Opposed Rolls and Success Levels

[Atgxtg]

This idea sprang up from the back & forth discussion on oppsed rolls in the BRP combat thread. As I've said, I don't like the idea of the opposed high (blackjack) game mechanic being used in conjunction with the old RQ roll low success levels. Now, assuming for the sake of argument that there is some truth to the idea that it takes two to fight, and that how well the attack does is at least partially dependant on how well the defender does then...

...why not set the success level by the difference between the two rolls?

Instead of basing the chance for a special, critical, fumble and whatnot by each characters individual success chance, we could figure out the success levels byt how much the winner beats his opponent by. A typical close rolls (what would be success vs success on the BRP combat matrix) would give the typical BRP parried result from the matrix. Each 20% or difference would shift the success level by 1. The exact progression can be hammered out later, I'm just considering the concept of using the margin for the result. A failed roll would be read as zero, and the difference would be the winner's roll.

It would simply the rules, by eliminating the trump effect of the success levels, and the need to figure out the % chances for specials, critical and fumbles.

Reactions? Thoughts?

[Zit]

Let's see... two 40%-level opponents fight each other. They can have only a maximum of 40 difference, i.e. 2 levels of success. Two guys at 80%, they can have up to 4 levels of success difference = twice as much as the 40% guys. This means that it is forbiden for a 40% guy to have more that 40 difference with whatever opponent, even against the dullest beginner with 05% skill rating, while two experienced fighters can win against each other with 80 difference. I'm not sure this realy takes into account how both opponents are relatively skilled.

[Atgxtg]

Let's see... two 40%-level opponents fight each other. They can have only a maximum of 40 difference, i.e. 2 levels of success. Two guys at 80%, they can have up to 4 levels of success difference = twice as much as the 40% guys. This means that it is forbiden for a 40% guy to have more that 40 difference with whatever opponent, even against the dullest beginner with 05% skill rating, while two experienced fighters can win against each other with 80 difference. I'm not sure this realy takes into account how both opponents are relatively skilled.

Yes, but we can tweak things a little.

The first tweak would probably be to reduce the margin needed for a SL. I need to crunch things a bit to try and get the best match to the table. Maybe 10% brackets would work. THat would give the 50% guy a chance at 5 SLs. Especially since getting a big difference is effect is hard against a skilled foe. Worse case scenario is about a 95% vs a 5%, and even that would average about 5 SLs vs. the 2SLs it would have now.

Another possibility would be to allow characters to up their difficulty to bump up their SL. Another would be to go with higher starting chances.

THe big difference between BRP and the oppsed methods used in other RPGs is that BRP die results are limited by the D100 while most other systems use an open method (Die+skill).

[Philotomy]

As I've said, I don't like the idea of the opposed high (blackjack) game mechanic being used in conjunction with the old RQ roll low success levels.

What is your criticism of the "blackjack" mechanism? (I know you've probably said that in the other thread, but it's a lot to dig through.)

[Zit]

btw, what do you all mean with "Blackjack" ? I only know the Belotte and, thanks Gianni, the Scopa .

[Atgxtg]

What is your criticism of the "blackjack" mechanism? (I know you've probably said that in the other thread, but it's a lot to dig through.)

I don't like uising it with RQ/BRP's normal roll-low success levels. For example somebody with a 60% skill faces off against somebody else with a 50% skill.

-If the first guy rolls a 15 and the second guy a 14 the first guy wins

-If the first guy rolls a 13 and the seond guy s 14, the second wins

-But if the first guy rolls a 12 and the second one rolls a 14, the first guy wins!

[Atgxtg]

btw, what do you all mean with "Blackjack" ? I only know the Belotte and, thanks Gianni, the Scopa .

The one who rolls hightest without going over (roiling over his skill). For example, a guy with a 60% skill wants to roll as close to 60 as possible without rolling over 60 (which would be a failure). People call it the blackjack method becuase it is similar to blackjackm where you want to roll as close to 21 as possible without going over (busting).

[Pete Nash]

The best way of thinking about the Blackjack method is this: Whoever gets the best level of success wins, but in case of a tie the highest roll wins. It helps make the numbers fade into the background.

[Philotomy]

I don't like uising it with RQ/BRP's normal roll-low success levels.

Ah, okay. You want consistency in the rolling; for example, if rolling low is good, then you'd like a lower roll to always be the better roll (if the degrees of success are equal).

I get your objection, and I understand why it bothers you. I can't say that it bothers me, though. I find the "blackjack" approach to be easy to remember and apply, even if it is not as aesthetically nice as "rolling low is always better." (Actually, the discussion reminds me of similar objections to the way surprise rolls work in 1e AD&D -- it's usually good to roll high, unless your roll falls within the range where you're surprised, in which case it's better to roll low.)

Anyway, I don't want to steer the thread away from discussing different approaches and into an argument about whether a different approach is needed; everybody has their own preferences. I was just curious about what your objections were.

[Atgxtg]

Ah, okay. You want consistency in the rolling; for example, if rolling low is good, then you'd like a lower roll to always be the better roll (if the degrees of success are equal).

Yup. plus it goes over easier when teaching the game to new players. Using my example above, it is a hard sell to a player than if he rolled a 12 he would win, a 14 he would tie and a 15 he would win, but he looses on a 13.

Every other opposed game mechanic uses high is always better or low is always better.

I get your objection, and I understand why it bothers you. I can't say that it bothers me, though. I find the "blackjack" approach to be easy to remember and apply, even if it is not as aesthetically nice as "rolling low is always better." (Actually, the discussion reminds me of similar objections to the way surprise rolls work in 1e AD&D -- it's usually good to roll high, unless your roll falls within the range where you're surprised, in which case it's better to roll low.)

There is a beaut in !E AD&D with weapon speeds and initiative. Fast Weapons, can attack twice if the initiative roll is a tie, but only once if they win!.

Anyway, I don't want to steer the thread away from discussing different approaches and into an argument about whether a different approach is needed; everybody has their own preferences. I was just curious about what your objections were.

It was a fair question. Even a fair argumnent if you want to start a thread for it. Of coruse I'd chime in and say the blackjack method was a different approach that was not needed, and RQ/BRP got along for over 3 decades without it. So it could be a lively debate.

I'd love to just use success levels, but I wish there were a few more grades to break up "success". Among skilled character, success vs. success is the single most common outcome.

The Rolemaster method would work, that is D100+skill, roll over 100 to succeed. Then the crticals and special could be put up[ on the high end of the table. For example, somebody with Sword 60% would have Fumble 01-02%, Special Success 89%-00%, Critical 98-00%

Edited September 17, 2012 by Atgxtg

[SDLeary]

The best way of thinking about the Blackjack method is this: Whoever gets the best level of success wins, but in case of a tie the highest roll wins. It helps make the numbers fade into the background.

Or to simply dispense with Levels of Success as we now use them in BRP and go back to the cRQ2 usage where, apart from success and failure, they simply invoked a special effect (Critical damage ,Slash, Crush, Impale, Failure, Fumble)

SDLeary

Edited September 17, 2012 by SDLeary

[Atgxtg]

Or to simply dispense with Levels of Success as we now use them in BRP and go back to the cRQ2 usage where, apart from success and failure, they simply invoked a special effect (Critical damage ,Slash, Crush, Impale, Failure, Fumble)

SDLeary

Yeah, but RQ2 didn't use opposed rolls. HE who got a better SL benefited in a fashion that was equivalent to winning. For instance a better attack did more damage, and abetter parry stopped more damage. So functionally, it worked the same as better SL winning.

I've been thinking of going with a higher success chance (it really is pretty easy to hit someone in melee. If you fail it is almost always because of something they did) oppsed rolls, and that raise concept, where you can up the difficulty to double the effect. Some of that can translate into BRP by using the difference for effect. A higher base chance would increase the possible spread for the difference, and reduce the amount of "zeroing out". It would also make better hits more a matter of skill and raises than luck. And it would take care of the skills over 100% fights. Guys at very hight skill levels would need to take raises to benefit from thier higher skill, and attack and defense raises would tend to cancel out.

