Players Roll All Dice Variant is Wrong! [Archive] - Wizards Community

Players Roll All Dice Variant is Wrong! [Archive] - Wizards Community

Post/Author/DateTime Post
ManinDarkness
08-01-04, 06:58 PM Look at this:

Let's say Abe (PC) has an AC 18 and Bail (NPC) has an attack bonus of +10. Bail hits Abe if he gets 8 or more. That is, he hits 13 times out of 20.

UA says Abe Defense Bonus is +8 and Bail's Attack Score is 21. That means Abe fends off the attack if he gets 13 or more. He doesn't gets hit 8 times out of 20. Which is the same as saying he gets hit 12 times out of 20.

??????

All 11's in the Variant should be 12's!!!!!!
schulerta
08-02-04, 04:26 PM Very true. All your calculations are dead on.

I have rationalized it this way: Tie goes to the roller.

Troy
Lina_Inverse
08-06-04, 01:05 PM Originally posted by ManinDarkness
Look at this:

Let's say Abe (PC) has an AC 18 and Bail (NPC) has an attack bonus of +10. Bail hits Abe if he gets 8 or more. That is, he hits 13 times out of 20.

UA says Abe Defense Bonus is +8 and Bail's Attack Score is 21. That means Abe fends off the attack if he gets 13 or more. He doesn't gets hit 8 times out of 20. Which is the same as saying he gets hit 12 times out of 20.

??????

All 11's in the Variant should be 12's!!!!!!

....um no,they should all be 10.5s average of a d20 is 10.5
Elzaban the Ethereal
08-07-04, 07:16 PM The "Player's Role All Dice" variant is correct (whether or not it's morally wrong to use it is a different matter ;)). The main error that I can see in your calculations is that one of the numbers in the results (13 and 8) overlaps. Because you win if you tie or roll higher, there will be one number that yeilds a different result depending on who's rolling. In this case, if Bail were to roll an 8, he would hit. If Abe, however, were to roll an 8, Bail would miss. It's as simple as that.
ManinDarkness
08-14-04, 09:48 PM Originally posted by Lina_Inverse
....um no,they should all be 10.5s average of a d20 is 10.5

It is not a matter of averages is a matter of algebra:

Meaning:

1d20 = 21 - 1d20

So: 1d20 + Att >= 10 + AC for Bail to hit.
Then: 21 - 1d20 + Att >= 10 + AC for Bail to hit.
Then: 11 + Att >= 1d20 + AC for Bail to hit.
Which means: 1d20 + AC >= 12 + Att for Abe to fend off the attack.

Originally posted by Elzaban the Ethereal
The "Player's Role All Dice" variant is correct (whether or not it's morally wrong to use it is a different matter ;)). The main error that I can see in your calculations is that one of the numbers in the results (13 and 8) overlaps. Because you win if you tie or roll higher, there will be one number that yeilds a different result depending on who's rolling. In this case, if Bail were to roll an 8, he would hit. If Abe, however, were to roll an 8, Bail would miss. It's as simple as that.

Nope, that is not what the rule says. If Abe rolls an 8 he is being hit. The rule says that Abe has to roll the attack score or more to fend off the attack.

And, isn't a DM naturally amoral? ;)
Elzaban the Ethereal
08-15-04, 02:09 AM Wow... I mustn't have been thinking clearly when I wrote that explaination. Let's hope I'm thinking clearer this time around.

The reason that one of them can have 13 possibilities of success and the other can have 8, is that each of them has a possible roll that creates a tie. A tie can go either way, so it would be better counted as a half in normal circumstances (12.5 and 7.5, respectively). However, in D&D, if they person who's rolling gets a tie, they're the one who succeeds, so each of the ties has to become whole possibility of success, creating the seemingly overlapping amounts of success of 13 and 8..
ManinDarkness
08-18-04, 11:17 PM Originally posted by Elzaban the Ethereal
A tie can go either way

Nope, Ties always go to the roller. Thus, there is no overlapping probability.
Elzaban the Ethereal
08-19-04, 02:02 PM Could you have possibly read the sentence after that one, too?
However, in D&D, if they person who's rolling gets a tie, they're the one who succeeds, so each of the ties has to become whole possibility of success,
A tie going either way was refering to usual non-D&D situations. Usually a tie represents a win and a lose simultaneously, but with D&D, it always goes to the roller, which is why you get that "extra" number - the two half-chances of winning the roll become full chances.
ManinDarkness
08-20-04, 11:48 PM Nevertheless, probabilities shouldn't have changed. Unless you think that making players roll the dice should grant them a +1 bonus on every extra roll.
Elzaban the Ethereal
08-21-04, 08:37 AM Hmm, you do bring up a good point there. But I don't feel that making the DC a 12+Mod is the way to solve it. It's almost like taking the tie away from them, yes, but I don't feel that the tie mechanic is great in the first place. That's where the error with the Players Roll All Dice varient is, so I propose the following:

Whenever anyone rolls a tie on a d20 check (attack roll, caster level roll, armor class roll, etc), roll another (unmodified) d20. If the result is 10 or less, the person making the check fails. If the result is 11 or above, however, the check succeeds.

