72 replies to this topic

[Bishop Steiner]

Tarogato, on 30 December 2015 - 01:16 PM, said:

Where were you guys when I was collecting this stuff a few months ago? hehe.

About 76 of the players reported being mostly solo players and only 5 players reported that they play mostly group queue. The rest didn't specify at all. Perhaps we can assume that solo players outnumber group players 15-to-1 in this sample, but take it with a very very large grain of salt.

I was probably either in Tennessee with no reliable access to a computer, or running for my life from the biggest hurricane ever recorded, in Mexico.......

[EgoSlayer]

Bilbo, on 30 December 2015 - 02:00 PM, said:

I understand how it works. The only time I see unbalanced (by weight class) matches is when valves are released due to the introduction of new mechs.

Your experience is different than mine, because I almost never see 4x3. I see 3 lights in a match in maybe 1 out of 10.

[RussianWolf]

Bilbo, on 30 December 2015 - 12:33 PM, said:

Again, if this were the case I'd expect to see near instant matchmaking. I do not. My guess is the underlying PSR number is used in the same fashion as elo. The only difference being the tier 1 players never see the those in the bottom two.

I do see near instant matches most of the time.

[RussianWolf]

DarthRevis, on 30 December 2015 - 12:37 PM, said:

ELO is designed to put two players of equal skill against each other...as far as i know w/l may be factored but ELO itself is not trying to give you a 50/50 its giving you a 50/50 chance at winning or close to as possible for the best skill match against players.

PSR works much the same, just because we got PSR doesn't mean the MM got thrown out the window. We still have a MM, what you are suggesting is NO MM which is false. Thats how CW works yes but not the normal queues.

let me rephrase.... the ******* version of ELO that PGI was using......

[Mystere]

Danth Reduviid, on 30 December 2015 - 01:23 PM, said:

Cool data, but just eyeballing that line it looks off. It seems like it should be steeper.

Actually, if you'd notice, the highest concentration of points is below the line and to the left of the 1.3 mark. Those dragged the line down.

[Appogee]

Just a point to add on that w/l vs k/d correlation graph...

Way back when AC40 Jagers were a thing, I set out to "achieve" a k/d of 2.0. Which I duly did... simply by taking a high-alpha build for two months.

So the observed variances between w/l and k/d can be explained by the kinds of Mechs people prefer to pilot. People who take dual-Gauss Jagers - or for that matter any high alpha PPFLD Mech - will get more kills without necessarily winning any more matches, I believe.

[Screech]

DarthRevis, on 30 December 2015 - 01:11 PM, said:

The only artificially inflated stat i see are the w/l when running with groups. Its actually the opposite when it comes to DMG and KILLS when you run with a group of very competent mechwarriors. You have to work very hard most the time to get kills and dmg in before mechs die due to focus fire.

When i solo play, i have no issue getting high score and kills 800, 900, 1000 dmg and 1-4 kills is not unheard of. But in group queue im lucky to dish out 250-350 dmg and steal a kill because we are all on the same level and doing the same thing.

So again, the only thing i see that truly can get inflated artificially is your w/l, unless you had a team feeding you kills and dmg.

KDR can be improved since it kills per death and not kills per match. Winning more matches typically will mean dying less and that can make up for any lowering of kills. I could kill twice as much as someone who dies half as much and we would have the same KDR.

[AlphaToaster]

Appogee, on 30 December 2015 - 03:09 PM, said:

Just a point to add on that w/l vs k/d correlation graph...

Way back when AC40 Jagers were a thing, I set out to "achieve" a k/d of 2.0. Which I duly did... simply by taking a high-alpha build for two months.

So the observed variances between w/l and k/d can be explained by the kinds of Mechs people prefer to pilot. People who take dual-Gauss Jagers - or for that matter any high alpha PPFLD Mech - will get more kills without necessarily winning any more matches, I believe.

I think you're correct. The opposite can be said for a player who always leads a push regardless, tends to get shot up badly without a kill, but those pushes win the game. That player will have a terrible K:D since even on a win they're focused down first usually, but maybe a stronger W/L to reflect their tanking-first-in-the-door playstyle.

[Moomtazz]

Not sure if OP stated his overall W/L but working on XP for new mechs means they won't be as effective as usual and so a drop win ratio is going to happen.