[SDLeary]

Yeah, but RQ2 didn't use opposed rolls. HE who got a better SL benefited in a fashion that was equivalent to winning. For instance a better attack did more damage, and abetter parry stopped more damage. So functionally, it worked the same as better SL winning.

Yes! So Better Success Level Wins wasn't really needed! ;-)

And as far as opposed rolls go... Lowest roll under skill wins sounds just fine, no need for fancy schmancy "success levels".

SDLeary

[Atgxtg]

Yes! So Better Success Level Wins wasn't really needed! ;-)

Nope. not in combat. Even MRQ1 didn't use opposed rolls for combat (although I'd bet money it was supposed to, it would have made more sense that way). The problem has always been when characters were opposed in some non-combat skill. Like a poker match, or a PC trying to sneak past a guard. And even then RQ always had a non opoosed way of handling it.

And as far as opposed rolls go... Lowest roll under skill wins sounds just fine, no need for fancy schmancy "success levels".

SDLeary

Not to decide who wins. SLs differentiate between successes. It's a quality thing. To use a baseball analogy it is the difference between a base hit, double, triple, or home run. And I hope I remember than analogy when I try to explain SLs to new players!

[Psychman]

The other issue with just "roll low" is that it doesn't automatically reward higher skill levels. If a in an opposed test a person with 50% rolls 13, and a the opposer with a 70% rolls 16 who has won. Simple roll low says the the 50% but doesn't seem fair.

The usual way to resolve this issue is "amount below skill" so the first guy has passed skill by 37, and the second guy has passed his skill by 54, so second guy won. It works but many people find subtraction on the fly awkward, even if math-capable. The Blackjack method enables high skill to have value without the awkward maths.

[Atgxtg]

The other issue with just "roll low" is that it doesn't automatically reward higher skill levels. If a in an opposed test a person with 50% rolls 13, and a the opposer with a 70% rolls 16 who has won. Simple roll low says the the 50% but doesn't seem fair.

It might not seem fair to you, but it is. With the low critical and specials, it is logical to assume that a low roll indicates a superior effort and/or result. That's why an 01 is better than a 10. So the guy who rolled a 13 put in a better effort than the guy who rolled a 16, even if the second guy has a higher skill score. Where he benifits from his skill is when the rolls and in the 51-70% range, where he wins all the time.

The usual way to resolve this issue is "amount below skill" so the first guy has passed skill by 37, and the second guy has passed his skill by 54, so second guy won. It works but many people find subtraction on the fly awkward, even if math-capable. The Blackjack method enables high skill to have value without the awkward maths.

But we still have awkward maths because of the special and critical %. You wind up with situations where a guy loses but would have won if he had rolled higher or lower. No other game that uses a blackjack mechanic combined it with low crits and specials. Pendragon uses a blackjack mechanic, but it uses high rolls for crtits. If they swapped the crti and fumble results and moved the specials to the high end, I'd be okay with it as a mechanism.

.

[axe-elf]

Reactions? Thoughts?

Why not just use the resistance table? Much faster with one roll. Special, critical and fumble can be calculated from the %-roll of the table also.

[Atgxtg]

Why not just use the resistance table? Much faster with one roll. Special, critical and fumble can be calculated from the %-roll of the table also.

Becuase anybody with a 10 point advatage on the table wins automatically. So a 90 vs. and 80 is a automatic win.Or do you mean use skill/5?

BTW, I was wondering just what people think the success chance should be. For example with 50% vs 70%, how do people think the results should be?

:30/70 (per the resistance table)

;42/58 (per the ratios)

:or what?

Maybe if we have the answer, we can look at the methods.

[SDLeary]

The other issue with just "roll low" is that it doesn't automatically reward higher skill levels. If a in an opposed test a person with 50% rolls 13, and a the opposer with a 70% rolls 16 who has won. Simple roll low says the the 50% but doesn't seem fair.

They have BOTH succeeded... that is the round is a tie, at least in my world. If combat, one might inflict a wound on the other, but in this case the level of success is not what determines that. The 70% person has a range above the skill of the other where they succeed (51-70) AND they have a higher chance of rolling a Special or Critical.