Post/Author/DateTime	Post
ManinDarkness 08-01-04, 06:58 PM	Look at this: Let's say Abe (PC) has an AC 18 and Bail (NPC) has an attack bonus of +10. Bail hits Abe if he gets 8 or more. That is, he hits 13 times out of 20. UA says Abe Defense Bonus is +8 and Bail's Attack Score is 21. That means Abe fends off the attack if he gets 13 or more. He doesn't gets hit 8 times out of 20. Which is the same as saying he gets hit 12 times out of 20. ?????? All 11's in the Variant should be 12's!!!!!!
schulerta 08-02-04, 04:26 PM	Very true. All your calculations are dead on. I have rationalized it this way: Tie goes to the roller. Troy
Lina_Inverse 08-06-04, 01:05 PM	Originally posted by ManinDarkness Look at this: Let's say Abe (PC) has an AC 18 and Bail (NPC) has an attack bonus of +10. Bail hits Abe if he gets 8 or more. That is, he hits 13 times out of 20. UA says Abe Defense Bonus is +8 and Bail's Attack Score is 21. That means Abe fends off the attack if he gets 13 or more. He doesn't gets hit 8 times out of 20. Which is the same as saying he gets hit 12 times out of 20. ?????? All 11's in the Variant should be 12's!!!!!! ....um no,they should all be 10.5s average of a d20 is 10.5
Elzaban the Ethereal 08-07-04, 07:16 PM	The "Player's Role All Dice" variant is correct (whether or not it's morally wrong to use it is a different matter ;)). The main error that I can see in your calculations is that one of the numbers in the results (13 and 8) overlaps. Because you win if you tie or roll higher, there will be one number that yeilds a different result depending on who's rolling. In this case, if Bail were to roll an 8, he would hit. If Abe, however, were to roll an 8, Bail would miss. It's as simple as that.
ManinDarkness 08-14-04, 09:48 PM	Originally posted by Lina_Inverse ....um no,they should all be 10.5s average of a d20 is 10.5 It is not a matter of averages is a matter of algebra: Meaning: 1d20 = 21 - 1d20 So: 1d20 + Att >= 10 + AC for Bail to hit. Then: 21 - 1d20 + Att >= 10 + AC for Bail to hit. Then: 11 + Att >= 1d20 + AC for Bail to hit. Which means: 1d20 + AC >= 12 + Att for Abe to fend off the attack. Originally posted by Elzaban the Ethereal The "Player's Role All Dice" variant is correct (whether or not it's morally wrong to use it is a different matter ;)). The main error that I can see in your calculations is that one of the numbers in the results (13 and 8) overlaps. Because you win if you tie or roll higher, there will be one number that yeilds a different result depending on who's rolling. In this case, if Bail were to roll an 8, he would hit. If Abe, however, were to roll an 8, Bail would miss. It's as simple as that. Nope, that is not what the rule says. If Abe rolls an 8 he is being hit. The rule says that Abe has to roll the attack score or more to fend off the attack. And, isn't a DM naturally amoral? ;)
Elzaban the Ethereal 08-15-04, 02:09 AM	Wow... I mustn't have been thinking clearly when I wrote that explaination. Let's hope I'm thinking clearer this time around. The reason that one of them can have 13 possibilities of success and the other can have 8, is that each of them has a possible roll that creates a tie. A tie can go either way, so it would be better counted as a half in normal circumstances (12.5 and 7.5, respectively). However, in D&D, if they person who's rolling gets a tie, they're the one who succeeds, so each of the ties has to become whole possibility of success, creating the seemingly overlapping amounts of success of 13 and 8..
ManinDarkness 08-18-04, 11:17 PM	Originally posted by Elzaban the Ethereal A tie can go either way Nope, Ties always go to the roller. Thus, there is no overlapping probability.
Elzaban the Ethereal 08-19-04, 02:02 PM	Could you have possibly read the sentence after that one, too? However, in D&D, if they person who's rolling gets a tie, they're the one who succeeds, so each of the ties has to become whole possibility of success, A tie going either way was refering to usual non-D&D situations. Usually a tie represents a win and a lose simultaneously, but with D&D, it always goes to the roller, which is why you get that "extra" number - the two half-chances of winning the roll become full chances.
ManinDarkness 08-20-04, 11:48 PM	Nevertheless, probabilities shouldn't have changed. Unless you think that making players roll the dice should grant them a +1 bonus on every extra roll.
Elzaban the Ethereal 08-21-04, 08:37 AM	Hmm, you do bring up a good point there. But I don't feel that making the DC a 12+Mod is the way to solve it. It's almost like taking the tie away from them, yes, but I don't feel that the tie mechanic is great in the first place. That's where the error with the Players Roll All Dice varient is, so I propose the following: Whenever anyone rolls a tie on a d20 check (attack roll, caster level roll, armor class roll, etc), roll another (unmodified) d20. If the result is 10 or less, the person making the check fails. If the result is 11 or above, however, the check succeeds.