97 replies to this topic

[ilikerice]

I could care less about stats.... BUT I would like to keep a positive k/d just because I don't want to be in the negative. This does not change or effect my playstyle in anyway, because for the most part my k/d fluctuates around 1:1

I play this game while I drink... a lot. So if I really cared about my stats I would have to go through all the hassle of being sober and not being able to enjoy my time off of work, or playing on an alternate drunk account. That all is too much effort for me, so my stats are my stats...

[Mercules]

None of the stats they use track measure team contribution accurately.

Win/Loss comes closest, but there are a bunch of circumstances not in anyone's control that can skew that as well. If your teammate suddenly disconnects you are down a mech. If they decided tonight was the night to get drunk and run around in an Urban mech with a flamer and MG as weapons you can't really make up for that.


Nothing measures a player who leads a charge and absorbs all the damage that allows a team to push into the enemy, in fact that person usually ends up with horrible stats for any match they do that as they tend to die with very little damage, kills, or assists.

Nothing measures a light who manages to get 1/4 of the enemy team chasing them so that the rest of their team can pound what is left.

Nothing measures the contribution a mech in a good location informing his team that the enemy is not coming from the direction they though so that the team can get set up better makes.

Edited by Mercules, 08 June 2015 - 10:11 AM.

[Jabilac]

EgoSlayer, on 08 June 2015 - 08:15 AM, said:

These patch notes list the rewards 2.0 payouts in XP and C-Bills, although I think there have been some updates mostly on the Conquest/capture side to increase them:
http://mwomercs.com/...tch-notes-13345

Great information but I was referring to the OPs original question about what stats matter to each player. I would like to be able to see how often I get rewarded for my role in combat besides whats already available in the stats page. Of the things tracked by the stats page xp per match and cbills per match tend to show me if im improving or not.

[WildeKarde]

Stats don't mean much to me playing, they don't tell me if I had fun in any of the games or had to grind the game out with no enjoyment.

I always find a new mech struggles for me so probably pulls my stats down and I don't play enough on mastered mechs to bring them back up I'd say.

Also I just need to look at HBK-4P for me where I've a 1.17 W/L and a K/D of 0.19. Means nothing on how I've performed as part of the team, if I maybe took a risk to help get a win.

[Mercules]

Jabilac, on 08 June 2015 - 08:44 AM, said:

Great information but I was referring to the OPs original question about what stats matter to each player. I would like to be able to see how often I get rewarded for my role in combat besides whats already available in the stats page. Of the things tracked by the stats page xp per match and cbills per match tend to show me if im improving or not.

No system is going to accurately measure that.

[Revis Volek]

Adiuvo, on 04 January 2015 - 10:18 AM, said:

KDR is more important here than in others due to the lack of respawns.

K/d can also be poached due to our sprites having lots of HP instead of death by a bullet or 2. Not many other games can you stand and watch your crew do dmg to mechs paper dolls and then wait to just walk up and get the final shot.

So IMO K/D is not indicative of much at all that i find good qualities in a team player. W/L is more important and mech and weapon stats as well they show me what you use best or use the most.

They are all just for us to get better on our own...

[ScarecrowES]

Mercules, on 08 June 2015 - 08:59 AM, said:

No system is going to accurately measure that.

Not true. The system MWO uses already tracks many stats that determine how a player is playing. You're rewarded money and Xp for almost all of it. But match scoring is vastly skewed to damage and kills. Try running conquest as a light and ONLY cap. Your team will beat the other easily, but you'll receive almost nothing for it.

What needs to happen is all these pieces of info MWO tracks need to be calculated to meaningful stats in the context of the match, and overall lifetime stats adjusted in relation to their value. Look at Battlefield as an example of persistent stat systems that mean something. Their stats can tell you instantly how skilled a player is, and what kinda teammates he is. Its just about making all those stats mean something.

[Bobzilla]

KD is a bad metric. How frequently do you leave that mech that's stripped of weapons to find an actual threat? It happens to me quite a bit, of course anyone else on my team usually swarms him or chase him for the 'KD' aspect. It's just a really bad metric.

[Roadkill]

Mcgral18, on 04 January 2015 - 08:29 PM, said:

So, a negative K/d is a bad thing, and higher generally implies competence, but can be inflated.

Add in that K/D isn't really a relevant measuring stick between players, and sure.

K/D is heavily influenced by play style, which makes it meaningless stat to use to compare players. A sniper will almost always have a higher K/D than a brawler, for example.

