Quote from: vgunn;556828[*1] If the rolls occur simultaneously, then the time would be roughly the same as a TN. Maybe a few seconds longer.

Considerably longer, since for an opposed roll you have to compare the values. The literal amount of time the dice spend rolling isn't a big deal, it's the fact that you're running at least one extra comparison operation between numbers that are being generated by different people.

Quote[*2] I would think the opposite would be the case, in that the bonuses becomes very important. In this particular system (d6 pool), Cleric gets 1D for Trade, 2D for Tools [Holy Symbol] [True Words], 1D for Trademark [Bless]. 4D total to roll. The fighter (since he's not resisting) would roll would roll 1D + 1P, and drop the highest die. The low die of the fighter is what the cleric needs to meet or beat for a successful roll. So, there's a very good chance the cleric's roll will be successful.

Opposed rolls make your bonus matter less, that's just math. It doesn't matter whether you're using a d20 or dicepools or a deck of cards to generate your results. The easiest way to think about it is that if you're doing opposed tests, there is some non-zero chance that the opposition will flub their roll and then your bonuses won't be important. But let's run through the actual math. And for that, I'll pretend that we're using the random number generator of 2d6+Bonuses, because the math on that comes out really easy.

Now you could do 2d6+Bonuses vs. a fixed TN of 7, or you could do 2d6+Bonuses vs. a TN generated by an opposed 2d6. You'd agree that's the same on average, right? But wait! That's the same as rolling 2d6 + Bonuses minus 2d6 against a TN of 0.You're just moving the dice from one side of the equals sign to the other by reversing their sign (or in this case, pushing them across the table and reversing the sign). But wait! Instead of doing that, you could invert the signs again and just roll 4d6 + Bonuses - 14 against a TN of 0. That would be the same as making an opposed roll.

Except here's the thing: I'll bet you'll agree that if your random number generator is 2d6, that a +1 makes a lot more difference than it does if your RNG is 4d6, right? Your RNG is only 11 numbers long instead of 21 numbers long, of course a +1 makes more difference! It only takes 5 of those things to push you completely off the RNG in one direction or the other instead of 10.

Whenever you have an opposed roll, you are in essence playing with a bigger RNG. And that makes each nominal bonus you have mean less. Opposed d20 rolls is statistically the same as rolling two d20s together. Dice pools being opposed functions similarly (if more complicatedly) in that putting dice on the other side of the table is statistically identical to putting dice on your side of the table and facing a higher success threshold.

-Frank

Quote from: FrankTrollman;556791I don't think we are in agreement. An "opposed roll" doesn't have to represent "the target taking active measures against your action", it represents "the difficulty of the task is a variable function for whatever reason."

The original question is system agnostic. You have to roll dice to determine whether the Bless succeeds in whatever scenario or magic system the OP (vgunn) is envisioning. But we don't know what the flavor is. And the truth is, we could put in flavor that either made Opposed Rolls "make sense" or not as we chose. If the flavor was that you had to slip magical effects past the target's mental defenses or trick them into thinking about your god's name or something in order to get your bless through, then it wouldn't make any sense for a roll to be opposed if the target was trying to help you. On the other hand, if the flavor was that the Tide of Magic™ ebbed and flowed and the relative difficulty of casting spells went up and down, then it would make perfect sense for your Bless roll to be "opposed" whether the target was in favor or against.

The real question is what you're hoping to get mechanically out of opposed vs. static rolls. And what you get for an opposed roll is that:

• It takes longer to resolve the action.
• The player character's bonuses matter less because the TN might be really high or really low.
• The player is more likely to pass or fail by a large amount, because the TN might be really high or really low.

And frankly, I don't see the latter two as advantages for the action "cast a minor buff on the Fighter", and the first is definitely a drawback in virtually all cases.

But not because I can't write flavor text where it would "make sense" to roll an Opposed Check. The opposition could be by anything, so of course you can write flavor text where it would work.

