Official Rules Updates (March 02, 2010)

[Aduro]

Want to break that down for those of us not mastered in probability? Actual math examples maybe?

 

[Victim]

Basically, you take half as much damage if people need 20 to hit you versus them needing a 19 (barring crits and such). It's only a change of a 5% percent increment, but the character only hit on a 20 will live twice as long. As you move further from that extreme, the returns are much less dramatic.

 

[Sonofbizzkitz]

New To D&D

Ok I have played D&D in the past (1st edition). Some friends and I decided to take it up again with me as the DM. They rolled characters and I inputed some of the info online. I used 4th edition online thing and the saving throws are a lot higher than 1st edition.
They asked me why? I said "Oreo" and we cant decide if we should just stick with first edition or go 4,(I went and bought 4E Manulals already and I would really like to run 4E!
we were all wondering why some of the stats are different.
Should we go 4E or stick with 1??

 

[KarinsDad]

The problem is that AC is just that good. And each point is better than the last one. Going from 55% to be hit to 50% chance to be hit is a 10% improvement; going from 10% chance to be hit to 5% chance to be hit is a 100% improvement!

The point is that it's a feat tax because many players think that the PC cannot be played without the feat (which btw I am doing for my LEB sorcerer, so it can be done, just most players don't see it).

A player of a Fighter might say "Meh. Plate armor? I don't want to waste the feat. Scale (and shield) is good enough."

Most players of Clerics don't say that. Most players of Wizards and Sorcerer don't say that.

They think that Chain with no shield and especially Cloth are not even close to "good enough". AC 14 or 15 when most of the rest of the team are at AC 16 to 20, seems unplayable. They think that they have to take that feat. It becomes a must take or a feat tax.

 

[BobTheNob]

Ah you see, there is the problem. The actual increase always is (P1-P2/P2) where P1 is the probability after the change and P1 is the original probability; that is because an increase in chance for something to happen from a previous state is a conditional probability. So all probabilities that involve a roll of 1d20 + modifiers face diminishing or increasing effect depending on the direction we are moving at. In short, in the d20 system any increase in defensive DCs presents increasing returns and any increase in offensive d20 rolls faces decreasing returns.
The same does not stand if we use a bell curve approximation (3d6 works well on that). This concept lies in the very foundation of the d20 system and largely defines the feel of combat in this system in a way that is poorly understood by most.

A very simple way of thinking about this...

Lets say an enemy would need a 11 to hit you (50% average). Lets say you somehow get a +1 boost to your defence, and the enemy therefore needs a 12 to hit you (now 45%).

Has the chance he hits you changed by 5%? Yes. Has the chance that he hits you decreased by 5%? No. Its 10%

By saying 5% you are comparing the factor of the change to 100% (i.e. 5/100) where you should be comparing the shift(5%) to the original probability (50%), which is 5/50, or 10%

Same scenario, but lets say the enemy needed a 19 (10%). You get the same +1 bonus to defence so he now needs a 20 (5%) (leave crits out of debate for now)

Has the chance he hits you changed by 5%? Yes. Has the chance that he hits you decreased by 5%? No. Its more like 50%

The change that he hits you has decreased by 5/10 or 50%

This is very rough an non-formulaic (just cant be stuffed right now) but this is the principal of how probability works on an ajusted flat d20 roll. 1 <> 5%, and it scales dramatically at the extreme end of the probability.

What you have to think is not how often every 20 rolls he hits, but how often he hits before, compared to how often he does after. By taking a creature fro 10 hits every 20 to 9 hits every 20, the impact is fairly minimal, but by taking a creature from 2 hits every 20 to 1 every twenty, you have halved hit hit rate...same +1, completely different effect.

If you feel you can buy into debates over when a class(/other) has a number that is too high (like the debates over avenger AC) is is ESSENTIAL you understand this principal, cause if you dont, your just wasting everyones time.

 

[DracoSuave]

Want to break that down for those of us not mastered in probability? Actual math examples maybe?

Okay, not so much text as above.

5% extra probability does not mean 5% more outcomes.

If your chances to hit are 10 in 20 outcomes, then a 10% extra chance to hit means 2 extra outcomes, which is actually 20% more outcomes.

20% more hits.

20% more damage.

 

[Elric]

A very simple way of thinking about this...

Lets say an enemy would need a 11 to hit you (50% average). Lets say you somehow get a +1 boost to your defence, and the enemy therefore needs a 12 to hit you (now 45%).

