Jump to content

Why Elo Doesn't Work Here


633 replies to this topic

#201 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 23 January 2014 - 04:22 PM

View PostRoland, on 23 January 2014 - 04:14 PM, said:

Having played with you, I would estimate that you're better than the vast majority of players in MWO currently. For you, your rating is going to rise up significantly faster and stabalize at the "better than most people" level quickly.. because really, for outliers on the edge of the Elo rank, it's much EASIER for Elo to work..because you are exerting a stronger force on it.

That is, in many games you participate in, you are going to have a much larger influence over the outcome than most players do, thus the actual result of the match actually corresponds more directly to your skill than it does to other players.


In all of those applications, each team is viewed as a single virtual player. The TEAM is rated, rather than any individual players.

That's how Elo works in those cases, because they aren't trying to determine individual ratings for players based on the outcomes of team games.


No, they determine each players Elo ranking from the team performance and victory. Sports teams also have variable players, 2nd and 3rd string but that's a separate factor.

Roland, you posted the link to the MIT paper on TrueSkill. It draws a players individual Elo ranking out of team performance and in fact, by drilling down on the telemetry of relative team compositions and the like is even able to predict rather well how well you'll play as a member of another team in a different sort of game, even of a different team size. TrueSkill is even able (when it's got sufficient data on a player to hit convergence) to predict how, for example, you'll perform in a FFA CoD game by taking into account how you played in team deathmatch games in LoL sorts of environments.

It's a thing of beauty but again, supreme overkill for what MW:O needs.

I'm all for adding both depth and breadth to the Elo calculation for MW:O. It's a lot of work though and without a thick enough player population won't actually change matchmaking results - the MM can't decide to place you with an LRM heavy group when playing against a sniper heavy group because you're in your 3L with tag because it's just happy you're in roughly the right weight range and roughly the right general skill range to save the match from being utterly lop-sided.

#202 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,629 posts
  • LocationChicago

Posted 23 January 2014 - 04:22 PM

View PostGhogiel, on 23 January 2014 - 03:48 PM, said:

Paul in his infinite wisdom addressed a nonsense point in his recent command chair. He said there were a few players who sent him screens of games that were stomps and he worked it out that the averaged team Elo rating between the teams is quite closely matched actually.


That's a ******* frighting statement from a dev...

Wait what.....how could he not know that's how Elo was working?? HE made a post over a YEAR ago explaining how a team's Elo was averaged out and how high and low players could be on the same team

http://mwomercs.com/...79-matchmaking/

#203 Boris The Spider

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 447 posts

Posted 23 January 2014 - 04:24 PM

The Elo system they are using on this game is totally unfair to certain players. I'm sure we have all seen it, you know when you have lead a charge into the enemy and went down in a blaze of glory. Then you start spectating the team, seeing if they have enough momentum to carry on through and break the enemy ranks. Then you see it it, an enemy with no weapons! You think your team haven’t noticed it, you go to type “look out team!” but before you do you see that one selfless hero has peeled off the group just in the nick of time to deal with this imminent threat to his comrades. Like a determined blood hound he chases the mech down, it doesn’t matter if it takes him 4 minutes to finally corner and kill this mech, his resolve is strong.

Surely, even if the rest of his team dies, and the round ends 11-12 he should at least be rewarded for his valiant deeds, no? He got over 600 damage, should his Elo not go up? And as for the rest of those noobs, especially the clearly overconfident charge leader and his supporters, well they should take an Elo hit even if the team wins!

We need an Elo system that rewards being able to get the high scores, not one that rewards winning stupid battles.

#204 Roadkill

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,610 posts

Posted 23 January 2014 - 04:28 PM

View PostSug, on 23 January 2014 - 03:14 PM, said:

Why do you think that?

Because this next part isn't true.

Sug said:

If your Elo can only change significantly if you beat a higher Elo team or lose to a lower one

I don't remember who made the post, but whoever it was at PGI that said this was wrong. In an Elo system, the only way that your rating doesn't change much is if your Elo is dramatically higher or lower than your opponent's and the expected outcome happens.