-Frank

OK, but aren't you needlessly complicating a simple situation in the rules?

Quote from: jeff37923;556851OK, but aren't you needlessly complicating a simple situation in the rules?

Dude, it's Frank.

Quote from: jeff37923;556851OK, but aren't you needlessly complicating a simple situation in the rules?

I would think you would be. I don't actually know, because I know almost nothing about the system in question. The question arises why you would be rolling for a minor buff at all rather than ticking it off and moving on with your life. Maybe you're in a game like Shadowrun, where every spell cast drains your energy reserves and has a variable effect that is itself limited by how much drain you gambled with when the spell is cast. Maybe you're playing a game like Call of Cthulhu, where every spell you cast is supposed to be risky and uncertain. Maybe you're playing a game like Role Master, where exploding things with critical spell failures and successes is half the fun. I don't know.

Whether the casting of a buff spell on the fighter is supposed to be a tense moment with potentially wildly different and possibly dangerous results or not is not something I could tell you without knowing more about the game that was being designed.

-Frank

@Frank:  The system doesn't have you roll unless the outcome is uncertain. So your not rolling for a minor bluff, except in certain situations.

The basic resolution is:


Whenever your character attempts to overcome a foe or peril and the outcome is not certain, you'll need to roll dice to see if you succeed. If you try something which is related to your trade you get one die [1D]. Add another die for each tool and trademark you can use. If your die pool doesn't seem like it's enough, you can lower your threat score on a one point for one die basis. Any of your companions can also tag one of their tools or trademarks to give you a die of their own. Keep in mind, however, that you cannot roll more than six dice at a time. This is known as the 'rule of six'. Finally, roll all the dice in your pool. Keep the highest die. If you have a pair for your highest dice, add them together. With three or more of a highest number, each die beyond the pair can be used to refresh your threat score, remove a trouble, or add a triumph to a successful result. If your roll is equal to, or higher than, the opposing score then you have succeeded.

Quote from: vgunn;556905@Frank:  The system doesn't have you roll unless the outcome is uncertain. So your not rolling for a minor bluff, except in certain situations.

The basic resolution is:


Whenever your character attempts to overcome a foe or peril and the outcome is not certain, you'll need to roll dice to see if you succeed. If you try something which is related to your trade you get one die [1D]. Add another die for each tool and trademark you can use. If your die pool doesn't seem like it's enough, you can lower your threat score on a one point for one die basis. Any of your companions can also tag one of their tools or trademarks to give you a die of their own. Keep in mind, however, that you cannot roll more than six dice at a time. This is known as the 'rule of six'. Finally, roll all the dice in your pool. Keep the highest die. If you have a pair for your highest dice, add them together. With three or more of a highest number, each die beyond the pair can be used to refresh your threat score, remove a trouble, or add a triumph to a successful result. If your roll is equal to, or higher than, the opposing score then you have succeeded.

Sorry to be blunt, but that's completely bugnuts. Why are you writing a game with such an incredibly wonky random number generator?

Do you have any idea what the actual probabilities that such a setup generates are? And whether you do or not, why do you think such a counter-intuitive setup is something you should use as your core resolution mechanic?

-Frank

chalk up another for TN.

Quote from: FrankTrollman;556959Sorry to be blunt, but that's completely bugnuts. Why are you writing a game with such an incredibly wonky random number generator?

Do you have any idea what the actual probabilities that such a setup generates are? And whether you do or not, why do you think such a counter-intuitive setup is something you should use as your core resolution mechanic?

-Frank

Yep, I have the probabilities. Really it's quite simple.

For example, a cleric wants to turn a skeleton. 1D for [trade/cleric] 2D for [tools/holy symbol and true words] 1D for [trademark/ward].

Roll 4d6, keep the highest die (unless highest are doubles).

Quote from: vgunn;556977Yep, I have the probabilities. Really it's quite simple.

For example, a cleric wants to turn a skeleton. 1D for [trade/cleric] 2D for [tools/holy symbol and true words] 1D for [trademark/ward].