Has the chance he hits you changed by 5%? Yes. Has the chance that he hits you decreased by 5%? No. Its 10%

By saying 5% you are comparing the factor of the change to 100% (i.e. 5/100) where you should be comparing the shift(5%) to the original probability (50%), which is 5/50, or 10%

Same scenario, but lets say the enemy needed a 19 (10%). You get the same +1 bonus to defence so he now needs a 20 (5%) (leave crits out of debate for now)

Has the chance he hits you changed by 5%? Yes. Has the chance that he hits you decreased by 5%? No. Its more like 50%

The change that he hits you has decreased by 5/10 or 50%

This is very rough an non-formulaic (just cant be stuffed right now) but this is the principal of how probability works on an ajusted flat d20 roll. 1 <> 5%, and it scales dramatically at the extreme end of the probability.

What you have to think is not how often every 20 rolls he hits, but how often he hits before, compared to how often he does after. By taking a creature fro 10 hits every 20 to 9 hits every 20, the impact is fairly minimal, but by taking a creature from 2 hits every 20 to 1 every twenty, you have halved hit hit rate...same +1, completely different effect.

If you feel you can buy into debates over when a class(/other) has a number that is too high (like the debates over avenger AC) is is ESSENTIAL you understand this principal, cause if you dont, your just wasting everyones time.

They problem with this kind of analysis, though, is that plenty of attacks don't target AC (though it's true that AC is probably targeted ~ as much or more than all other defenses put together, depending on your level).

People apply this logic to (incorrectly) argue that you should increase high FRW defenses more than low ones, even if the low ones aren't at "would be hit on a 1" level and the high ones aren't any more targeted than the low ones. Quoting from a post on WotC's boards:

I believe the rule of thumb is that each increasing attack bonus brings diminishing returns, and by the same token each further increase in defense brings increasing returns. Increasing your low defenses (assuming it goes from a 2 to hit to more) actually reduces more damage than increasing a high defense.

This isn't true. If a defense is very low relative to the attacker's attack bonus, a plus to that defense won't give you the same benefit. For many characters/opponents, you won't get the full benefit of a FRW-boosting feat because the monster would have hit you on a 1 if not for the auto-miss rule, pre-feat (e.g., "it hits on a -1 or better"), so some of the benefit of a bonus (e.g., that from a feat) will be lost. Setting that possibility aside,

Dropping a 70% hit rate does less than dropping a 40%. I don't think it's worth it.

Here's the essential reason this isn't correct: If you have two enemies, one of whom needs an 18 to hit you, and the other of whom needs a 2 to hit you, and they do equal damage, the average damage you take if both attack you once is the same whether you get a +2 bonus to the defense where you're hit on an 18 or the defense where you're hit on a 2. Just because you chose to take a +2 to your stronger defense doesn't mean you can ignore the attacks on your weak defense! AC is much more frequently targeted than any FRW defense, though, so given the choice between boosting AC by 1 and boosting one FRW by 1, AC almost certainly a better choice.

 

[DracoSuave]

They problem with this kind of analysis, though, is that plenty of attacks don't target AC (though it's true that AC is probably targeted ~ as much or more than all other defenses put together, depending on your level).

Plenty of d20 rolls have nothing to do with attacking either.

That doesn't change how the probabilities work one bit.

People apply this logic to (incorrectly) argue that you should increase high FRW defenses more than low ones, even if the low ones aren't at "would be hit on a 1" level and the high ones aren't any more targeted than the low ones. Quoting from a post on WotC's boards:

If and only if all NADs are targetted equally as often, then you have:

1/3 of all attacks against Fortitude
1/3 against Will
1/3 against Reflex

Which means that the total number of outcomes for 'gets hit' is the same over all NAD attacks.

Taking 4 outcomes off of a low NAD has no difference on the total number of hits than 4 outcomes off of a high NAD.

Therefore, it doesn't matter WHICH NAD you boost. All are just as good.

Otherwise, the only rational choice is to boost the one targetted most often, regardless if it is high or low.

 

[KarinsDad]

If and only if all NADs are targetted equally as often, then you have:

1/3 of all attacks against Fortitude
1/3 against Will
1/3 against Reflex

Which means that the total number of outcomes for 'gets hit' is the same over all NAD attacks.

Taking 4 outcomes off of a low NAD has no difference on the total number of hits than 4 outcomes off of a high NAD.

Therefore, it doesn't matter WHICH NAD you boost. All are just as good.

If one does not consider riders. Attacks vs. Reflex, for example, almost always have damage riders with maybe some minor modifier to defense or to hit. Not so bad. But Stun or Daze or Immobilize or Blinded can be quite a bit more devastating.

Otherwise, the only rational choice is to boost the one targetted most often, regardless if it is high or low.

Typically true.

 

[Colmarr]

Never mind. Forked instead