If I'm rated 2000 and you're rated 1000 and I win, not much changes. If you win, our ratings will change by the K value (50 in the case of MWO).

If we're both rated 1500, then whoever wins is going to earn half the K value and whoever loses will lose half the K value.

So even if you always play people with exactly the same rating, if you keep winning your Elo will keep going up until you start losing.

#205 RichAC

    Member

  • PipPipPipPipPipPipPip
  • 661 posts

Posted 23 January 2014 - 04:28 PM

View PostMischiefSC, on 23 January 2014 - 04:15 PM, said:


No, the problem is that 'common sense' in the context you're using it is actually 'in my uneducated opinion'.

The problem is that someones opinion of using win/loss has absolutely nothing to do, what so ever, with how effective it is. I entreat you to read the links I posted on statistics, statistical analysis and the actual principles of predictive analysis so you understand why your opinion is wrong.

Which it is. Saying Elo doesn't work for MW:O, or any other win/loss competition, is flat out wrong. Team or individual it doesn't matter. You could have 100 teammates that change every single game it it still works - it would just take more games.

Let me put it to you this way.

Suppose you took LeBron James and put him randomly in to replace one player each game in high-school basketball games. He would play each game with a different team and against a different team.

At the end of the season the teams he played with would be far, far more likely to have won because he was there.

Do the same thing with Payton Manning in high-school football games. Every team he played with would be far more likely to win, right?

You don't have to be world-shaking good to impact your teams performance, that just makes it easier to identify. Those guys could trend their winning in 20 games. For comparison sake they'd have a win/loss in that situation of like 5.0 or 6.0 instead of a 1.2 or the like.

Elo works. You're welcome to research the math behind why it works but trying to say it doesn't work is literally like saying math doesn't work because you're writing 1+1+1+1+1+1+1+1 = 8 instead of 4+4=8. The value of each factor (+1 eight times vs +4 twice) may be reduced but the result is the same.


So what your saying once again, it boils down to "don't worry eventually it will even out". Well I'm sorry but thats not good enough. Apparenlty most people quitting this game dont' feel like its evening out buddy. And its only going to get worse they less people are playing.

One problem is in this random team game called MWO, noone is even playing to win. They are playing for personal stats.

Similar to how people complain about pro NBA players just play for personal stats and prefer to watch college ball instead. Because those personal stats are what determines how much Lebron is getting paid. Just like it does in MWO!!

Call it uneducated if you want...But The only math that matters to people, is the math on the scoreboard at the end of a match!

Edited by RichAC, 23 January 2014 - 04:34 PM.


#206 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 23 January 2014 - 04:30 PM

View PostRichAC, on 23 January 2014 - 04:12 PM, said:


LoL apparenlty just switched to trueskill, becaues they realized an ELO system is not designed for a random team game. But Ironically trueskill is still ELO based.

Glicko, is not a game, its another ELO/ chess like based rating system. http://en.wikipedia....o_rating_system

LoL will eventually come the same conclusion, that trueskill is no different then any other ELO system. What LoL has going for it, is that it has millions of players, so there is less of a skill gap felt. And there is no need for them to try and create another industry standard.

But you still have threads like these http://forums.na.lea...d.php?t=3817616 that beg the same question.

How do scouts rank professional athletes? Do you think they do it based only on their team wins and losses? Absolute utter nonsense! But I expect nothing less from a nerdy industry that doesnt' even sell sports games anymore.

Its sad that bad decisions in quakelive killed that game, because their rating system was pioneering a whole new world once again in online mutliplayers. Just like they have in the past....But more then ID, we have to blame the gaming community that exists nowadays.

its common sense as to why you can't rate and individual based on their team performance and needs no explanation.


So... and I want to get clear on this...

All the statisticians at MIT, all the literally worlds greatest experts on predictive statistical analysis who've helped create these systems and use Elo as the basis of every serious matchmaking calculator in existence....

they're all just wrong? Because you just 'feel' that way?