The usual way to resolve this issue is "amount below skill" so the first guy has passed skill by 37, and the second guy has passed his skill by 54, so second guy won. It works but many people find subtraction on the fly awkward, even if math-capable. The Blackjack method enables high skill to have value without the awkward maths.

High skills still have value in roll-low. Greater overall chance of success, greater chance of special or critical and their resulting effects.

In contests of various types there is not always a winner or looser.

SDLeary

[Atgxtg]

They have BOTH succeeded... that is the round is a tie, at least in my world. If combat, one might inflict a wound on the other, but in this case the level of success is not what determines that. The 70% person has a range above the skill of the other where they succeed (51-70) AND they have a higher chance of rolling a Special or Critical.

I'd love that. The problems with that are twofo9ld:

1) some types of contests don't allow for a "push" (yet another blackjack term).

2) With the basic success and/or the faulire blocks being so large, a disproportionate amount of contestest end up as ties/pushes (the same SL).

For example, if two characters both have 100% skill. they match SLs over 66% of the time, and with both at 50%, ties occur over 49% of the time.

If BRP differentiated the SLs a bit more, it would reduce the number of ties.

High skills still have value in roll-low. Greater overall chance of success, greater chance of special or critical and their resulting effects.

Yes, in fact, when you do the math, you see just how much of an edge the higher skill actually has. Most of the "it's unfair" statments arise becuase people are looking are setting up "unfair" conditions by using a vey narrow range of the possible outcomes, and then saying that it is wrong. Such as the 50% vs. 70% example above.Let's put that into perpspective.

A guy with 30% goes up against a guy with 90% The 30% guy rolls rolls an 02. Now according to the logic being used here to justify the blackjack method the 90% is somehow getting screwed because if he rolls anything over an 05 he looses! Everyone cries unfair, but that is a very rare occurrence (less than 1%). but it is all due to the great 02 roll by the 30% guy. What people seem to miss is all the times the 30% guy will not roll so good. Same with the 50%er above.

In contests of various types there is not always a winner or looser.

SDLeary

Nope, and I'd say most contestes allow for a "push" of some sort. Just not all. could certainly see using a "push" to resolve most situations, though. Some Examples:

1) Baseball, batting (Bat% vs. Pitch %) TIE= Foul Ball

2) Sneaking Past a Guard (Sneak vs. Spot): TIE= Not spotted but hand to stop and duck behind cover for a bit

3) Spellcasting (POW vs POW roll): TIE=Still duking it out mentally, spell resolution delayed 1 SR/DEX Rank

4) Armwrestling (STR vs. STR): TIE= No arm moves, yet.

But there are some cases where a push doesn't make sense.

The old RQ method was to give ties to the defender (2nd roll).

[Atgxtg]

How doy you set up a pool on the forum?

I'd like to see what the consuses (if any) we have for the winning% of 50 vs 70. Just to see what people think is "right".

[Evilschemer]

Personally, I wish the official rules were "roll highest without goving over" ala Pendragon. Critical = your skill level, Fumble = 01. It's just the simplest and easiest to convey to the players.

When I was playtesting Swords of Cydoria, I got a lot of complaints about BRP's convoluted "roll low, lower is better unless its contested, then higher is better if you are both within the same success level". All my players hated it and it was a big barrier towards getting them to accept BRP.

[Atgxtg]

Personally, I wish the official rules were "roll highest without goving over" ala Pendragon. Critical = your skill level, Fumble = 01. It's just the simplest and easiest to convey to the players.

Well, no opposed rolls is the simplest and easiest. It is how RQ/BRP worked for 30 years or so. But the Pnedragon method is fine, although it limits you to just 2 success levels, and breaks down at 200% (I had a guy with a 30+ Sword skill in my last Pendragon campaign, and it gets boring pretty quickly).

When I was playtesting Swords of Cydoria, I got a lot of complaints about BRP's convoluted "roll low, lower is better unless its contested, then higher is better if you are both within the same success level". All my players hated it and it was a big barrier towards getting them to accept BRP.

