4 replies to this topic

[Amaris the Usurper]

The purpose of this topic is to facilitate discussion (elsewhere) on weapon balance issues by building a database of observed per-missile damage numbers for various builds of the game. The focus is on typical use, not on extreme cases (e.g., shooting small 'mechs that take amplified splash damage).

Please post screenshots of your in-game stats along with a description of the circumstances under which they were collected. For example, you might post your stats as they appeared just before patching along with post-patch stats following a few hours of play, allowing damage numbers to be computed specifically for the patch in question.

Please keep the discussion on-topic and confined to official PGI in-game stats. This is not a place to debate the relative merits of various weapons as presently implemented.

Note: In what follows, "patch" refers to the 19 March patch (version 1.2.204).

I will kick off the discussion with the following analysis.

On the night before the hotfix, I dusted off my CPLT-C1, put 2 x LRM 20 with TAG on it, and played a handful of games, largely out of curiosity about LRM performance.

It turns out that these games were the only time that I have used LRM 20s since tracking of detailed stats began on 5 March. Therefore, my LRM 20 stats reveal typical pre-hotfix (but post-patch) damage numbers for LRMs. The stats are shown here: http://i.imgur.com/VDWmuyp.jpg.

LRM 20 w/ TAG post-patch, pre-hotfix
(3077 damage)/(1166 hit) = 2.64 damage/missile

This got me thinking about what else I could glean out of my stats. There turned out to be quite a bit of useful information.

Several days before the patch, I reported the following numbers from my stats, but (unfortunately) didn't get a screenshot. (See http://mwomercs.com/...ed-2013-03-15/; the numbers are in the first post.) The LRM 15s were fired from a Stalker with TAG.

LRM 15 w/ TAG pre-patch
206/179 = 1.15 damage/missile

SRM 4 pre-patch
808/302 = 2.68 damage/missile

SRM 6 pre-patch
4739/1977 = 2.40 damage/missile

Streak SRM 2 pre-patch
3884/1274 = 3.05 damage/missile

After I posted these numbers, I went on firing SRM 6s and 4s, but I didn't use LRM 15s or Streak SRM 2s. I stopped firing SRM 4s at some point before the patch, but I kept firing SRM 6s (from various platforms) right through both the patch and hotfix. My stats, as of last night, are shown here: http://i.imgur.com/VDWmuyp.jpg. (This is the same image linked above.)

Overall, since detailed stats tracking began, I had connected with 8543 missiles fired from SRM 6s, inflicting 25,938 damage. We can get a lower bound for how much damage SRMs were doing between the patch and hotfix by subtracting out the pre-patch hit and damage numbers.

SRM 6 post-patch, pre-hotfix (lower bound)
(25,938-4739)/(8543-1977) = 3.23 damage/missile

I suspect that the real numbers are quite a bit higher. Unfortunately, there is no way to get the information now, unless you know (for instance) that you used SRM 4s only between the patch and the hotfix. Here's hoping that you're out there somewhere.

After taking (last night) the previous shot of my stats, I played more games with LRM 15s, SRM 6s, and Streak SRM 2s, in order to determine some post-hotfix damage numbers. The resulting stats are shown here: http://i.imgur.com/bxpiaSi.jpg.

LRM 15 w/ TAG post-hotfix
(1207-206)/(1119-179) = 1.06 damage/missile

SRM 6 post-hotfix
(27516-25938)/(9325-8543) = 2.02 damage/missile

Streak SRM 2 post-hotfix
(4334-3884)/(1479-1274) = 2.20 damage/missile

Edited by Amaris the Usurper, 25 March 2013 - 04:48 PM.

[focuspark]

since your stats did not reset after the patch... how can you derive real value from this data?

[Amaris the Usurper]

focuspark, on 25 March 2013 - 02:58 PM, said:

since your stats did not reset after the patch... how can you derive real value from this data?

If I read this correctly, you are saying that we can't know what a patch has done to our stats unless they were reset after the patch. Correct?