Please, stop for a moment. Seriously, stop and think about what you're saying. Drill down on the basis from which you are making that assumption because it's a big assumption you're making here.

This is exactly why 'common sense' isn't that common. What you're talking about here isn't 'common sense', it's you attempting to assert that your opinion is more valid and reliable than mathematical formula and scientific analysis.

Not trying to be mean here or insulting but.... really. That's exactly what you're saying. I'm not going to be able to sum up years of courses in mathematics or statistics or condense the published works on statistical analysis that support just exactly how and why Elo works for you, not in a small enough bite to be useful here.

The gist of what you're saying though just blows me away though. It's like saying you've seen birds flying so clearly gravity doesn't really work on birds unless they want it to. I'm not really sure where to even begin fixing everything that's wrong in that.

#207 Roland

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 8,260 posts

Posted 23 January 2014 - 04:32 PM

View PostMischiefSC, on 23 January 2014 - 04:22 PM, said:


No, they determine each players Elo ranking from the team performance and victory. Sports teams also have variable players, 2nd and 3rd string but that's a separate factor.

No they don't.

Those examples you cited don't try to derive rankings for individual players at all from team games. They are only rating the teams themselves.


Quote

Roland, you posted the link to the MIT paper on TrueSkill. It draws a players individual Elo ranking out of team performance and in fact, by drilling down on the telemetry of relative team compositions and the like is even able to predict rather well how well you'll play as a member of another team in a different sort of game, even of a different team size. TrueSkill is even able (when it's got sufficient data on a player to hit convergence) to predict how, for example, you'll perform in a FFA CoD game by taking into account how you played in team deathmatch games in LoL sorts of environments.

Yes, Trueskill can determine an individual player's rating based on team games much better than Elo rating can.

That's why they made Trueskill and use it for multiplayer games on XBL.

#208 Roadkill

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,610 posts

Posted 23 January 2014 - 04:35 PM

View PostAbivard, on 23 January 2014 - 03:55 PM, said:

Elo Was created as a means to rate chess players.

It works fine for that, it works fine for any one on one sport, or even team sports where the teams are consistent. Once you get outside of what it was designed for, it stops working.

No, it doesn't. It works just fine, the caveat being that it takes longer for ratings to stabilize.

That's all. It's just as effective and accurate in the long run.

Could we do better? Maybe, but as MischiefSC pointed out any system that depends on other attributes such as damage or kills or spotting assists or whatever can be gamed. You might be able to create a "better" system under ideal circumstances, but once you add gamers to the equation that goes right out the window.

A big part of Elo's appeal is its simplicity. It just works. What you do with the ratings, though, might not work.

#209 100mile

    Member

  • PipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 1,235 posts
  • LocationAlegro: Ramora Province fighting Pirates. and the occasional Drac

Posted 23 January 2014 - 04:36 PM

View PostMustrumRidcully, on 23 January 2014 - 08:28 AM, said:

??? Where did you get that information?


http://mwomercs.com/...79-matchmaking/ <<<<<Go to this link and read, don't skim it...actually read it...

View PostRoland, on 23 January 2014 - 08:36 AM, said:

No, because of the various factors which will obfuscate your rating, it will actually take far MORE than 1200 games. The multiplying factor isn't simply the number of players. In terms of what the actual factor would be, I actually have no idea. Some of it could be figured out through statistical analysis, but some of it starts to play into the feedback resulting from mech configurations since this isn't a game like chess where all players are "playing with the same pieces". Then you throw things like map selection and synergy between maps and mechs into it, and it becomes further obfuscated.

Mischief was operating with the belief that you could simply multiply it by 12 and it'd be the same, but that's not accurate. The number of games required would be the number required by a game like chess, multiplied by some (large) number that's much more than 12.

It's true that it could, potentially, arrive at a correct seating, but I'm not sure at what point that would happen.


No, you're incorrect.
Your Elo rating is based purely on whether you win or lose, and the amount your rating moves is based on who you played. That is all that goes into it. It does not account for ANY of your actual performance metrics in the game.