Exactly! That's the reaction I've seen and gotten from everybody, save the handful of MRQ/Blackjack fans.

The more I look at the math, the less I like the blackjack method.

Look at 5% vs 95%, about the most lopsided case we could really expect to see in play.

By old rules, the 5% is in a tough spot, but he has a 1.15% chance a winning (critical vs. other, or success vs. failure or fumble), and about a 3% chance of getting a tie (success vs. success).

With low roll wins and success level, he has about s 2% chance of winning.

With the blackjack method, he has less than 1% chance of winning, as anything other than a critical success turns into a automatic win for the other guy.

And then there are the cases where both fail, and the higher skilled guy gets the win due to the blackjack method. Like when a the 50% rolls a 51 and looses to the 70% guy's roll of 95 (just one short of failure). That pushes me more towards SD's method, Pendragon method (both would fail), or skill multiples (the 50% guy wins since he only rolled 1.02 times his skill, where the 70% guy rolled 1.36 times his skill).

Edited September 26, 2012 by Atgxtg

[axe-elf]

Or do you mean use skill/5?

Yes. This is how the optional rule is described in BRP.

BTW, I was wondering just what people think the success chance should be. For example with 50% vs 70%, how do people think the results should be?

:30/70 (per the resistance table)

;42/58 (per the ratios)

:or what?

Maybe if we have the answer, we can look at the methods.

I think the table gives the lower skill an OK chance.

Im not sure about the calculation behind the 42/58 (I would say: 50% x (100%-70%)= 15% => 15/85), but it does not differ so much from 30/70.

(Edit: Lower not lover )

Edited September 26, 2012 by axe-elf

[Evilschemer]

Well, not opposed rolls is the simplest and easiest. It is how RQ/BRP worked for 30 years or so. But the Pnedragon method is fine, although it limits you to just 2 success levels, and breaks down at 200% (I had a guy with a 30+ Sword skill in my last Pendragon campaign, and it gets boring pretty quickly).

Naw, you could still have success levels.

If I had a skill of 70%, a special success would occur on a roll of 56-69. A roll of 70 = critical.