But that isn't true. The stats represent cumulative totals of damage, e.g., number of shots, number of hits, etc. We can always tell what a patch has done by recording our stats, playing a few games, and then recording the new stats. For example,

[hits between Time 2 and Time 1] = [hits up to Time 2]-[hits up to Time 1],

and

[damage between Time 2 and Time 1] = [damage up to Time 2]-[damage up to Time 1].

The quantities on the right come from looking at our stats page at two different times. We can then find the average damage per hit according to

[average damage between Time 2 and Time 1] = [damage between Time 2 and Time 1]/[hits between Time 2 and Time 1].

As long as no patch or hotfix occurs between the two times in question, the numbers obtained are valid statistics for a single version of the game. If a patch or hotfix does occur, then they represent a weighted average between two (or among several) versions of the game, which would be less valuable.

With one exception, the numbers that I am reporting are of the former type, i.e., statistics for a single version of the game. The sole exception was the lower limit that I established for SRM damage post-patch but pre-hotfix, which involved data from several versions. But the reasoning leading to that number is also correct.

Edited by Amaris the Usurper, 25 March 2013 - 04:47 PM.

[Cyke]

Crap, I should have recorded my SRM stats shortly after the patch hit.
Now my pre-patch and post-patch stats are muddled together

Still, at least 85-90% of my SRMs fired are before the patch, so average of my total stats should be (more or less) pre-patch stats. Meanwhile, the new increment to each stat from my new games should reflect the new stats.

I'll post again later in the week after I've had a chance to make some heavy SRM use (I don't use LRMs).


Also, by the way, I read on another thread that the total damage shown in the stats page does not (and did not) record additional damage inflicted by splash; it only shows direct impact damage.
If this is true, it would unfortunately make this endeavor quite fruitless..

Edited by Cyke, 25 March 2013 - 05:08 PM.

[Amaris the Usurper]

Cyke, on 25 March 2013 - 05:07 PM, said:

Also, by the way, I read on another thread that the total damage shown in the stats page does not (and did not) record additional damage inflicted by splash; it only shows direct impact damage.
If this is true, it would unfortunately make this endeavor quite fruitless..

It appears that things are working correctly, or, at least, if there is a problem, it's not that splash damage isn't being recorded. If only direct damage (a constant value) were being recorded, we ought to see exactly the same pre-patch damage numbers for everyone and for all types of SRM. But we don't.

The counterargument to this is that, for example, I have 26,266 damage from 1279 hits with the AC/20, resulting in 20.54 damage per hit. Why is it higher than 20? We should expect to see less than or equal to 20, given that damage falls off with range. Are the stats for direct damage broken in some subtle way?

Currently, my best hypothesis is that it is also counting the HP (armor and structure) remaining in limbs blown off by torso hits. This would offset the reduced damage from shots taken beyond the nominal "range" of the weapon. Alternatively, it might be counting damage dealt to items (which have various HP) through critical hits. Possibly both.

Anyway, there's a strong argument that the above--if it is actually happening--represents the meaningful "damage" dealt under battlefield conditions, i.e., it shows how effective the weapon is in destroying the enemy, as it is actually used.

Also, splash damage is almost certainly being counted post-hotfix, even if it was not before. The reason for this is that SRMs are now supposed to deal 1.5 direct damage to whatever component they strike, in addition to splash, but our post-hotfix stats show significantly more than 1.5 damage per missile, even when no limbs have been blown off or items destroyed.

You can enlist the help of a friend to check that the 1.5 direct damage is, indeed, being dealt. For example, have him put 60 armor on the CT of his Cataphract. You will find that it takes exactly 40 SRMs to completely remove the CT armor, with no bleed-through to the structure (you will see splash damage dealt to adjacent components). The simplest explanation for the extra damage appearing in the stats is that it is the splash damage observed in-game.

Edited by Amaris the Usurper, 25 March 2013 - 05:44 PM.