This is actually one of the problems... because it really should account for your match score.

Match score, as it stands now, is actually a good metric for player skill. The best players generally always have the highest match scores in any given game. Or at least, relative to their own team members.


No your wrong... please see link above and read the whole thing carefully so you actually have some research to back up your arguments.....Notice this was the first post on ELO and there has been other stuff added to how it works since... i excerpted the pertinent info but please read the info above it as well....

"We now use the probability value of 0.41 to determine the change in player ratings.


Case 1: 1350 Player WINS over 1410 Player

Since the lower rated player won despite the odds against him/her, they are rewarded a much higher change in score than the higher player is. The higher player score will actually be reduced.

One variable that is set by winning or losing is the WinFlag (as seen in Figure 3). The WinFlag is a binary (true/false) value of either 1 (Player has won) or 0 (Player has lost). In this case, the WinFlag value is 1.

Let’s begin the calculation:

Old Rating = 1350
Maximum Change Allowed = +50 for a win, -50 for a loss (as seen in Figure 1)
WinFlag = 1
Probability of Winning = 0.41

1350 Players new ranking = 1350 + 50 x (1 – 0.41)
= 1380

1410 Players new ranking = 1410 – 50 x (1 – 0.41)
= 1381

A player’s rating will only go down if they are beaten by a player who has a lower rating than theirs. In this case, if the 1350 player lost, their score would not change since the Match Maker was correct in its prediction.

So how does this affect Match Making – Phase 3?

This formula and scoring system is run against match data that the current dedicated servers are giving us. Basically it’s being tested on real world numbers generated by you, the community.

Once we get a full understanding of how accurately the Match Maker is working, we are going to add some additional parameters to the mix. These include a more defined player skill rating and a Mech weight class balancing system. More info on these when the first pass of Elo testing is done."

Keep in mind that the above describes what happens in a pug situation...
In a team situation (12 Man) then the Team ELO is factored and averaged and the score is applied to the individuals depending on which team wins and which team loses......

#210 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 23 January 2014 - 04:37 PM

View PostRoland, on 23 January 2014 - 04:32 PM, said:

No they don't.

Those examples you cited don't try to derive rankings for individual players at all from team games. They are only rating the teams themselves.



Yes, Trueskill can determine an individual player's rating based on team games much better than Elo rating can.

That's why they made Trueskill and use it for multiplayer games on XBL.


All the esports that use Elo, as far as they've listed, derive individual players ranks from team performance using Elo as the basis of that ranking. The only ones I listed that don't are ones that don't rank individuals, only the teams.

I'm going to repeat this about TrueSkill -

TrueSkill uses your win/loss ranking as the basis for your ranking. Just like Elo does. It even uses the Elo calculation as its foundation. All it does is then look at the players you played with and against, how you played and what you played, to then predict how you'll perform in other systems and other teams.

TrueSkill could not significantly improve matchmaking in MW:O - not more than switching to a Gaussian distribution, splitting pug/premade Elo and matching to range not high/low to target would do. It can not create more people to match you with to give it a wider range of matchmaking options.

#211 RichAC

    Member

  • PipPipPipPipPipPipPip
  • 661 posts

Posted 23 January 2014 - 04:39 PM

View PostMischiefSC, on 23 January 2014 - 04:30 PM, said:


So... and I want to get clear on this...

All the statisticians at MIT, all the literally worlds greatest experts on predictive statistical analysis who've helped create these systems and use Elo as the basis of every serious matchmaking calculator in existence....

they're all just wrong? Because you just 'feel' that way?

Please, stop for a moment. Seriously, stop and think about what you're saying. Drill down on the basis from which you are making that assumption because it's a big assumption you're making here.

This is exactly why 'common sense' isn't that common. What you're talking about here isn't 'common sense', it's you attempting to assert that your opinion is more valid and reliable than mathematical formula and scientific analysis.

Not trying to be mean here or insulting but.... really. That's exactly what you're saying. I'm not going to be able to sum up years of courses in mathematics or statistics or condense the published works on statistical analysis that support just exactly how and why Elo works for you, not in a small enough bite to be useful here.