Roll 4d6, keep the highest die (unless highest are doubles).

Do you add anything to those?

How do you determine the TN?  

Do TNs go beyond 6?  

Have you calculated the benefit of reducing the number of dice rolled versus reducing the TN by 1?  

If the TN is above 6 and you don't add anything to your roll, I think you're basically always best served reducing the TN to the lowest possible and at most rolling two dice (to give you a chance at doubles).  Since doubles only matter when they occur on the highest die roll, rolling 3 or 4 dice doesn't really help much, since the doubles might from something below the highest roll.

Quote from: deadDMwalking;556996Do you add anything to those?

How do you determine the TN?  

Do TNs go beyond 6?  

Have you calculated the benefit of reducing the number of dice rolled versus reducing the TN by 1?  

If the TN is above 6 and you don't add anything to your roll, I think you're basically always best served reducing the TN to the lowest possible and at most rolling two dice (to give you a chance at doubles).  Since doubles only matter when they occur on the highest die roll, rolling 3 or 4 dice doesn't really help much, since the doubles might from something below the highest roll.

Yes, there are some dice tweaks I didn't go into. You have trumps, which are special abilities that can effect the roll. You can also spend threat points for dice.

TNs go from 1-12.

But if you have extra multiples of the same dice (beyond the doubles), for example, roll 5D (5,5,5,3,2). 5+5=10. The other 5 gives you an extra success, which comes with benefits.

If the roll isn't a big deal but you insist on the cleric having to roll, possibly the spell should only fail on some sort of 'critical failure' result (i.e. a very low TN). The system should make sure that if the opponent is actively resisting a spell or whatever, the opposed roll difficulty would be much higher.

Another reason to use opposed rolls rather than set TNs might be because there's a pool of 're-rolls' floating around, and you want to set the difficulty as a dice roll so that resource becomes available to spend (though there certainly are drawbacks to an opposed roll, pretty much as others such as Frank have outlined). However, in the fighter/cleric example above, this gets complicated because by default lower rolls are better, and so you'd need to add rules to explain how characters can trade in bonuses for extra penalties if you're going down that path.

Quote from: Bloody Stupid Johnson;557055If the roll isn't a big deal but you insist on the cleric having to roll, possibly the spell should only fail on some sort of 'critical failure' result (i.e. a very low TN). The system should make sure that if the opponent is actively resisting a spell or whatever, the opposed roll difficulty would be much higher.

Right. Certainly would be the case, if I went this way.

Quote from: Bloody Stupid Johnson;557055Another reason to use opposed rolls rather than set TNs might be because there's a pool of 're-rolls' floating around, and you want to set the difficulty as a dice roll so that resource becomes available to spend (though there certainly are drawbacks to an opposed roll, pretty much as others such as Frank have outlined). However, in the fighter/cleric example above, this gets complicated because by default lower rolls are better, and so you'd need to add rules to explain how characters can trade in bonuses for extra penalties if you're going down that path.

Yes, re-rolls are possible in the system. Something for me to think about. Thanks!

If I must choose from those 2 then TN.


I am not sure why you would set up a system where the cleric giving a minor buff to the fighter would not be automatic. I am accustomed to Vancian magic systems. So either the bless would "go off" or it would be fully negated (e.g. by the cleric taking combat damage before he could get the spell off).

Why wouldn't you have the bless be automatic and then have the fighter roll to see if he succeeds at his own check?

While outputs nominally go from 1 to 12, the 1 is only possible on a one die roll. For all others, the minimum output is 2. Furthermore, outputs of 7, 9, and 11 cannot happen under any circumstances. So you only really have 8 potential output numbers. The way you're describing it, it's incredibly easy to roll 4 dice and the maximum you can ever roll is 6, so the inputs are even more compressed.

