Filling my dream of a DB scrapper.

Arcanaville did a pretty good job of explaining tohit vs acc in this PM i sent her about a month ago. She gave me this general rule of thumb, as i was comparing set bonus +acc to the kismet +tohit, while trying to factor in base acc (from normal enhancements) as well as other +tohit (from focused acc, or tactics) So this shoud answer your question abit about which is better, and at which point one can overcome the other.

Alright, finally responding again. That rule of thumb above is probably going to be really handy for you. Let's make sure you've got all the information you need to apply it.

First, let's make sure you know about the "base toHit" Arcanaville mentioned. This is the basis of the number you see in your in-game attribute monitor if you monitor your toHit Chance. Against even-level foes, that's 75%. If the foe is higher level, the chance goes down, like so.

+0: 75%
+1: 65%
+2: 56%
+3: 48%

It keeps going down from there, but that's as much of the table as I've ever memorized. Let us know if you really want to know +4s or higher - I'd have to dig it up.

In the absence of all other factors, like defense (de)buffs or toHit (de)buffs, your chances of hitting a foe with a given attack is that attack's accuracy times your base toHit chance for that foe.

A power's accuracy actually has two components. The more well-known one is the part we can slot, and comes out to 100% + slotted accuracy. If we have any set bonuses or powers granting us global accuracy, they add to our final slotted accuracy. (By "final" I mean the game calculates ED's effect on our slotted ACC and then adds power and set bonuses.) All powers also have an inherent accuracy. Most damaging attacks and single-target mezzes have a 100% inherent accuracy, but some have more or less than 100%. For example, snipes have 120%, most AoE mezzes have 80%, and attacks in the Martial Arts, Broadsword and Katana powersets have 105%. You get a power's total accuracy by multiplying the inherent part times the final slotted/bonus part.

Let's look at an example. Say you've got a MA Scrapper with Cobra Strike. You have it slotted with level 40 Touch of Death set pieces, giving 43.4% slotted Accuracy . You have 20% other global ACC from set bonuses.

Your slotted accuracy is (100% + 43.4% + 20%). The power's inherent accuracy is 105%. So it's final accuracy is 163.4% * 105% = 171.6%

Let's say you're attacking a +3 foe with Cobra Strike. Your chance of hitting him is 171.6% * 48% = 82.4%.

It sounds like you understand how toHit and defense (de)buffs play in this, but just in case... Defense on your target makes its effective toHit number go down by the defense percentage. So if your +3 foe had 10% defense, that would make his effective toHit 38% instead of 48%, and your total chance would lower to 171.6% * 38% = 65.2%. Likewise, toHit would make the effective toHit number go up by the toHit buff percentage. Defense and toHit debuffs are just defense and toHit buffs with negative signs, and so just move things in the opposite direction.

So to answer your original question, accuracy doesn't really counter anything in a simple, direct sense in the way that toHit counters defense. (At least not in PvE. They added something that it counters in PvP, but as far as I know it doesn't exist in PvE.)

A far more comprehensive write-up of this is in Arcanaville's guide to (toHit and) defense.

Here's what I consider a simpler rule of thumb with almost no math:

1. Install Mids Hero Designer if for some reason you haven't.

2. Options -> Configuration -> Exemping & Base Values -> Base ToHit -> Enter the number for how high uplevel you're fighting:

+0: 75%
+1: 65%
+2: 56%
+3: 48%
+4: 39%
+5: 30%

3. Try out both the to hit buff and the accuracy buff you're considering. Mouse over the powers to see which helps your final chance to hit more.

1. To hit buffs are better than accuracy buffs.

While it is true, that point for point tohit is better then acc, due to the higher base chance tohit, and the fact everyone slots acc, it can makes things seem misleading. as Arcanaville goes onto explain: