349 replies to this topic

[Jay Leon Hart]

xe N on, on 05 January 2018 - 06:09 AM, said:

PGI could add this in their database.

They would have to track it for every single match though, it's a waste of resources.

[Q]

Jay Leon Hart, on 05 January 2018 - 09:37 AM, said:

They would have to track it for every single match though, it's a waste of resources.

Pretty sure tier level is considered one of those private information things, too, not that it much matters anymore.

[Q]

xe N on, on 05 January 2018 - 12:48 AM, said:

Interesting. If you lock up the top 10, top 20 or top 30 even, mediums mechs are quite under-represented. Most high scoring players play either Assault or Heavy mechs, even Light mechs apparently frequently score higher.

Conclusion: Medium mechs need a major buff to compensate for being less armed, less armored but only minimally more mobile and slightly smaller than heavy mechs.

Dangerous to make too many conclusions. The data is fun and useful individually but as many others have pointed out the stats are skewed to a particular playstyle and you can "game" the leaderboards by a fair degree.

[Scurro]

Q, on 05 January 2018 - 10:34 AM, said:

the stats are skewed to a particular playstyle and you can "game" the leaderboards by a fair degree.

The goal of the formula provided in the Jarl's list for ranking was to reduce the amount of cheesing for stat ranking.

My opinion is that it is fairly accurate in representation of a players skill.

I have however noticed some players smurfing accounts to game their rank. It may cause me to make some adjustments to the formula and ranking system.

[MischiefSC]

Wow. Updated early 2/1/2018.

So sexy.

This is an excellent source of viewing stats. Thank you, well done.

[Sebaztien Hawke]

...If I started checking these stats I would have to care about whether I survived the game or not. Also, you’d never experiment or use a new and unskilled mech. I better leave it alone.

But if the top person only has 84 games, are they not a very experienced player using a separate account? Doesn’t that break the whole leaderboard? I know my stats or performance now would be much improved if I erased my first 500 games or something...

[Bombast]

Sebaztien Hawke, on 01 February 2018 - 03:52 PM, said:

...If I started checking these stats I would have to care about whether I survived the game or not. Also, you’d never experiment or use a new and unskilled mech. I better leave it alone.

If that's you're reaction to it...

Quote

But if the top person only has 84 games, are they not a very experienced player using a separate account? Doesn’t that break the whole leaderboard? I know my stats or performance now would be much improved if I erased my first 500 games or something...

General rule of thumb - Never use general leaderboards like this as an absolute. The Number 1 position probably isn't the best player, and the Number 40,000 probably isn't the worst. But your general position on it can still be a useful indicator. If you're in, say, the 10th percentile, you're probably not there because some dastardly alt account robbed you of positions.

EDIT: Also, newer matches have more weight than older matches, so you can 'push' old, bad matches out of your score, in effect.

Edited by Bombast, 01 February 2018 - 04:01 PM.

[Scurro]

Sebaztien Hawke, on 01 February 2018 - 03:52 PM, said:

...If I started checking these stats I would have to care about whether I survived the game or not. Also, you’d never experiment or use a new and unskilled mech. I better leave it alone.

But if the top person only has 84 games, are they not a very experienced player using a separate account? Doesn’t that break the whole leaderboard? I know my stats or performance now would be much improved if I erased my first 500 games or something...

There will always be those that cheese the system. I try to minimize this by having a requirement of 200 games for full score and they have to keep playing otherwise their account will be retired.

I may increase the amount of games needed to receive a full score if it becomes too much of a problem.

[UnofficialOperator]

Any plans to make it an onscreen overlay tracker like Hearthstone tracker?

Edited by UnofficialOperator, 01 February 2018 - 04:14 PM.

[Scurro]

UnofficialOperator, on 01 February 2018 - 04:14 PM, said:

Any plans to make it an onscreen overlay tracker like Hearthstone tracker?

I've been contacted by someone who is creating one using the Jarl's list database for the overlay. I've given them permission to use it. I don't have any more information to give at this time.

[Xavori]

As long as the matchmaker is broken, because the in-game ranking system is broken, none of these leaderboards matter at all.

PSR is effectively an XP bar. Unless you are the most potato potato that ever potatoed, you will go up. Hell, when I was actively working against going up, I was going up. Once I quit that, I started going up a decent pace.

The matchmaker then tries to use this broken PSR to make matches. Now, that works to keep newbies away from Tier 1 non-smurf accounts, but that's it. Everything else about the matchmaker might as well be random. And since the matches have random skilled players, the stats produced are random. Random numbers are random.

So the Jarl's list, while kinda fun to browse, ultimately becomes as meaningless as the official leaderboard.

But, but Xav...what about...blah blah blah

- A good matchmaker produces matches where both sides have a 50/50 chance of winning. That means that in a good matchmaking system, everyone should eventually have a W/L ratio that approaches 1. This is the easiest way to see just how bad PGI's matchmaker is.

-No matter how many matches, no matter what your actual W/L is, you can't say with any degree of certainty that you are responsible for your success because again, random numbers are random. Just because you have a positive W/L, that doesn't mean you are a good player. Most likely, it means you play on teams where you're eliminating as much of the randomness of the matchmaker as possible. But it's also possible that you just log on and get lucky with your teams.