The gist of what you're saying though just blows me away though. It's like saying you've seen birds flying so clearly gravity doesn't really work on birds unless they want it to. I'm not really sure where to even begin fixing everything that's wrong in that.


Are those MIT guys professional scouts or managing professional athletes or professional sport teams? hahah I guarantee they would all laugh about this.

You should watch that movie moneyball with Brad Pitt, and see how a real MIT guy changed the face of baseball as we know it today. Someoen should give him a call lol.

I still think the rating system made by gaimtheory for quakelive, is the best ever made to date. And would of broke new ground if the game didn't die off for other reasons. Trueskill might better then a basic ELO rating system, supposedly, but its still an ELO rating system to me. and would only be good for rating a team not the individual.

I consider online multiplayers like MWO, sports. And treat them like such.

Stats are what mainly determine a players worth to a team. And stats should be what determines his "skill" in MWO. Stats determine how much a pro athlete gets paid, and it also determines how many cbills an MWO player gets paid.

not just one stat like W/L, or K/D or damage or assists, Unless your trying to fill a role/gap on a permanent team, but in reality all of the above. If you have a problem with that fix the point system. Which is the root problem of many issues with the game.
\
How long are we gonna wait for it to even out? For the game to die first? That statement is something us uneducated people can trust? You know us guys thats actually play sports? When we can see and understand the scores that we are playing for with our own eyes?

Again, the only math that matters to people, is the math on the scoreboard at the end of a match.

Edited by RichAC, 23 January 2014 - 04:52 PM.


#212 Ghogiel

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • CS 2021 Gold Champ
  • CS 2021 Gold Champ
  • 6,852 posts

Posted 23 January 2014 - 04:44 PM

View PostSug, on 23 January 2014 - 04:22 PM, said:


That's a ******* frighting statement from a dev...

Wait what.....how could he not know that's how Elo was working?? HE made a post over a YEAR ago explaining how a team's Elo was averaged out and how high and low players could be on the same team

http://mwomercs.com/...79-matchmaking/
Yes we all knew it would place players closest to target elo since the first days. We didn't know the actual thresholds until this week.

And no one questioned that is how it was working. He's just playing stupid if you thinks having high rated elo players and low rated elo players on a team isn't what we wanted to know about.

As in not the difference in the teams averaged elo, but the average individual elo divergence of the players on the teams.

#213 Roadkill

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,610 posts

Posted 23 January 2014 - 04:51 PM

View PostDoctor Proctor, on 23 January 2014 - 03:33 PM, said:

Worse, match score is heavily weighted on damage. I've had more kills and assists than players and received a lower match score due to have lower damage.

Yeah, no kidding. In my best game ever in a Jager (2 AC/10, 4 ML), I had 7 kills, 5 assists, and 700-ish damage. But I finished 3rd in match score.

It wasn't planned, but I ended up working with a Stalker missile boat. Everything I'd start chasing, he'd unload on (in no small part because I actually hit 'r' just in case). I'd get in a couple of shots, the missiles would start to rain, and a few seconds later I'd get the kill. Oh yeah, I was also deliberately legging Mechs once I figured out that the missile boat was working with me. Might as well help his missiles hit, right?

The missile boat ended up with 2 kills and 10 assists. And something ridiculous like 1400 damage.

The other guy who ended up with a higher match score had the remaining 3 kills, something like 6 assists, and something like 800-ish damage.

The missile boat I can figure out - he did double my damage so that's how he got a higher match score. But the other guy I just don't get. I had twice as many kills, more total enemies engaged (kills + assists), and nearly as much damage yet he somehow got a match score 20 points higher than mine. (IIRC mine was 90-ish, the missile boat was 120-ish, and the other guy was 110-ish.)

So... no. I don't think match score is the solution. It might be part of the solution, but it's not enough by itself.