In the 4+ dice world, your chances of rolling a 6+ are very very high, heck even at just 2 dice your chances of rolling a 6+ are 7/18 - more than one third of rolls are clustered at the incredibly wonky top. And by incredibly wonky top, I mean that at all dice totals other than 1 and 2 it is flat more likely to roll a 12 than it is to roll an 8. Even on three dice, the 12 comes up on 8/108 rolls while the 8 comes up on only 5/108.

So seriously, what the hell? There's two bulges, where a 6 is more likely than a 5 or 8, but a 12 is also more likely than an 8. What is that even for? There's a probability curve, but numbers adjacent to each other on the probability distribution aren't next to each other numerically. You might as well use a spinner. At least that would graphically display peoples' chances to them.

-Frank

Quote from: FrankTrollman;557065*snip*


Count of how many times player scores X with 1D+0R+0P (100000 total tests)
  1     16725  16.7%
  2     16698  16.7%
  3     16582  16.6%
  4     16633  16.6%
  5     16678  16.7%
  6     16684  16.7%
Average score: 3.50

Count of how many times player scores X with 2D+0R+0P (100000 total tests)
  2      8452   8.5%
  3     11115  11.1%
  4     19227  19.2%
  5     22199  22.2%
  6     30597  30.6%
  8      2788   2.8%
 10      2812   2.8%
 12      2810   2.8%
Average score: 5.06

Count of how many times player scores X with 3D+0R+0P (100000 total tests)
  2      1844   1.8%
  3      5652   5.7%
  4     14313  14.3%
  5     21962  22.0%
  6     38103  38.1%
  8      4603   4.6%
 10      6164   6.2%
 12      7359   7.4%
Average score: 6.03

Count of how many times player scores X with 4D+0R+0P (100000 total tests)
  2       398   0.4%
  3      2451   2.5%
  4      9135   9.1%
  5     19591  19.6%
  6     41373  41.4%
  8      5185   5.2%
 10      8839   8.8%
 12     13028  13.0%
Average score: 6.77

Count of how many times player scores X with 5D+0R+0P (100000 total tests)
  2        90   0.1%
  3      1082   1.1%
  4      5565   5.6%
  5     16220  16.2%
  6     42137  42.1%
  8      4754   4.8%
 10     10631  10.6%
 12     19521  19.5%
Average score: 7.38

Count of how many times player scores X with 6D+0R+0P (100000 total tests)
  2        13   0.0%
  3       401   0.4%
  4      3223   3.2%
  5     13047  13.0%
  6     41318  41.3%
  8      4208   4.2%
 10     11474  11.5%
 12     26316  26.3%
Average score: 7.91

Count of how many times player scores X with 6D+1R+0P (100000 total tests)
  2         6   0.0%
  3       177   0.2%
  4      1878   1.9%
  5     10088  10.1%
  6     39707  39.7%
  8      3184   3.2%
 10     11636  11.6%
 12     33324  33.3%
Average score: 8.38

Count of how many times player scores X with 6D+2R+0P (100000 total tests)
  2         1   0.0%
  3        71   0.1%
  4      1021   1.0%
  5      8007   8.0%
  6     37737  37.7%
  8      2446   2.4%
 10     11498  11.5%
 12     39219  39.2%
Average score: 8.76

Count of how many times player scores X with 6D+3R+0P (100000 total tests)
  3        26   0.0%
  4       569   0.6%
  5      5984   6.0%
  6     35119  35.1%
  8      1875   1.9%
 10     10778  10.8%
 12     45649  45.6%
Average score: 9.14

Count of how many times player scores X with 1D+0R+1P (100000 total tests)
  1     30460  30.5%
  2     25075  25.1%
  3     19606  19.6%
  4     13751  13.8%
  5      8318   8.3%
  6      2790   2.8%
Average score: 2.53

Count of how many times player scores X with 2D+0R+1P (100000 total tests)
  2     19943  19.9%
  3     19390  19.4%
  4     26954  27.0%
  5     16587  16.6%
  6     11489  11.5%
  8      3274   3.3%
 10      1880   1.9%
 12       483   0.5%
Average score: 4.09

