a Grandmaster with 240%.

See, yeah. There's a part of me that thinks, "Dude. Why are we even rolling when one character has 20% and the other has 120%? Fugettaboudid, the 120% wins." I mean, if there's 100% between skills, that's like two different worlds.

But then I realize that there should be a slight chance for the 20%er to do something mind bogglingly good. So I'll probably wind up having them both roll, and if the 20% critis and the 120% fumbles, I'll give it to the 20%er.

 

[GbajiTheDeceiver]

How about each rolls until they fail, halving their skill each time, then count the number of successes (same as number of rolls - 1)? Highest number of successes wins. Or even simpler, first to fail loses. It's more rolls, admittedly, but that doesn't bug me too much.

 

[atgxtg]

UH wait a minute...isn't

How about each rolls until they fail, halving their skill each time, then count the number of successes (same as number of rolls - 1)? Highest number of successes wins."

exactly the same as

Or even simpler, first to fail loses.

:?:


Uh oh I fell myself getting illuminated :idea:

 

[GbajiTheDeceiver]

Yeh, they're the same. :lol:

I was making it up as I went along, and only realised right at the end. Quick jump back to edit.

 

[bluejay]

You could do something simple along the lines of 'if both succeed' then the guy with the higher skill wins or perhaps 'if both fail then the guy with higher skill wins' (not both of course).

It would sway in favour of the higher skill guy quite a bit but I haven't really thought about how much yet.

 

[Rurik]

BlueJay and KA basically pointed out that what matters with opposed rolls is the spread. TaKe for example 60 vs. 40. If both players roll below 40 it is 50/50 who wins. If both players roll above 60 it is 50/50 who wins.

Point 1) Where the higher skill gets the advantage is the range between 40 and 60. If he rolls in that spread, he cannot lose the opposed tests.

Point 2) However, if the higher skill rolls above 60, that range belongs to the lower skilled character - he cannot lose if he rolls between 40 and 60 because of the rule: both fail lowest roll wins.

To compensate for the penalty to the higher skill due to halving take the case in Point 2 away from the lower skilled character. To do this use the following rule: When halving and both sides fail, highest roll wins.

So for normal opposed tests, both fail low roll wins. For Halved opposed tests both sides fail high roll wins.

The effect is that if the real skills involved were 120 and 80, halved to 60 and 40, the lower skilled character cannot win by rolling between 40 and 60 even if the higher skilled character fails.

Make sense?

BlueJay how does this look to you.

I am working out the math but you have a head start on me

 

[atgxtg]

Rurik, isn't it the same thing mathematically?

I mean high win or low win is a coin toss.

 

[Rurik]

Not at all. Both fail, high roll win favors the high skill much more than the low skill.

Look at skill 80 vs 20. With both fail low roll wins, should the high skill character roll above an 80, the low skill character wins on any roll 01-80

With both fail high roll wins, if the high skill character rolls above an 80 the low skill character only wins if he rolls 1-20. If he rolls 21-80 he loses because high roll wins, where in the first example he wins because the low roll wins. Big swing in favor of the higher skill.

And the swing is relative to the spread between skills.

 

[atgxtg]

So


Both roll
Result Favors
01-20 Even Money
21-80 High skilled (autowin)
81-99 also High Skill
00 The one who didn't fumble.



It's sort of wierd, since everything other than a critical is higher better, but I think you are onto something here.

Maybe we can move the critical range? or go with doubles?

 

[bluejay]

Hmm, I tried the 'both fail and high number wins' approach and it didn't seem to help that much unfortunately. Similar problems through the 100% break.

 

[Rurik]

Here are some simplified tables to show what I am talking about.

Say skills are from 1-10, and rolls are made on a d10. Ties are rerolled. The principle is the same as the % opposed rolls.

The first character (column) has a skill of 6, the second character (row) has a 4.

This table shows the results of both fail low roll wins. Note what happens when the skill 4 rolls a 5 or 6 and the High character rolls a 7 or above.

  1 2 3 4 5 6 7 8 9 0
1 - L L L W W W W W W
2 W - L L W W W W W W
3 W W - L W W W W W W
4 W W W - W W W W W W
5 W W W W - W W W W W
6 W W W W W - W W W W
7 L L L L L L - L L L
8 L L L L L L W - L L
9 L L L L L L W W - L
0 L L L L L L W W W -

Now note what happens to that same area when the both fail high roll wins rule is used.

  1 2 3 4 5 6 7 8 9 0
1 - L L L W W W W W W
2 W - L L W W W W W W
3 W W - L W W W W W W
4 W W W - W W W W W W
5 W W W W - W W W W W
6 W W W W W - W W W W
7 L L L L W W - L L L
8 L L L L W W W - L L
9 L L L L W W W W - L
0 L L L L W W W W W -

There are 90 possible results. In the first table the higher skill wins 54 of those possible results (or about 59.4% of the time). In the second table he wins 62 of those results (about 68.2% of the time)

Does that help?

Edit: Fixed table results - they were borked when I first posted.

 

[Rurik]

Hmm, I tried the 'both fail and high number wins' approach and it didn't seem to help that much unfortunately. Similar problems through the 100% break.

The whole idea was use the both fail, low roll wins for opposed rolls that are not halved. Use both fail high roll wins if the roll is halved.

That gives a bump to the higher skilled character to offset the penalty for halving.

 

[Rurik]

So


Both roll
Result Favors
01-20 Even Money
21-80 High skilled (autowin)
81-99 also High Skill
00 The one who didn't fumble.



It's sort of wierd, since everything other than a critical is higher better, but I think you are onto something here.

Maybe we can move the critical range? or go with doubles?

Well, High roll is good in opposed roll no matter what, high roll under skill wins.

Are we sure there are crits in opposed rolls? It goes against the logic of high roll wins. I had thought we ibnored crits for opposed rolls.

I think BlueJays calculator does (ignores crits).

Another question for Matt I suppose.

 

[atgxtg]

Well, High roll is good in opposed roll no matter what, high roll under skill wins.

Are we sure there are crits in opposed rolls? It goes against the logic of high roll wins. I had thought we ibnored crits for opposed rolls.

I think BlueJays calculator does (ignores crits).

Another question for Matt I suppose.

Sure, no. I sure that we aren't sure of anything. But I sort of hope there are criticals outside of combat. THe more they made combat and non-combat skill rolls different the qeasier I feel.

Doubles?

 

[GbajiTheDeceiver]

Doubles break over 100%.

Drop the "ones" digit for your crit chance, I say.

 

[King Amenjar]

See, yeah. There's a part of me that thinks, "Dude. Why are we even rolling when one character has 20% and the other has 120%? Fugettaboudid, the 120% wins." I mean, if there's 100% between skills, that's like two different worlds.

Yes, me too. I think there's far too much rolling for the sake of rolling in just about all RPGs. It would a bizarre and random world we'd live in it worked according to anyone's game mechanics. If I arm wrestle someone from the WWF, it's fairly obvious I'll lose - and lose every time, not just 80% of the time.

But then I realize that there should be a slight chance for the 20%er to do something mind bogglingly good. So I'll probably wind up having them both roll, and if the 20% critis and the 120% fumbles, I'll give it to the 20%er.

Ah, but why should there be? In a very complicated and unpredicatable scenario, like a fight, true, anything can happen. Perhaps this is the problem: when RPGs first evolved out of wargaming, all you did was fight. But what does Mr. 20% think he's up to, exactly, trying to arm wrestle The Rock or beat Gary Kasparov at chess? Of course he'll lose! Why should anyone have a complaint when he does?