Opposed Rolls Confusion

[Girandural]

This topic has probably been discussed already, but don't the rules on opposed rolls work counter to the basic principle that the lower the roll the better? The rule says that if both characters succeed in an opposed roll, the one with the higher roll wins - so if I roll a critical I lose! Does this make any sense?

How about:
- If both have an average success they have to roll again (until there is a clear winner)
- critical success beats average success

Girandural

 

[weasel_fierce]

What you propose is exactly how Stormbringer works, and works well. Though SB has crits happening on a fifth of skill instead of tenth. You also, commonly, have skills above 100, even at creation.


The way I usually treat it is that a critical may loose, but you make the bastard pay somehow, or seriously complicate his victory. Alternatively, but a bit confusing, just say a crit beats a regular roll, if you dont have a crit, high roll wins

 

[FatPob]

Just treat a crit as a result of 100

The idea is the higher the skill the better you are, therfore the higher the roll you can make (similar to Pendragon),
Anyone can roll low, so in opposed skills apart from the obvious advantage that you arre more likely to suceed with a high skill, the low roller does seem to be more likely to win if he does suceed as he needs a lower number in the first place

eg,
Bob the Broo Slayer has a stealth of 65 (after -24% skill for armour)
Gustav the Fat Sleepy beer swilling drunk guard has a perception of 17 (-30 for beer swilling drunk fat sleepiness)

Bob rolls a 45, Gustav rolls a 12.
So if we go lower the better, both suceed their rolls and Gustav spots Bob as he had to roll low to suceed.

If you go for higher roll but under skill, bob is clean past gustav on almost every occasion.

If bob rolls a 04 and gustav rolls a 14, bob would be caught, but hold on there, 04 is a crit, so bob gets 100, again clean through, Bob steals the princess's amulet and away he goes...

seems simple to me, plus rewards high skills in opposed checks.

 

[gamesmeister]

From a probability point of view I'm not sure I agree with the 'higher roll wins'

Using my simplistic mathematically challenged brain, if Bob has 10% in a skill and Ernie has 50%, Ernie is five times as skilled as Bob. In other words, using the base skills only, Bob has a 20% chance of winning a competition with Ernie.

But, working on the assumption that both make their skill roll, even if Bob rolls a 10 (the highest possible success), there is still only a 20% chance that Ernie will succeed and still roll less than Bob. This means that Bob has only a 4% chance (roughly) of winning a competition between the two. That seems mighty low to me, given their comparitive skills.

And that's why when I start my campaign in the new year, I'm going to use lowest always wins opposed rolls. Sure, if both succeed in the above example, Bob is likely to win, but the chance of Bob succeeding in the first place is sufficiently small to make the system work.

As mentioned above, crits will still trump a normal success.

 

[Rurik]

From a probability point of view I'm not sure I agree with the 'higher roll wins'

Using my simplistic mathematically challenged brain, if Bob has 10% in a skill and Ernie has 50%, Ernie is five times as skilled as Bob. In other words, using the base skills only, Bob has a 20% chance of winning a competition with Ernie.

But, working on the assumption that both make their skill roll, even if Bob rolls a 10 (the highest possible success), there is still only a 20% chance that Ernie will succeed and still roll less than Bob. This means that Bob has only a 4% chance (roughly) of winning a competition between the two. That seems mighty low to me, given their comparitive skills.

And that's why when I start my campaign in the new year, I'm going to use lowest always wins opposed rolls. Sure, if both succeed in the above example, Bob is likely to win, but the chance of Bob succeeding in the first place is sufficiently small to make the system work.

As mentioned above, crits will still trump a normal success.

The math behind opposed rolls doesn't quite work the way you describe. It was covered at length months back, but High Roll wins is definately better than Low Roll Wins in opposed tests. The odds favor the lower skilled character when you go Low Roll Wins, which is obviously undesireable.

In fact, using High Roll wins if both succeed OR both lose has better results than High Roll Wins if both Suceed and Low Roll Wins if both fail (in that it favors the higher skilled character).

