Is an 18/19/20 an absolute must?

[Blackbrrd]

IMHO this distinction is unnecessarily confusing for the exact people who aren't good at math.

IMHO a more intuitively clear way to say it is: +1 to attack is usually +5% expected damage.

Cheers, -- N

No it isn't a +1 to attack is usually about +10%* to expected damage, that is why I brought it up.

As noted above, you go from hitting 50% to hitting 55% of the time. An increase of 5 percentage points (very different from 5% Percentage point - Wikipedia, the free encyclopedia ).

Let us say you are using a longsword for 1d8+5 damage,
your average damage with 50% to hit chance is: 9.5*50% = 4.75
your average damage with 55% to hit chance is: 9.5*55% = 5.225

5.225 is 10% more than 4.75.

*The base scenario for this is that you hit 50% of the time before the bonus to hit.

 

[Mengu]

IMHO a more intuitively clear way to say it is: +1 to attack is usually +5% expected damage.

No it isn't. It's more like +10% (ish) expected damage, but is a bit more complicated then that because of a multitude of reasons, from the fact that hit chance is not always 50%, to crits, multiple attacks, rerolls, etc.

 

[Blackbrrd]

Devoted Challenge is the Dwarf's solution to that, if it becomes a problem.

Looked up the feat, and I agree, which then makes the Dwarf Fighter with 16 starting str seem quite viable.

It is just about the only character with tricks like that up the sleeve though.

 

[Nifft]

Let us say you are using a longsword for 1d8+5 damage,
your average damage with 50% to hit chance is: 9.5*50% = 4.75
your average damage with 55% to hit chance is: 9.5*55% = 5.225

5.225 is 10% more than 4.75.

Yes, if you chop a number in half (9.5 * 50%) then the increase of 5% to the original number "doubles" to 10% relative to the new number, because the new number is half of the original number.

That's what I'm saying is confusing, and it's entirely possible that I phrased my previous post poorly, because I do indeed find this stuff confusing.

How about: +1 to attack grants you approximately a 5% increase in the proportion of your base damage you can expect to inflict. Unlike the internal/external view, this proportion doesn't depend on assuming any particular base expectations.

No it isn't. It's more like +10% (ish) expected damage, but is a bit more complicated then that because of a multitude of reasons, from the fact that hit chance is not always 50%, to crits, multiple attacks, rerolls, etc.

The rider effects can be calculated, but it is tougher. Look at a Str-based Cleric for example: his at-will attack has an x% chance of giving the next PC a +Str bonus.

If you look at the next PC's attack in terms of expected damage, though, you can see directly how a +1 to the Cleric's attack translates into a +(Str*5)% increase in the proportion of base damage that can be expected to result. So it's pretty clear that a +1 to the Cleric's attack yields a very significant payout.

Cheers, -- N

 

[Infiniti2000]

As noted above, you go from hitting 50% to hitting 55% of the time. An increase of 5 percentage points (very different from 5% Percentage point - Wikipedia, the free encyclopedia ).

If you knew this, then why'd you ask what 5% difference he was talking about? In any case, statistics is "great" just for this reason. You can both take that 5% and massage it with other numbers and words to color your statements in the best way possible. You compute the difference in rolls with a heavy caveat on the base chance to hit. He simply subtracts the new chance to hit from the old chance and (correctly) calls that a difference (no caveat). It's a lower value, thus it looks like less. Fwiw, I agree with viewing it as a 10% increase.

As yet another aside, the percent difference is 9.5%.

 

[Infiniti2000]

How about: +1 to attack grants you approximately a 5% increase in the proportion of your base damage you can expect to inflict. Unlike the internal/external view, this proportion doesn't depend on assuming any particular base expectations.

Exactly. This would be easier to follow with using a base damage of 10 (not 9.5). For each +1, you get +0.5 points of damage (+5%). The percentage increase, however, from a theoretical to-hit roll of 10 to 11 is +10%. The percent difference between the 10 and 11, on the other hand, is only 9.5%.

 

[Mistwell]

No it isn't a +1 to attack is usually about +10%* to expected damage, that is why I brought it up.

As noted above, you go from hitting 50% to hitting 55% of the time. An increase of 5 percentage points (very different from 5% Percentage point - Wikipedia, the free encyclopedia ).

Let us say you are using a longsword for 1d8+5 damage,
your average damage with 50% to hit chance is: 9.5*50% = 4.75
your average damage with 55% to hit chance is: 9.5*55% = 5.225

5.225 is 10% more than 4.75.

*The base scenario for this is that you hit 50% of the time before the bonus to hit.

Hypothetically, if I already have a +20 to attack at first level, making it +21 to attack isn't really increasing the damage at all. In that hypothetical, you're going to hit against virtually anything, almost all the time. Obviously I am demonstrating using an unrealistic example, but it clearly shows one flaw with always analyzing based on a 50-50 chance.

Let's take a 4th level dwarven guardian fighter with Dwarven Weapon Training, Weapon Expertise, and Plate Armor Proficiency, that started with 16 str, 16 con, and 16 wis. You attack with brash strike (which grants +2 attack). So you attack with a +3 (str) +2 (craghammer proficiency) +1 (weapon expertise) +2 (brash strike) +2 (level) +1 (magic weapon) +1 (weapon talent) for a total of +12 vs. AC.

The difference between a +12 and a +13 at 4th level isn't as much as your hypothetical average example. Most foes have an AC of about "level + 14". So I am hitting on an average roll, regardless of whether it is a +12 or +13 attack bonus. The additional +1 will make a difference sometimes, but not as often as the "I hit exactly half the time" analysis.

So while I agree it's OK to start from a 50-50 hit probability and 10% change from a +1 attack bonus, that's not where it ends. When you get into actual likely builds, it's sometimes not a particularly accurate method of computing the impact of an additional +1 to attack.

 

[AbdulAlhazred]

Hypothetically, if I already have a +20 to attack at first level, making it +21 to attack isn't really increasing the damage at all. In that hypothetical, you're going to hit against virtually anything, almost all the time. Obviously I am demonstrating using an unrealistic example, but it clearly shows one flaw with always analyzing based on a 50-50 chance.

Let's take a 4th level dwarven guardian fighter with Dwarven Weapon Training, Weapon Expertise, and Plate Armor Proficiency, that started with 16 str, 16 con, and 16 wis. You attack with brash strike (which grants +2 attack). So you attack with a +3 (str) +2 (craghammer proficiency) +1 (weapon expertise) +2 (brash strike) +2 (level) +1 (magic weapon) +1 (weapon talent) for a total of +12 vs. AC.

The difference between a +12 and a +13 at 4th level isn't as much as your hypothetical average example. Most foes have an AC of about "level + 14". So I am hitting on an average roll, regardless of whether it is a +12 or +13 attack bonus. The additional +1 will make a difference sometimes, but not as often as the "I hit exactly half the time" analysis.

So while I agree it's OK to start from a 50-50 hit probability and 10% change from a +1 attack bonus, that's not where it ends. When you get into actual likely builds, it's sometimes not a particularly accurate method of computing the impact of an additional +1 to attack.

On the other hand your analysis is entirely correct as far as it goes, but there are plenty of additional things to consider. Conditions are of course one that is often mentioned because they are all or nothing and thus value to-hit more than damage does. Miss effects and reliable powers also have a VERY considerable effect and Reaping Strike for instance does on average a fairly phenomenal amount of damage and is little affected by to-hit chance on a proportional basis.

However even all of these things are not taking into consideration the tactical implications of hitting and doing damage. Hitting weak monsters is often less valuable than hitting powerful ones. Weak monsters usually have weak powers used against them for one thing. They have generally weak defenses and thus boosting your to-hit in cases where you're facing a weak opponent is less valuable. OTOH they are also a lot less of a threat, so it matters less. Hitting the STRONG opponents though, which have generally higher defenses is even more critical. You don't want to miss them and you often deploy encounter or daily powers against them that you REALLY want to hit with. Even a character that is nominally hitting equal level monsters on a 6+ may find that it still makes sense to have a bit higher to-hit bonus just for that time when you're after the level+5 Solo Soldier monster that's eating your lunch. Of course you now have to factor in what kind of leader support you get in that case and decide if its better to up your to-hit more or just get the leader to stock some extra buffs.

Really its also not worth studying any of these numbers in isolation from other things you may be gaining/losing by increasing a starting stat. It won't do you a lot of good to have a really high to-hit number if your character drowns, falls off the side of a cliff, or gets killed by a pit trap because he only had one really good stat.

 

[Felon]

It's worth stating--and perhaps even restating--that aggregate bonuses are what you should be looking at. If you're a fighter without a racial bonus to Strength, you need to cover the spread in some other area. Maybe instead of taking a +2 weapon with some extra magical property, you should try to acquire a +3 weapon. Or perhaps design your character around a sword of some kind for its +3 proficiency bonus.

 

[Mengu]

Maybe instead of taking a +2 weapon with some extra magical property, you should try to acquire a +3 weapon.

With a handful of exceptions, when presented the option of +2 weapon with property vs +3 vanilla weapon, I think most characters would choose the vanilla +3 anyway. +1 attack and damage as a "weapon property" if you will, is just too good to pass up in most cases. But that can get boring fast, and fortunately DM's provide the items for a game, often working off wishlists, and as such you are rarely likely to see a vanilla +X item.