
Now consider the following match summary, which was the last match immediately prior to the above. It was a loss. For that matter, I was the last on the team, left against two enemies, and only scored that one kill against one of those two in a failed attempt. 1K, 8A, Match Score 408. The difference? Besides having one kill fewer and losing, I mean? Oh yeah, the image was crap and left out the DAMAGE. 600+.

THIS is what's wrong with match scoring right now. I can drop two enemies in a win, and score lower than four other players who were killed out and killed NO ONE ELSE AT ALL (barring an accidental TK, though I don't recall that happening). But have me pull a lot of damage in a LOSS, and I can score 2.5 times as high with fewer kills.
COME ON!
Really? REALLY?!?!?! Look, I'm not bashing the other players in either of these two matches, nor am I trying to say that I do any better than MEDIOCRE on a good day. IF THAT. That's not the point at all. That 600+ damage that I contributed on the loss? It didn't save the match. Could have been 6,000, but it was still a LOSS. I only managed to kill ONE enemy mech (had it been TWO, it would have been a win, but there's no sense crying over that now). HOW IN THE FREAKIN' UNIVERSE can that be 2.5 times as good a performance as the other, in which I killed two enemies, survived, and contributed to a WIN? HOW?!
(And no, I didn't need Caustic Valley for the challenge, and yes, the Forest Colony score was one that I DID need.)
Simply rewarding DAMAGE at such a rate, that one can achieve a better match score by spamming high-DPS weapons without regard to their effect on target, doesn't recognize anything at all about player performance and contribution to the outcome of the match.
I don't pretend to know what would be a better match scoring system. That whole, "If you're going to criticize then you need to offer an alternative" thing is B***SH*T. Heard that for 22 years in the Army, and it was never true one single day of that career. I don't know because it's not my job, as the CUSTOMER/CONSUMER, to know. But I certainly have a role to play in pointing out that it is junk as is.
But don't listen to me. Check the stats on the Territorial Domination Challenge. Most common mechs selected? Gonna be the Dire Wolf, Timber Wolf, and Hellbringer. The first two are well known to be Tier One mechs in this game, and the third is not far behind and carries ECM (to keep the Lurmpocalypse at bay, because LRMs buff DMG numbers, which buffs match score, which brings the rewards in the challenge, without being all that effective for getting actual kills because they spread damage out). I'd bet a kidney that DWF and TBR are among the top three mechs chosen this weekend, and HBR is no worse than fifth place on that list. Probably 1-2-3 right there. And that's for one reason.
It's not that they're necessarily any more entertaining to play, and they're certainly not all that relatively challenging (you see that I handicapped myself with a SHC). It's that they bring DMG, DMG brings Match Score, and Match Score is what the challenge is based upon.
Some suggestions for match score balancing, if I may...
- Weigh the score for damage done by the type of mech doing the damage. Lights should get a little better score for hitting 250, 500, 750 DMG, because they generally carry fewer and less powerful weapons, so achieving the same damage as the guy in the KGC is relatively impressive when you're in a RVN. This should reflect in Match Scores, and if it somehow already does, then it should reflect even more.
- Weigh the relative value of the mech killed in scoring a kill. That is, if I kill a stock CDA, it's not so big a deal as killing a very custom DWF. Sure, the Cicada moves faster and is smaller, but the Whale has a LOT more armor and structure, and has a LOT more weapons with which to return fire. Killing the latter is usually considerably more impressive than killing the former. With this talk of rebalancing perhaps taking some BV-like mech value into account, it shouldn't be so challenging to consider THAT value when deciding what a K is worth.
- Perhaps weigh the relative value of the mech from which the kill is made, like above. If I kill your Marauder (THANK YOU SO MUCH FOR THAT, PGI!) with a LCT and had the most-damage-on-kill bonus, that should weight more heavily on my Match Score than if I did it from an AS7 with LRMs from a distance, and only just finished it off. Seriously, I had a match once where I did exactly 4 points of damage and had a kill. Last shot fired in the match, didn't even get the full burn of a single ML on the target before it fell, and that was the ONE shot I hit that match. Got the same credit for the kill as if I had worked that mech over for a whole minute. But that's just not right.
- Some extra bonus for winning by kill when you're the last mech on your team, or when it's a come-from-behind win. Say, as I saw happen yesterday I think, a guy in a RVN-3L(C) finds himself alone, outnumbered 5-to-1, and manages to finish off the match. Sure, he already had 7 kills and some ridiculously high damage output, to buff his score. Still has to be worth more that his kill ended the match in a win, and that he accomplished the turnaround win. SOLO. Dude's match score damned well should have been a four-digit number, but it wasn't. (Probably WOULD have been, had he done more damage.)
- Different DMG scores for damage done to mechs that were eventually killed, and those that were NOT. Say we do .25 Match Score for every point of DMG done to enemy mechs that survive, but .40 for every point of DMG done to enemy mechs that are eventually killed. Add 0.10 for every point, across-the-board, if it's a win, and another 0.05 if the player survives (win or lose). Whatever. Something like that. You get the idea.
The current match scoring system is pushing players to play heavier, deadlier mechs. That's fine if all you want to see are heavy and assault mechs in every match, or only new players out there grinding up lights to earn money for heavier mechs. But I don't think this is that sort of game, and I don't believe that's what PGI intends.
Who knows, maybe I'm preaching to the choir, and PGI already suspected as much. It would explain the challenge event's parameters. "Let's put a match-score-based challenge out there, and see what everyone chooses to use for a mech and loadout. You know, see how our match scoring system is working." Would not surprise me.
If so, then you have your answer. Current match scoring system is TOO biased toward damage, with little emphasis on things that matter (like WINNING, there, Charlie Sheen)...