#214 Roland

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 8,260 posts

Posted 23 January 2014 - 04:51 PM

View PostMischiefSC, on 23 January 2014 - 04:37 PM, said:


TrueSkill uses your win/loss ranking as the basis for your ranking. Just like Elo does. It even uses the Elo calculation as its foundation. All it does is then look at the players you played with and against, how you played and what you played, to then predict how you'll perform in other systems and other teams.

Dude, just because it uses wins and losses as a basis doesn't mean its the same as Elo.
The point of the system is that they IMPROVED Elo (or rather, Glicko, which was itself an improvement of Elo) to be better able to determine player skill from team results.

View PostMischiefSC, on 23 January 2014 - 04:37 PM, said:

TrueSkill could not significantly improve matchmaking in MW:O - not more than switching to a Gaussian distribution, splitting pug/premade Elo and matching to range not high/low to target would do. It can not create more people to match you with to give it a wider range of matchmaking options.

I would agree that in the given matchmaking situation, Trueskill would not somehow magically fix things. The matchmaker has other fundamental problems with it.

However, at the same time, simple Elo OR Trueskill may be insufficient for this game anyway.

Both Elo and Trueskill are essentially just generalized rating algorithms... For MWO, since you aren't trying to generalize across all possible games, you could come up with a BETTER rating system which actually took into account specific aspects of the game itself.

For instance, as others have pointed out... The match score at the end of the game. THAT could be leveraged in a rating system... because the folks topping the scoreboard are generally better players.

There's really no reason to limit ourselves to only using simple win/loss for rating.

#215 Rofl

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 435 posts
  • LocationTrash can around the corner.

Posted 23 January 2014 - 04:55 PM

View PostMischiefSC, on 23 January 2014 - 04:15 PM, said:

Which it is. Saying Elo doesn't work for MW:O, or any other win/loss competition, is flat out wrong. Team or individual it doesn't matter. You could have 100 teammates that change every single game it it still works - it would just take more games.



So what you're saying is the more random elements in the ELO system (represented by 11 other team mates) turns into a longer time in games to calculate your true ELO, and during that time your ELO will most likely be accurate (albeit getting ever so slightly more accurate over time).

My only complaint would be, how many games would it take with 11 random elements to figure out how good a player really is? Some people in LoL say about 200 games will average you out. However, in 200 games your skill has most likely increased, so we could say your LAST 200 games is your average, and is expected to rise unless you are not trying or run into a steak of bad match ups.

With LoL you have 4 team mates, compared to 11. How long is it taking new players to get into their true ELO? How about experienced players?

400? 800? Ten million games? How long is it until the system has averaged enough games to say with any amount of reliability: This is you.

#216 Roland

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 8,260 posts

Posted 23 January 2014 - 04:55 PM

View Post100mile, on 23 January 2014 - 04:36 PM, said:

No your wrong... please see link above and read the whole thing carefully so you actually have some research to back up your arguments.....Notice this was the first post on ELO and there has been other stuff added to how it works since... i excerpted the pertinent info but please read the info above it as well....

100Mile, what part of that do you think supports your belief that the matchmaker is taking into account anything other than win/loss in determining rating?

You cut out an excerpt which points out that your change is rating is affected by the rating of your opponent, which is exactly what I said... but it doesn't take into account anything other than whether you won or lost. It doesn't take into account your score, or kills, or assists, or anything.

What part of what you posted do you think says otherwise?

#217 White Bear 84

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 3,857 posts

Posted 23 January 2014 - 04:58 PM

View PostArtgathan, on 22 January 2014 - 02:07 PM, said:

You could make the argument that your performance was not "flawless" if your team ended up losing.


Yeah but when you get 600 damage, the only kills and the rest of the team have a score that adds up to less than your own score, pretty sure it is safe to say you played 'flawlessly' in comparison to the rest of your team. Arguably, one mech could perform amazingly, but it takes more than one mech to hold up a team.

#218 RichAC

    Member

  • PipPipPipPipPipPipPip
  • 661 posts

Posted 23 January 2014 - 05:02 PM

Can you just imagine if a general manager of a pro athletic team, or pro athletic scouts, used an ELO system.