K/D is also heavily influenced by whether or not you play on an organized team. Organized teams do better in general than PUGs, which means you die less often, which means your K/D is higher.

Even a negative K/D (by which I assume you mean K/D < 1.0, and not actually negative since last I checked there's no "resurrect enemy" skill in the game) isn't necessarily a bad thing if you play scouts. Depending on the Mech you use for scouting and the way you do your scouting, scouts may not do much damage at all themselves but they enable the rest of the team to be much more efficient.

So in a very loose sense, yes, K/D can be a very general measure of player skill provided you have sufficient additional information for it to be relevant. On it's own it's pretty meaningless.

[Mercules]

ScarecrowES, on 08 June 2015 - 10:43 AM, said:

Not true. The system MWO uses already tracks many stats that determine how a player is playing.

I pointed out three examples of how it doesn't when someone is actively contributing to a success in the game.

It tracks "many" but most of those don't tell the story.

Point in fact. My unit won a group game the other night and we all contributed to the success. Good communication and movement and good pushing at the right moment. Most of the game I spent shifting around and keeping my team informed as to enemy movements so that they could shift their line to counter. Occasionally I would take a potshot with a medium laser. At one point I chased a Locust that was harassing one of our Assaults off with SRMs. I think my damage was right around 100 in that match, but I kept eyes on the enemy and contributed to our success.

The game didn't track that, you can't see that I was in the right place at the right time to spot and help counter and enemy flank or push. You can't tell from my stats that at the crucial moment I was the one who led the push in my Commando zig-zagging into the enemy and getting them to lay fire that missed me out then pushing so deep into them that some of them turned around.... right as my teammates came out of cover and cored their backs.

[Chuck Jager]

I find the win/loss a better stat. Even if match maker is borked, it should be equally borked for the mechs of that class. After 50 then 100 games certain mechs are just better than others in creating a win. The secret seems to be the alpha in 300-750m range. Closer range mechs may have great KDR, but the mech was not able to do much before the game was won or lost.

I also like to keep track of this stat if I have noticed a playstyle that needs to be improved. The biggest issue is it takes 25-35 games for the sample to be close to valid and that is still questionable.

[Rampancy]

Mercules, on 08 June 2015 - 11:33 AM, said:

I pointed out three examples of how it doesn't when someone is actively contributing to a success in the game.

It tracks "many" but most of those don't tell the story.

Point in fact. My unit won a group game the other night and we all contributed to the success. Good communication and movement and good pushing at the right moment. Most of the game I spent shifting around and keeping my team informed as to enemy movements so that they could shift their line to counter. Occasionally I would take a potshot with a medium laser. At one point I chased a Locust that was harassing one of our Assaults off with SRMs. I think my damage was right around 100 in that match, but I kept eyes on the enemy and contributed to our success.

The game didn't track that, you can't see that I was in the right place at the right time to spot and help counter and enemy flank or push. You can't tell from my stats that at the crucial moment I was the one who led the push in my Commando zig-zagging into the enemy and getting them to lay fire that missed me out then pushing so deep into them that some of them turned around.... right as my teammates came out of cover and cored their backs.

Nonsense. W/L tracks that. If you're consistently contributing in ways that don't show up on other aspects of the stat sheet, your W/L will reflect that.

[Lugh]

There are two stats that mean the most to me. Did I have fun? And did we win?

Winning increases fun. The thing that wins games is focus fire and damage dealt. So a high damage number and a high assist number means you are doing right by your team. (Stow the but lrms do high damage and they are worthless argument because you are wrong)

[Mercules]

RampancyTW, on 08 June 2015 - 11:54 AM, said:

Nonsense. W/L tracks that. If you're consistently contributing in ways that don't show up on other aspects of the stat sheet, your W/L will reflect that.

Not when you PuG. Because I can disrupt an entire enemy team but if my team sits behind a ridge and does nothing with it then we are down a mech(me) and everything else is still the same.

Lugh, on 08 June 2015 - 11:56 AM, said:

Winning increases fun.

No it doesn't. I'd rather play a good solid close game that we lost than a play a game that we won without effort.

[Jabilac]

We have a reward system in place that gives you cbills and exp for flanking, hit and run, protection, lance in formation, solo kill, most damage kill, savior kill, ect. The game already tracks those actions to reward the player after a match is over but none of those rewards are tracked in the stats section. Those are the types of things I would like to see. How often was I in formation or spotting. What is my average solo kill during play. How many times have I fallen into a protection formation to help cover a damaged friendly.