Ex. my W/L is 2137/1644 or a 1.29 ratio. I don't think it's an unusual ratio from reading here. If I went on a losing streak of only 1 win per 5 matches, it would take 821 matches to drop to a 1.0 W/L ratio.

Edited by Moomtazz, 30 December 2015 - 08:16 PM.

[Revis Volek]

Bishop Steiner, on 30 December 2015 - 02:00 PM, said:

I believe W/L was the stat we were talking about?

I was more clearing the air, not really so much towards you good sir. Though i quoted you, i was gonna post it anyway because in my travels K/D wasn't very easily inflated with the groups i ran/run with. They are all as good or better then me so it hard sometime to be useful even lol.

Screech, on 30 December 2015 - 03:17 PM, said:

KDR can be improved since it kills per death and not kills per match. Winning more matches typically will mean dying less and that can make up for any lowering of kills. I could kill twice as much as someone who dies half as much and we would have the same KDR.

Good point, so i guess to an extent then yes K/D is also inflated, though maybe not so easily.

Thanks and i agree 100%.

RussianWolf, on 30 December 2015 - 03:04 PM, said:

let me rephrase.... the ******* version of ELO that PGI was using......

Do you have the exact equation/formula they were using? I'd surely like to see it! Maybe would shed some light on a few things for us lol.

Edited by DarthRevis, 30 December 2015 - 05:07 PM.

[9thDeathscream]

I think that damage output is more related to Win loss, then KDR. (Pugging not CW, but probably applies)

If your doing 200 damage a match and team is loosing, your not carrying hard enough.

If your doing 500+ damage and your going 50/50 that's about average.

If your doing 700+ damage and your teams are winning then your definitely performing where you should be.

Ok in this Tourny ive been plodding in my Dire Wolf B average about 650+ ish damage with a 2-4 kill. my PSR (based on WL ratio) bar has moved about half an inch (27" 1080p). Now considering I have about 5600 matches since the stats system was refreshed. I would say that my damage output has been a significant factor.

I may not be killing huge numbers but ripping the enemy's armour off is helping out the friendlies a lot!

In saying all of this my KDR hasn't changed. 1.97 but my win loss ratio has!

Edited by Akulla1980, 30 December 2015 - 07:13 PM.

[Hit the Deck]

Tarogato, on 30 December 2015 - 11:20 AM, said:

Your win/loss ratio should correlate roughly to your kill/death ratio. At least, according to my research. So if you're bad at the game, you'll lose more. If you're good at the game, you'll win more. If you're good at the game and losing more, then you are a statistical anomaly and can take comfort in the knowledge that it will come to pass and you'll be out of the rut eventually.

This tells us that they become more related the lower your KDR is.

[J0anna]

Appogee, on 30 December 2015 - 11:12 AM, said:

I've PUGed exclusively for the past week, and played about 60 matches.

My win/loss ratio, levelling my IIC Mechs, has been about 1 win from every 5 matches.

While I'm only a Tier 2 player, I do consistently finish in the top 3 damage and kills on my team. So it's not like I'm personally causing my team's to consistently fail.

I do seem to get lumped with a lance of quite inexperienced players every match. These players are unfamiliar with concepts such as sticking together to win etc. They frequently do 100 damage or less, and tend to spend most of the match taking cover instead of shooting enemies.

Is there anything in the current matchmaking algorithm that should lead to about a 50:50 win:loss ratio? Or is this what I should expect from MWO going forward ... mostly losses as I try to level Mechs and prove incapable of carrying inexperienced PUG players?

This is a genuine question. I'm wondering how I can be losing so many matches so consistently despite generally achieving personally reasonable scores?

I have never seen a quote from PGI that stated they want or hope players achieve a 50% win ratio. From everything I've read the match maker tries to balance teams and normally does pretty good at it.

However I have never seen a statement that the teams are perfectly balanced. So in all probability one team is probably slightly favored to win over the other. Additionally, PGI has never even hinted that if you lose say 10 in a row, that the match maker will try to place you on the 'favored' team. In all probability you've been getting a run of drops on the less favored team. Additionally, most of us probably perform worse in mechs before they're mastered, so you are probably playing a bit below your (normal) level.