-The reason number of matches doesn't matter is that player skill in MWO (or really any online game) doesn't follow a normalized distribution. This isn't rolling dice where every roll has an equal chance of producing the numbers 1-6. This is closer to rolling a dice where the numbers are 1,1,1,3,5,6. There are vastly more bad, below average, and just casual players than there are great players. And even among the good players, if they are playing a new, unskilled mech, or just goofing around, or are tried, etc. they would be a hindrance, not help, to their team. So, no Law of Large Numbers for you!

-The stats the game tracks actually have little bearing on team success, and quite a few match types don't even reward objectives in a meaningful way. Do you know how many conquest matches I've carried by focusing on objectives? Nope? Neither do I, although I have screenshots of a few memorable ones. And what about those 'boring' base rush matches. Do you see any way to identify how effective a pilot is at helping his team win those ones?

-A great player who can CT 90% of his alpha strikes will actually have lower damage stats, lower KMDD's, lower solo kills, earn less c-bills, and less XP than a potato who lives on streaks, ATM's, LRM's, and other targeted weapons. This comes up every time PGI runs a faction-based event that is tied to match score or damage done. You are far better off being a bad aim and padding your damage than in being an effective pilot. And even kills, solo kills, and KMDD's suffer from this. Kill is just last hit. A KMDD requires getting the most damage which goes back to padding damage being better than being efficient. And solo kills combine these two bad stats into one terrible stat.

p.s. I'm really looking forward to the 1v1 and 2v2 Solaris style games where we can finally see who the good pilots are. This doesn't mean they'd necessarily be good quick/faction/comp play teammates, but for the purposes of finding out who can pilot their leaderboard class mech the best, it'll work.

[Gas Guzzler]

I mean, all players have to contend with the matchmaker, so... all else equal, those who score higher on the leaderboards on a regular basis are generally better.

[Bombast]

Xavori, on 01 February 2018 - 04:20 PM, said:

p.s. I'm really looking forward to the 1v1 and 2v2 Solaris style games where we can finally see who the good pilots are. This doesn't mean they'd necessarily be good quick/faction/comp play teammates, but for the purposes of finding out who can pilot their leaderboard class mech the best, it'll work.

Surely you see the problem with claiming Jarls is flawed (It is, of course, but its not that far off), yet being excited about Solaris telling you who 'the real good players are' when its a game mode that's devoid of most of the games relevant mechanics?

[Gas Guzzler]

Bombast, on 01 February 2018 - 04:28 PM, said:

Surely you see the problem with claiming Jarls is flawed (It is, of course, but its not that far off), yet being excited about Solaris telling you who 'the real good players are' when its a game mode that's devoid of most of the games relevant mechanics?

Yeah, I'm inclined to agree.

A 1v1 and 12v12 team play have very different attributes that make a player good.

[UnofficialOperator]

Scurro, on 01 February 2018 - 04:20 PM, said:

I've been contacted by someone who is creating one using the Jarl's list database for the overlay. I've given them permission to use it. I don't have any more information to give at this time.

Nice...

[Bombast]

Scurro, on 01 February 2018 - 04:20 PM, said:

I've been contacted by someone who is creating one using the Jarl's list database for the overlay. I've given them permission to use it. I don't have any more information to give at this time.

Wait, we're getting an XVM?

Man, people are going to be pissed.

[Xavori]

Bombast, on 01 February 2018 - 04:28 PM, said:

Surely you see the problem with claiming Jarls is flawed (It is, of course, but its not that far off), yet being excited about Solaris telling you who 'the real good players are' when its a game mode that's devoid of most of the games relevant mechanics?

I said "pilots" not "players". I always try to be careful with word choices, cuz it matters I even pointed out exactly what you were talking about in the second sentence.

But if you put two players into roughly equivalent mechs and have them fight each other, than you can see between them which is the better pilot. PILOT. That obviously doesn't translate to player directly because there is more than just driving and firing your mech that goes into being a good player.

So ya, I think the leaderboards are junk. I also am looking forward to the new playstyles because I'd like to know how well I do when it's just me that determines by success/failures. These are not mutually exclusive concepts.

[Xavori]

Gas Guzzler, on 01 February 2018 - 04:26 PM, said:

I mean, all players have to contend with the matchmaker, so... all else equal, those who score higher on the leaderboards on a regular basis are generally better.

Nope.

Players who score higher on the leaderboards want to score higher on the leaderboards. That's really it.

You can play on teams thereby eliminating most of the randomness I talk about. You can only play the meta mechs. You can intentionally pad your damage stats which will end up boosting your other stats (except kills, those are still pretty random since short of having your teammates avoid kill shots, not much you can do there).

But if you play solo queue, or you play whatever mech you feel like playing regardless of whether it's a top tier or fully skilled or not, then your stats will suffer because you will play with more bad players than good, and in a team based game, much of your success (in MWO 87.5% of it) is tied to your team.

[Mystere]

Seriously, do people actually care about this stuff?

[Gas Guzzler]

Mystere, on 01 February 2018 - 04:58 PM, said:

Seriously, do people actually care about this stuff?

I know that really bad players claim it doesn't matter to make themselves feel better, but sorry if your average match score is less than 200 and your KDR is less than 0.5, you not credit to team, regardless of what mechs you are piloting.