[Rampancy]

Mercules, on 08 June 2015 - 12:06 PM, said:

Not when you PuG. Because I can disrupt an entire enemy team but if my team sits behind a ridge and does nothing with it then we are down a mech(me) and everything else is still the same.

Then you aren't actually accomplishing anything, you're just throwing your mech away.

[Templar Dane]

RampancyTW, on 08 June 2015 - 01:08 PM, said:

Then you aren't actually accomplishing anything, you're just throwing your mech away.

I've experienced exactly what he's talking about. I've run into the enemy team to throw up a UAV, then got chased by half their team for 5 minutes.

Friendlies still haven't moved because an AC2 blackjack has them suppressed.

Or when your team is pinned by LRMs, and you track down the LRM boat and blow off his ears......then when you're making your getaway an enemy firestarter comes after you and after a running battle you lose.......and your team insults you for being stupid and getting killed by yourself.

There are two stages to a match. The poking stage, and the flanking stage. You can lose the match at either but its usually the team that doesn't flank that dies. Huddling up in one location is a sure way to get yourself surrounded and that's the fatal mistake a lot of pugs make.

[ScarecrowES]

Jabilac, on 08 June 2015 - 12:13 PM, said:

We have a reward system in place that gives you cbills and exp for flanking, hit and run, protection, lance in formation, solo kill, most damage kill, savior kill, ect. The game already tracks those actions to reward the player after a match is over but none of those rewards are tracked in the stats section. Those are the types of things I would like to see. How often was I in formation or spotting. What is my average solo kill during play. How many times have I fallen into a protection formation to help cover a damaged friendly.

Pretty much what he said. The game DOES track this stuff. If you watch your scoring carefully, or review it in depth at the end of the match, you'll find the game knows more about what you did than you think.

The game knows when you're targeting enemy mechs that your friendlies hit. The game knows when a UAV you popped is letting your team ram LRMs down the enemy's throat. If you're scouting, the game knows it. The scoring system just doesnt favor that as equally as doing straight up damage. But look it up. All the scouting indicators are there. The stats are there. The game just doesn't care.

[CptGier]

If I care about my WR its only because it affects my KDR...

How? Cuz when my side melts in 2 1/2 minutes, or im on the losing side, the game typically doesnt last long enough for me to get any kills....ending up with 0k, and 1 death. And, I find it annoying to be on the losing side that got stomped that badly.

[Mercules]

RampancyTW, on 08 June 2015 - 01:08 PM, said:

Then you aren't actually accomplishing anything, you're just throwing your mech away.

People who sit in cover and hide lose. I don't suicide rush but if I notice 3 mechs are separated from the enemy team and can bait them while guiding my team into them I'll do it, even at the cost of my mech.

In the group queues against semi-decent shots I can often make two passes through the enemy team in a Commando and still have a mech left. Literally run through the whole enemy team ducking and weaving. My Unit will take advantage of that in a heartbeat because most of the time I can get several to focus on me just long enough for my team to take one down. I can also break up their firing line enough that my team can break off and move to better positioning. In doing so I usually spray a couple mechs in passing nipping at their backs or heels, and generally tick them off, but don't destroy anything. I can also use that to secure a kill on a fleeing damaged mech.

Most PuGs however tend to focus on stick in cover and screw anyone else. In a PuG match where I often run into less focussed aim, I can often make pass after pass harassing the enemy or even run through and solo down an LRM mech. Nothing sucks more than distracting an LRM or Sniper mech and seeing that my team stayed in cover despite me saying in VOIP, "I have their LRM boat locked down you can move." and "Might want to push while their LRM mech and another mech are focused on me."

ScarecrowES, on 08 June 2015 - 01:30 PM, said:

If you're scouting, the game knows it.

No it doesn't. You get scouting points for being the first mech to target lock it. After that all the, "Enemy is flanking right through Gulf3 prepare for them to hit you in Echo3." earns you exactly zilch. I've frequently secured most of the "Scouting" bonuses in a match by being in an ECM Hellbringer with a Targeting Computer, CAP, and Sensor Range module. I just get high fairly quick and can cycle through the opponents quickly before the "Scouts" even enter their smaller sensor range. I don't even have to hide since at that distance lasers barely poke and if I am moving at all Gauss have a hard time hitting.

Edited by Mercules, 08 June 2015 - 02:28 PM.