I've found if you want to increase your winning percentage, it's best to drop in a group....

[Bilbo]

Screech, on 30 December 2015 - 03:17 PM, said:

KDR can be improved since it kills per death and not kills per match. Winning more matches typically will mean dying less and that can make up for any lowering of kills. I could kill twice as much as someone who dies half as much and we would have the same KDR.

I think you are right, but my drunk self just got a headache trying to follow your logic.

[NextGame]

Matchmaker likes to book players to job with a team of idiots who wont ever push under any circumstances in order to create the most frustrating experience possible.

Why have one part of the player base upset because they cant win a game when you can annoy the entire player base by forcing them to drop with terribad teams.

Edited by NextGame, 30 December 2015 - 11:42 PM.

[adamts01]

If you're consistently doing well and losing then you have to bring cheesy meta builds and carry harder. Also, top damage doesn't mean you're doing what's best for the win. Not saying this is you, but it's easy to sit back, pad your damage and kdr while your teammates are mixing it up on the front lines. I'm frequently at the top of my matches in fun mechs but my W/L is close to 1. My try-hard Timby has a 1.93 W/L after over 1,000 drops.

Edit: This was almost entirely in solo drops as well.

Edited by adamts01, 30 December 2015 - 11:46 PM.

[Anjian]

Appogee, on 30 December 2015 - 11:12 AM, said:

I've PUGed exclusively for the past week, and played about 60 matches.

My win/loss ratio, levelling my IIC Mechs, has been about 1 win from every 5 matches.

While I'm only a Tier 2 player, I do consistently finish in the top 3 damage and kills on my team. So it's not like I'm personally causing my team's to consistently fail.

I do seem to get lumped with a lance of quite inexperienced players every match. These players are unfamiliar with concepts such as sticking together to win etc. They frequently do 100 damage or less, and tend to spend most of the match taking cover instead of shooting enemies.

Is there anything in the current matchmaking algorithm that should lead to about a 50:50 win:loss ratio? Or is this what I should expect from MWO going forward ... mostly losses as I try to level Mechs and prove incapable of carrying inexperienced PUG players?

This is a genuine question. I'm wondering how I can be losing so many matches so consistently despite generally achieving personally reasonable scores?

The answer should be no. Not without violating or licensing someone else' patents.

Wargaming.net, the developers for World of Tanks, -Warships, -Warplanes, owns the patent for the matchmaker algorithm that equalizes player win/loss ratios.

[Anjian]

DarthRevis, on 30 December 2015 - 12:34 PM, said:

If you are bad, you will have a 50/50 ratio

If you are decent you will have a 52/48 ratio

Good players will start getting into 55/45 and 60/40 ratios....



I dont consider myself "good" merely decent and i have about 600 more wins then losses and its creeping up all the time.


Currently have 600 more wins then losses out of about 6,000 matches in total.

In World of Tanks, average or normalizing means you get at least a 44% ratio. If you are getting a 50% ratio, you are good. Getting 52%, you are very good.

[El Bandito]

Appogee, on 30 December 2015 - 11:12 AM, said:

Is there anything in the current matchmaking algorithm that should lead to about a 50:50 win:loss ratio? Or is this what I should expect from MWO going forward ... mostly losses as I try to level Mechs and prove incapable of carrying inexperienced PUG players?

This is a genuine question. I'm wondering how I can be losing so many matches so consistently despite generally achieving personally reasonable scores?

I currently have 1.3 WLR and it is still going up in Solo-Q. So carry harder, I guess. Also, use GXP to level the most important of the basic trees on the new mechs. Especially heat and accel/decel related ones. Especially since we got tons of GXP from the New Year event.

Edited by El Bandito, 30 December 2015 - 11:54 PM.

[Anjian]

Tarogato, on 30 December 2015 - 11:20 AM, said:

Your win/loss ratio should correlate roughly to your kill/death ratio. At least, according to my research. So if you're bad at the game, you'll lose more. If you're good at the game, you'll win more. If you're good at the game and losing more, then you are a statistical anomaly and can take comfort in the knowledge that it will come to pass and you'll be out of the rut eventually.

Perhaps in this game. My experience in War Thunder and World of Warships shows no correlation at all. As it turns out, damage output is the one that is more important. This is also my experience with MWO:CW as well.