Count of how many times player scores X with 3D+0R+1P (100000 total tests)
  2      5590   5.6%
  3     13102  13.1%
  4     26135  26.1%
  5     22322  22.3%
  6     18659  18.7%
  8      7255   7.3%
 10      5297   5.3%
 12      1640   1.6%
Average score: 5.09

Count of how many times player scores X with 4D+0R+1P (100000 total tests)
  2      1479   1.5%
  3      7061   7.1%
  4     20420  20.4%
  5     24574  24.6%
  6     23388  23.4%
  8      9853   9.9%
 10      9619   9.6%
 12      3606   3.6%
Average score: 5.87

Count of how many times player scores X with 5D+0R+1P (100000 total tests)
  2       368   0.4%
  3      3700   3.7%
  4     14422  14.4%
  5     24618  24.6%
  6     25913  25.9%
  8     10940  10.9%
 10     13803  13.8%
 12      6236   6.2%
Average score: 6.48

Count of how many times player scores X with 5D+1R+1P (100000 total tests)
  2        79   0.1%
  3      1660   1.7%
  4      9736   9.7%
  5     23146  23.1%
  6     27279  27.3%
  8     11012  11.0%
 10     17685  17.7%
 12      9403   9.4%
Average score: 7.01

Count of how many times player scores X with 5D+2R+1P (100000 total tests)
  2        16   0.0%
  3       769   0.8%
  4      6407   6.4%
  5     20379  20.4%
  6     28804  28.8%
  8      9779   9.8%
 10     20414  20.4%
 12     13432  13.4%
Average score: 7.46

Count of how many times player scores X with 5D+3R+1P (100000 total tests)
  2         2   0.0%
  3       434   0.4%
  4      4639   4.6%
  5     19228  19.2%
  6     31059  31.1%
  8      7930   7.9%
 10     20349  20.3%
 12     16359  16.4%
Average score: 7.66

Count of how many times player scores X with 5D+4R+1P (100000 total tests)
  2        62   0.1%
  3      1149   1.1%
  4      6892   6.9%
  5     21879  21.9%
  6     32834  32.8%
  8      7311   7.3%
 10     16737  16.7%
 12     13136  13.1%
Average score: 7.21

Count of how many times player scores X with 1D+0R+2P (100000 total tests)
  1     42267  42.3%
  2     28081  28.1%
  3     17265  17.3%
  4      8747   8.7%
  5      3227   3.2%
  6       413   0.4%
Average score: 2.04

Count of how many times player scores X with 2D+0R+2P (100000 total tests)
  2     31911  31.9%
  3     22910  22.9%
  4     26332  26.3%
  5      8677   8.7%
  6      6731   6.7%
  8      2522   2.5%
 10       829   0.8%
 12        88   0.1%
Average score: 3.51

Count of how many times player scores X with 3D+0R+2P (100000 total tests)
  2     11304  11.3%
  3     19380  19.4%
  4     31685  31.7%
  5     14306  14.3%
  6     12944  12.9%
  8      7138   7.1%
 10      2928   2.9%
 12       315   0.3%
Average score: 4.47

Count of how many times player scores X with 4D+0R+2P (100000 total tests)
  2      3490   3.5%
  3     12716  12.7%
  4     28339  28.3%
  5     19262  19.3%
  6     17097  17.1%
  8     11541  11.5%
 10      6675   6.7%
 12       880   0.9%
Average score: 5.27

Count of how many times player scores X with 4D+1R+2P (100000 total tests)
  2       984   1.0%
  3      7579   7.6%
  4     22874  22.9%
  5     22623  22.6%
  6     18589  18.6%
  8     14857  14.9%
 10     10665  10.7%
 12      1829   1.8%
Average score: 5.88