Skill	Fumble	Fumble %	Success	Special	Special %	Critical

100	01	1%		01 - 79	80 - 99	20%	100

99	01	1%		01 - 78	79 - 98	20%	99

98	01	1%		01 - 77	78 - 97	20%	98

97	01	1%		01 - 76	77 - 96	19%	97

96	01	1%		01 - 75	76 - 95	19%	96

95	01	1%		01 - 75	76 - 94	19%	95

94	01	1%		01 - 74	75 - 93	19%	94

93	01	1%		01 - 73	74 - 92	19%	93

92	01	1%		01 - 72	73 - 91	18%	92

91	01	1%		01 - 71	72 - 90	18%	91

90	01	1%		01 - 71	72 - 89	18%	90

89	01	1%		01 - 70	71 - 88	18%	89

88	01	1%		01 - 69	70 - 87	18%	88

87	01	1%		01 - 68	69 - 86	17%	87

86	01	1%		01 - 67	68 - 85	17%	86

85	01	1%		01 - 67	68 - 84	17%	85

84	01	1%		01 - 66	67 - 83	17%	84

83	01	1%		01 - 65	66 - 82	17%	83

82	01	1%		01 - 64	65 - 81	16%	82

81	01	1%		01 - 63	64 - 80	16%	81

80	01	1%		01 - 63	64 - 79	16%	80

79	01	1%		01 - 62	63 - 78	16%	79

78	01	1%		01 - 61	62 - 77	16%	78

77	01	1%		01 - 60	61 - 76	15%	77

76	01	1%		01 - 59	60 - 75	15%	76

75	01	1%		01 - 59	60 - 74	15%	75

74	01	1%		01 - 58	59 - 73	15%	74

73	01	1%		01 - 57	58 - 72	15%	73

72	01	1%		01 - 56	57 - 71	14%	72

71	01	1%		01 - 55	56 - 70	14%	71

70	01	1%		01 - 55	56 - 69	14%	70

69	01	1%		01 - 54	55 - 68	14%	69

68	01	1%		01 - 53	54 - 67	14%	68

67	01	1%		01 - 52	53 - 66	13%	67

66	01	2%		01 - 51	52 - 65	13%	66

65	01	2%		01 - 51	52 - 64	13%	65

64	01	2%		01 - 50	51 - 63	13%	64

63	01	2%		01 - 49	50 - 62	13%	63

62	01	2%		01 - 48	49 - 61	12%	62

61	01	2%		01 - 47	48 - 60	12%	61

60	01	2%		01 - 47	48 - 59	12%	60

59	01	2%		01 - 46	47 - 58	12%	59

58	01	2%		01 - 45	46 - 57	12%	58

57	01	2%		01 - 44	45 - 56	11%	57

56	01	2%		01 - 43	44 - 55	11%	56

55	01	2%		01 - 43	44 - 54	11%	55

54	01	2%		01 - 42	43 - 53	11%	54

53	01	2%		01 - 41	42 - 52	11%	53

52	01	2%		01 - 40	41 - 51	10%	52

51	01	2%		01 - 39	40 - 50	10%	51

50	01	2%		01 - 39	40 - 49	10%	50

49	01	2%		01 - 38	39 - 48	10%	49

48	01	2%		01 - 37	38 - 47	10%	48

47	01	2%		01 - 36	37 - 46	9%	47

46	01	2%		01 - 35	36 - 45	9%	46

45	01	2%		01 - 35	36 - 44	9%	45

44	01	2%		01 - 34	35 - 43	9%	44

43	01	2%		01 - 33	34 - 42	9%	43

42	01	2%		01 - 32	33 - 41	8%	42

41	01	2%		01 - 31	32 - 40	8%	41

40	01	3%		01 - 31	32 - 39	8%	40

39	01	3%		01 - 30	31 - 38	8%	39

38	01	3%		01 - 29	30 - 37	8%	38

37	01	3%		01 - 28	29 - 36	7%	37

36	01	3%		01 - 27	28 - 35	7%	36

35	01	3%		01 - 27	28 - 34	7%	35

34	01	3%		01 - 26	27 - 33	7%	34

33	01	3%		01 - 25	26 - 32	7%	33

32	01	3%		01 - 24	25 - 31	6%	32

31	01	3%		01 - 23	24 - 30	6%	31

30	01	3%		01 - 23	24 - 29	6%	30

29	01	3%		01 - 22	23 - 28	6%	29

28	01	4%		01 - 21	22 - 27	6%	28

27	01	4%		01 - 20	21 - 26	5%	27

26	01	4%		01 - 19	20 - 25	5%	26

25	01	4%		01 - 19	20 - 24	5%	25

24	01	4%		01 - 18	19 - 23	5%	24

23	01	4%		01 - 17	18 - 22	5%	23

22	01	5%		01 - 16	17 - 21	4%	22

21	01	5%		01 - 15	16 - 20	4%	21

20	01	5%		01 - 15	16 - 19	4%	20

19	01	5%		01 - 14	15 - 18	4%	19

18	01	6%		01 - 13	14 - 17	4%	18

17	01	6%		01 - 12	13 - 16	3%	17

16	01	6%		01 - 11	12 - 15	3%	16

15	01	7%		01 - 11	12 - 14	3%	15

14	01	7%		01 - 10	11 - 13	3%	14

13	01	8%		01 - 9	10 - 12	3%	13

12	01	8%		01 - 8	9 - 11	2%	12

11	01	9%		01 - 7	8 - 10	2%	11

10	01	10%		01 - 7	8 - 9	2%	10

9	01	11%		01 - 6	7 - 8	2%	09

8	01	13%		01 - 5	6 - 7	2%	08

7	01	14%		01 - 4	5 - 6	1%	07

6	01	17%		01 - 3	4 - 5	1%	06

5	01	20%		01 - 3	4 - 4	1%	05

4	01	25%		01 - 2	3 - 3	1%	04

3	01	33%		01 - 1	2 - 2	1%	03

2	01	50%		NA	NA	0%	02

1	NA	NA		NA	NA	0%	01

Edited September 26, 2012 by Evilschemer