If I find the time later I will try to post an example that can explain this relatively simply.

 

[canology]

I will continue to say that the amount you succeeed by should determine who wins. That way low rolls are good, and guys with higher skill levels will win more often.

It seems the most sensible and easy to use system, at least to me...

I hope that simple subtraction wouldn't be enough to scare people away from it... :roll:

 

[gamesmeister]

I will continue to say that the amount you succeeed by should determine who wins. That way low rolls are good, and guys with higher skill levels will win more often.

It seems the most sensible and easy to use system, at least to me...

I agree this works too, and the subtraction isn't a problem, but I just prefer a simple comparison in my old age...

 

[gamesmeister]

The math behind opposed rolls doesn't quite work the way you describe. It was covered at length months back, but High Roll wins is definately better than Low Roll Wins in opposed tests. The odds favor the lower skilled character when you go Low Roll Wins, which is obviously undesireable.

In fact, using High Roll wins if both succeed OR both lose has better results than High Roll Wins if both Suceed and Low Roll Wins if both fail (in that it favors the higher skilled character).

Yeah, I've just been playing around with the numbers here, and it seems that both systems seem to skew the numbers too far in one direction or the other.

Canology's system works the best, but a lot of people won't like the subtraction - perhaps we should all go back to the RQ2 method of skill increments of 5%

 

[simonh]

I will continue to say that the amount you succeeed by should determine who wins. That way low rolls are good, and guys with higher skill levels will win more often.

Sure but it's statistically identical to highest roll wins, you're just counting up from 1 instead of down from you're skill number, but with a superfluous mathematical step before you can compare rolls.

 

[elgrin]

I will continue to say that the amount you succeeed by should determine who wins. That way low rolls are good, and guys with higher skill levels will win more often.

It seems the most sensible and easy to use system, at least to me...

I hope that simple subtraction wouldn't be enough to scare people away from it... :roll:

Having thought about this after the last airing, I came to a similar way of thinking to Canology. For me it is nothing logical to do with the mathematics, simply it looks odd to have the higher roll winning. So I will be using 'whoever rolls furthest below their skill wins', but with a critical for the lower skilled character giving a chance of a less disastrous outcome.
(Of course for an Old Crusty like me, the RQ2 resistance table may be kept at hand - just in case!)

 

[duncan_disorderly]

It was covered at length months back, but High Roll wins is definately better than Low Roll Wins in opposed tests. The odds favor the lower skilled character when you go Low Roll Wins, which is obviously undesireable.

.

This depends on whether you are the lower skilled character or not...

 

[Rurik]

It was covered at length months back, but High Roll wins is definately better than Low Roll Wins in opposed tests. The odds favor the lower skilled character when you go Low Roll Wins, which is obviously undesireable.

.

This depends on whether you are the lower skilled character or not...

I am the higher skilled character.

 

[weasel_fierce]

Im definately keeping the resistance table handy. Its worked well for years, no reason to throw it out

 

[soltakss]

I will continue to say that the amount you succeeed by should determine who wins. That way low rolls are good, and guys with higher skill levels will win more often.

Sure but it's statistically identical to highest roll wins, you're just counting up from 1 instead of down from you're skill number, but with a superfluous mathematical step before you can compare rolls.

I oiriginally thought that "succeeded by most" was the best way to do it, but when I ran my comparisons, it worked out to be exactly the same as Highest Roll wins. So, even though I hate the idea of trying to roll low and trying to roll high at the same time, Highest Roll Wins is the easiest method.

Since I am a firm believer in being able to influence dice rolls just by wanting to roll low, I'll use the "Succeeds by Most" method as a low roll is always good.

 

[canology]

Another advantage that the "whoever succeeds by the most wins" system, is that you don't have to worry about modifying skills over 100%. In that way I think it is superior to the "highest roll wins" (even if they are statistically the same).

The only caveat that I would add is that critical successes trump regular successes.

But with skills over 100% you just roll normally and figure how much they made the roll by.