Count of how many times player scores X with 4D+2R+2P (100000 total tests)
  2       656   0.7%
  3      5663   5.7%
  4     19998  20.0%
  5     25568  25.6%
  6     18536  18.5%
  8     14448  14.4%
 10     12772  12.8%
 12      2359   2.4%
Average score: 6.09

Count of how many times player scores X with 4D+3R+2P (100000 total tests)
  2      1471   1.5%
  3      7785   7.8%
  4     22684  22.7%
  5     25272  25.3%
  6     17903  17.9%
  8     12337  12.3%
 10     10534  10.5%
 12      2014   2.0%
Average score: 5.79

Count of how many times player scores X with 4D+4R+2P (100000 total tests)
  2      3509   3.5%
  3     12673  12.7%
  4     28529  28.5%
  5     19130  19.1%
  6     17043  17.0%
  8     11633  11.6%
 10      6571   6.6%
 12       912   0.9%
Average score: 5.27

Count of how many times player scores X with 4D+5R+2P (100000 total tests)
  2      1018   1.0%
  3      7414   7.4%
  4     23326  23.3%
  5     22592  22.6%
  6     18465  18.5%
  8     14749  14.7%
 10     10673  10.7%
 12      1763   1.8%
Average score: 5.87

Count of how many times player scores X with 1D+0R+3P (100000 total tests)
  1     51896  51.9%
  2     28347  28.3%
  3     13480  13.5%
  4      5043   5.0%
  5      1153   1.2%
  6        81   0.1%
Average score: 1.75

Count of how many times player scores X with 2D+0R+3P (100000 total tests)
  2     43466  43.5%
  3     22557  22.6%
  4     22989  23.0%
  5      3843   3.8%
  6      5152   5.2%
  8      1638   1.6%
 10       344   0.3%
 12        11   0.0%
Average score: 3.13

Count of how many times player scores X with 3D+0R+3P (100000 total tests)
  2     18196  18.2%
  3     22632  22.6%
  4     32415  32.4%
  5      7539   7.5%
  6     12137  12.1%
  8      5566   5.6%
 10      1439   1.4%
 12        76   0.1%
Average score: 4.04

Count of how many times player scores X with 3D+1R+3P (100000 total tests)
  2      9770   9.8%
  3     20060  20.1%
  4     33086  33.1%
  5     11746  11.7%
  6     14075  14.1%
  8      8397   8.4%
 10      2717   2.7%
 12       149   0.1%
Average score: 4.51

Count of how many times player scores X with 3D+2R+3P (100000 total tests)
  2     11769  11.8%
  3     20530  20.5%
  4     32790  32.8%
  5     11800  11.8%
  6     12881  12.9%
  8      7546   7.5%
 10      2534   2.5%
 12       150   0.1%
Average score: 4.40

Count of how many times player scores X with 3D+3R+3P (100000 total tests)
  2     17894  17.9%
  3     22651  22.7%
  4     32797  32.8%
  5      7689   7.7%
  6     11991  12.0%
  8      5488   5.5%
 10      1428   1.4%
 12        62   0.1%
Average score: 4.04

Count of how many times player scores X with 3D+4R+3P (100000 total tests)
  2      9691   9.7%
  3     20112  20.1%
  4     33238  33.2%
  5     11504  11.5%
  6     14115  14.1%
  8      8487   8.5%
 10      2701   2.7%
 12       152   0.2%
Average score: 4.52

Count of how many times player scores X with 3D+5R+3P (100000 total tests)
  2     11776  11.8%
  3     20492  20.5%
  4     32769  32.8%
  5     11773  11.8%
  6     12853  12.9%
  8      7652   7.7%
 10      2536   2.5%
 12       149   0.1%
Average score: 4.40

Count of how many times player scores X with 3D+6R+3P (100000 total tests)
  2     18105  18.1%
  3     22579  22.6%
  4     32855  32.9%
  5      7672   7.7%
  6     11719  11.7%
  8      5580   5.6%
 10      1429   1.4%
 12        61   0.1%
Average score: 4.04