Improved Scorch Calculations
#1
I've been playing around with my mage for a bit in light of the new improved scorch changes, and my experience has been that it's certainly a bunch of fun getting huge scorch hits when full vulnerability debuffs have been applied and with somebody to curse of elements around.

As a note, this is verbatim copy/pasted the same post I've made to a couple other forums, in an effort to get some sort of other input, which so far has not happened.

However the talent build I've been using has provided me with a bit of a speculatory problem.

I'm aiming for ultimately a breakdown as follows:

30 Fire / 21 Arcane

Arcane:

2 Arcane Focus (or whatever, the main point here being to get to PoM)
5 Improved Arcane Missiles
5 Improved Arcane Explosion
5 Arcane Concentration
2 Improved Counterspell
1 Evocation
1 Presnce of Mind

Fire:

5 Impact
5 Ignite
2 Flame Throwing
2 Incinerate
3 Improved Flamestrike
1 Blast Wave
3 Critical Mass
4/5 Improved Scorch
4/5 Fire Power

The dilemma being whether to go with 5/5 Improved Scorch and 4/5 Fire power or the other way around. I'm not looking to criticize or develop this build, I'm playing around with it and am curious about the calculations below, not the merits/problems with the build.

I've tried a few more calculations. The basic idea here being considering constant scorching against a single target, one with 5/5 imp. scorch and the other with 5/5 fire power. I'm not terribly aware of all the interesting effects of crits or clearcasts on these calculations, but if somebody wants to help me work through it, I'd be glad to recalculate. These are basically just numbers lifted from the highest rank of scorch without regard for much else.

Consider, in the case with improved scorch at 5, and fire power at 4, expected damage over 10 scorches is like this:

01) 285.12
02) 290.82
03) 296.52
04) 302.23
05) 307.93
06) 313.63
07) 313.63
08) 313.63
09) 313.63
10) 313.63
-----
Total = 3050.78

Now, if we're using firepower, we can only make estimates based on probability of improved scorch taking effect.

So, the first scorch starts out higher, from the base 2% increase.

1) 290.4

Now the second scorch has a .8 probability of doing an additional 2% damage. As such, I'm going to try and do the rest of the calculations based on the binomial distribution with n trials (n being the number of the scorch we're on) and .8 chance of success.

From this, the second scorch is pretty easy. Either the first one did or didn't succeed, with p = .8 for success. So, damage_2 = base_avg + .02 * .8 * base_avg

2) 295.05

Here, it gets a little more complicated, though I'll detail the steps so people can double check me as I won't claim to be an expert on any of this. The probability of having 2 scorches succeed out of 2 previous scorches = (.8)^2 where 2 here is the number of previous trials = .64. Similarly, the chance of two failures is (.2)^2 = .04 And finally, the chance of one success and one failure is .32 = (2 choose 1)*(.8)*(.2). So from this, we get the expected damage to be (.64)*(base_avg + .04*base_avg)+(.32)*(base_avg + .02*base_avg)+(.04)*(base_avg) =

3) 299.69

I'm going to spare you the math on the rest of these, to save on readability which I imagine I've already completely abandoned, but, if these look off, let me know and I'll try and clarify or correct myself.

4) 304.34
5) 308.99
6) 313.63
7) 316.76
8) 318.36
9) 319.05
10) 319.31

For an expected average total of: 3085.56

In table form, these line up as follows (scorch on top, firepower on bottom, and I apologize if the tabs don't seem to paste in quite right):
Code:
1      2      3      4      5      6      7      8      9      10      Total
285.12    290.82    296.52    302.23    307.93    313.63    313.63    313.63    313.63    313.63    3050.78
290.40    295.05    299.69    304.34    308.99    313.63    316.76    318.36    319.05    319.31    3085.56
For reference, the equation I used for the damage of the 10th scorch using firepower rather than improved scorch looks like this:
Code:
=BINOMDIST(0;9;0.8;0)*(D44)
+ BINOMDIST(1;9;0.8;0)*(0.02*D44 + D44)
+ BINOMDIST(2;9;0.8;0)*(D44 + D44*0.04)
+ BINOMDIST(3;9;0.8;0)*(D44 + 0.06*D44)
+ BINOMDIST(4;9;0.8;0)*(D44 + D44*0.08)
+ BINOMDIST(5;9;0.8;0)*(D44 + D44*0.1)
+ BINOMDIST(6;9;0.8;0)*(D44 + D44*0.1)
+ BINOMDIST(7;9;0.8;0)*(D44 + D44*0.1)
+ BINOMDIST(8;9;0.8;0)*(D44 + D44*0.1)
+ BINOMDIST(9;9;0.8;0)*(D44 + D44*0.1)
where BINOMDIST(a,b,c,d) gives the binomial distribution for a specifically a successes in b trials with c = probability of success. D44 is the cell in which the average damage for a scorch is.

And this is ignoring things like the fact that incinerate with more firepower will do more damage on crits because the incinerate damage is calculated, so I'm told, with bonuses like firepower in mind, among a vast number of other things.

Now from these numbers, I conclude that it seems that there is no point in a single mage hitting a single target with continuous scorches that there is justification from a damage standpoint to go with 5/5 Imp. Scorch.

However, here's where I really need some more help: I want to try and consider situations such as a warlock with an imp or something, or a hunter or shaman doing flame damage as well, or hell, even another fire mage around. In any of these situations, does maxing the vulnerabilty debuff immediately give sufficiently large enough expected damage increase to justify using Imp. Scorch 5/5 instead of Fire Power? That is, if I knew I was going to be playing mainly with some friend of mine who was also capable of doing fire damage, would I choose talents differently?
Reply


Messages In This Thread
Improved Scorch Calculations - by waloo - 05-02-2005, 04:04 PM
Improved Scorch Calculations - by savaughn - 05-02-2005, 09:11 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)