Almost as much as including an unknown to hit vs and unknown AC. lol I do agree if your not looking for max possible damage and your looing for average damage not including accuracy means your results are not accurate. *ta da chi*
...But I would also understand posting at 100% hit since you get stuck in Range vs opponent X and Melee vs opponent y then you get in consistent answers in the same agreement over and over again that does actually add anything but "player and GM choices make comparing damage shifting sand." Which is also true.
Hmm. Posting up values already produces a wall of text, and it is sometimes difficult to know where to strike the balance of what to include.
First the characters are points buy, in order to compare like to like. Level required thought. People sometimes "prove" that something is okay (or not okay) by choosing an arbitrary level where the numbers stack up the way they want them (due to access to relevant features). I use the following = Tier 1 is 4th, Tier 2 is 8th, Tier 3 is 12th. My first concern was to choose levels that represent what people might realistically play. The chosen levels span the main arc of play for most campaigns in my experience, and according to the designers. I distribute ASIs to stat, then key feat, then stat again. A character could go earlier into the key feat, but accuracy and survivability concerns justify waiting (for instance, the power-attacks rely on a threshold of accuracy to be effective). Finally, I give all T3 characters two basic magic items relevant to their class. Usually a +1 weapon and a +1 defence or stat buff.
Then for foes, I use the DMG guidelines for encounters and monster scaling, so that -
Level 4 = 375xp = CR 1 or 2 = AC 13
Level 8 = 1400xp = CR 4 or 5 = AC 14 or 15
Level 12 = 3000xp = CR 7 or 8 = AC 15 or 16
That then suggests the ACs / saves -
T1 AC 13 +3 save
T2 AC 14 +4 save
T3 AC 15 +5 save
The result is that accuracy slightly out scales foe defenses (mirroring what I have seen in play at the table). So that Barbarian is T1= 0.7 T2 = 0.7 T3 = 0.8, then always Reckless, therefore the power-attack is T2 = 0.7 and T3 = 0.8. The maths is straightforward e.g. for T2 (0.7-0.25)=0.45 and 1-(1-0.45)^2 = 0.7... surprisingly good, but achieved only with persistent advantage.
There are some other factors to consider. For instance, in my experience it is very common for melee to lose one or two turns a fight closing or repositioning. Ranged lose a turn every few fights to kiting. Any character with insufficiently good defences, who is in the path of trouble, loses a few turns every few fights to dying. Thus for an assumed three encounters per day, five rounds per encounter, I factor melee as 12 effective rounds and ranged as 15. That is why it is so important to see the "maximum average" which removes such factoring in case some group's experiences are quite at odds with that assumption.