I imagine that might also affect the statistical "mirror-image" with the highest roll wins system.

If this made no sense it's because the caffeine hasn't kicked in yet :roll:

 

[Halfbat]

Another advantage that the "whoever succeeds by the most wins" system, is that you don't have to worry about modifying skills over 100%. In that way I think it is superior to the "highest roll wins" (even if they are statistically the same).

And that is a very good point.... ::thinks::

The only caveat that I would add is that critical successes trump regular successes.

That is the way I run it as well, atm: multi-staged. I find it adds something to the game by being able to have a 'trump' like approach, even if it might throw the calculations again. It's also another way of making high skills more useful.

 

[bluejay]

Yup, as long as criticals aren't important then there's no statistical problem with highest roll wins...

Feel free to check with my calculator.

 

[Rurik]

Out of curiosity what mechanic are you using with the "makes roll by most" mechanic when both parties fail? Fails by least?

A 'feature' of the Mongoose system as written I have come to like is that in opposed rolls is that a player has very little idea of how they did. This appeals to me as a GM in a couple of ways. You can have the players roll dice at random (the 'phantom' perception test) and not really know how well they rolled compared to you. The old alternatives are to have players roll themselves and pretty much know if they succeeded or to roll in secret, which takes away the feeling of players controlling their own destiny.

I plan on using opposed rolls a lot. Hidden objects are going to have a difficulty, so when a player searches a room for example he can roll and you can roll and he will not be much the wiser as to how he did. Also very useful for when you have to 'fudge' results, such as faced with the 'certain roll must be made to continue scenario plot' situation.

 

[canology]

Out of curiosity what mechanic are you using with the "makes roll by most" mechanic when both parties fail? Fails by least?

Yeah, in situations where one person *has* to win I would probably go with who failed by the least. Probably describing how poorly both parties performed in the contest.

A 'feature' of the Mongoose system as written I have come to like is that in opposed rolls is that a player has very little idea of how they did. This appeals to me as a GM in a couple of ways. You can have the players roll dice at random (the 'phantom' perception test) and not really know how well they rolled compared to you. The old alternatives are to have players roll themselves and pretty much know if they succeeded or to roll in secret, which takes away the feeling of players controlling their own destiny.

I understand what you mean here, but won't players know at least whether the succeeded or not? In the RAW you still have to roll under your skill to succeed, so players would know whether they had done that or not, wouldn't they?

I plan on using opposed rolls a lot. Hidden objects are going to have a difficulty, so when a player searches a room for example he can roll and you can roll and he will not be much the wiser as to how he did. Also very useful for when you have to 'fudge' results, such as faced with the 'certain roll must be made to continue scenario plot' situation.

But can't you do that when using low rolls? Just have the players roll Perception and you know that they need to make the roll by at least (for example) 20%, if they don't make it by that amount they don't see the hidden item. Seems easy, but perhaps I am misunderstanding what you mean.

 

[Rurik]

An example is as follows. Character has a 60% Perception. You have an object hidden with 50% difficulty to spot. He rolls a 10 - not really a good roll, if my roll is 11-50 he fails. Even if he rolls a 60 he knows he did as good as he can, but maybe he is looking for a concealed object with a difficulty of 80, and I still rolled better - or maybe there is nothing to find. His roll doesn't tell him much.

And if they fail their roll they don't know automatically that they failed the test. Very useful for scenarios for finding that one secret passage is necessary to advance the plot (but that is a separate subject )

Now look at a straight skill roll. He makes a perception test and rolls a 10. He knows he did pretty well and that there was likely nothing to find (unless it had a difficulty modifier of more than -50, in which case he was pretty much doomed).

So in the past, when the characters are camping and one player is on watch, your choice is to let them roll and possibly know how well they did, or roll in secret, which takes the dice out of the players hand, which is less fun for the players IMHO.

There are exceptions, if your skill is 95 and you roll a 95 you can pretty sure you rolled well, but in most common cases you can have players roll and not know how they did.