ELO is for chess or 1v1.

http://en.wikipedia.org/wiki/Moneyball

http://en.wikipedia....iki/Sabermetric

Sabermetrics is what the MIT guy at the Oakland A's came up with. Its the standard used today to rate players.

But MWO doesn't have to be complicated. Basicaly sabermetrics still uses indiviudal players stats for VORP for a permanent team. Since even in baseball your don't drop with random players on a team each match.

But rating on stats, is still basically the same thing. For example when it comes to lebron james. His stats are what determines his worth or how much hes getting paid. Just like stats already determine how much cbills an MWO player is getting paid. People want to be matched up accordingly to their match scores. That is what they are playing for.

If there is a problem with the point system. Change the point system. There have been many threads about how the point system takes away incentives to win. Takes away incentives to use other weight classes. Give more points for winning a game, give more points for capping....etc... Tweak the point system a little and match people up accordingly. After all we have skirmish now.

Otherwise the complaints will never end, and people will always feel mismatched.

Edited by RichAC, 23 January 2014 - 05:06 PM.


#219 Sug

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 4,629 posts
  • LocationChicago

Posted 23 January 2014 - 05:02 PM

View PostGhogiel, on 23 January 2014 - 04:44 PM, said:

As in not the difference in the teams averaged elo, but the average individual elo divergence of the players on the teams.


Team A Player Elo Scores: 2000, 2000, 2000, 2000, 1900, 1800, 1800, 1800, 500, 500, 500, 500

Team B Player Elo Scores: 1700, 1700, 1600, 1600, 1500, 1500, 1400, 1400, 1400, 1300, 1200, 1200

Average Elo of Team A: 1460

Average Elo of Team B: 1460

The matchmaker sees these teams as equal so no one's Elo score is going to move very far if at all no matter what the outcome. No big deal. Elo working as intended.

Would this be a good, fun game to be in? Probably not for one of the teams. Which is what people ***** about, which is a problem with how the MM uses our Elo scores and not Elo itself.

#220 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 23 January 2014 - 05:03 PM

View PostRichAC, on 23 January 2014 - 04:39 PM, said:


Are those MIT guys professional scouts or managing professional athletes or professional sport teams? hahah I guarantee they would all laugh about this.

You should watch that movie moneyball with Brad Pitt, and see how a real MIT guy changed the face of baseball as we know it today. Someoen should give him a call lol.

I still think the rating system made by gaimtheory for quakelive, is the best ever made to date. And would of broke new ground if the game didn't die off for other reasons. Trueskill might better then a basic ELO rating system, supposedly, but its still an ELO rating system to me. and would only be good for rating a team not the individual.

I consider online multiplayers like MWO, sports. And treat them like such.

Stats are what mainly determine a players worth to a team. And stats should be what determines his "skill" in MWO. Stats determine how much a pro athlete gets paid, and it also determines how many cbills an MWO player gets paid.

not just one stat like W/L, or K/D or damage or assists, Unless your trying to fill a role/gap on a permanent team, but in reality all of the above. If you have a problem with that fix the point system. Which is the root problem of many issues with the game.
\
How long are we gonna wait for it to even out? For the game to die first? That statement is something us uneducated people can trust? You know us guys thats actually play sports? When we can see and understand the scores that we are playing for with our own eyes?

Again, the only math that matters to people, is the math on the scoreboard at the end of a match.


So, the answer is yes. You believe that 'all those people at MIT' are just wrong and math is some big sneaky conspiracy that smart people use to, what, help keep people down for THE MAN?

For giggles I looked up the documentation available for the ranking system for Quake Live - it was based on Elo, more to the point it was a broken implementation of TrueSkill which is based on the Elo system. Win/loss. Clearly a conspiracy, right?

So your opinion is that we should use metrics like damage/kills/etc even though it's an absolutely unreliable basis for representing skill but it'll make you 'feel better' about how you're ranked.

I'm not going to spend the time to argue that. I'll just go with 'no thank you'.





23 user(s) are reading this topic

0 members, 23 guests, 0 anonymous users