Analyses of the ratings - Spotting the issues

Inflation causes elo manipulation on different ladders cause of the initial boost, as it was mentioned several months ago.

Elo points distribution for multiplayer games gives usually way less points to prevent inflation, it was like that on zone, igzones,voobly and game park, idk why DE devs thought that giving 17 points each player for a 4x4 was a good idea, on 1x1 you can’t stack but on team games you can, manipulating once more the ladder.

1 Like

I really really really hope the devs will fix this issue in the next patch… It will also solve part of the Alt+F4 reasons from players.

@OMGBoxShark

3 Likes

Now we also get EW, it is really time that the devs fix this. Otherwise the EW 1v1 and TG ladders will start as a mess as well, like the current TG ladder. I posted the following in the announcement thread of ranked EW:

I would even advice to just fix the formula (this look easy, just replace the max by the average) and then reset all ladders. That is the most easy fix with the least effort for the dev team. If you dont reset the ladders, then TG ladder will still be a mess. And also the newly EW will become still a mess, since they will use the ratings of other ladders as starting point. Instead of sorting out their mess, the most easy solution will be a full reset of the ladders. But that might be a unpopular decision…

Edit: Looks like @ChristheCo is tracking the bugs. I see his name pop up in the bug report section on behave of the dev team. So i will tag him here too. Hopefully he can make sure this issue will also be tracked quickly. I started this thread 1 year ago, but it is still not part of the Known Issues & Solutions – Age of Empires Support section?!

4 Likes

Fix the formula for tg rating! Reset the ladder!

As long as this doesnt show up in known/tracked bugs I can only conclude that this forum has absolutely no purpose when it comes to reporting bugs. How can this be ignored for more than one year?
The only logic explanation is that the devs have no intention of actually fixing their game and this is just a place to confort the players.

2 Likes

Some fixes coming.

3 Likes

It is great to see that the devs finally try to fix this issue. The calculation will be fixed in during the current server maintainance. This means that there will be no inflation any more in games after the current maintainance!

I only hope they will also take care about the current inflated players (high rating, low win rate). They said

but this takes really really long in my opinion. In their blog they speak about

I think this is really needed. I posted that under the blog as well ([BLOG] Age of Empires II: Definitive Edition — Updates to Ranked Team Game ELO Acquisition - #3 by WoodsierCorn696)

I am glad the devs do take these issues seriously. It will make playing the game much more fun for many players. Fixing this issue, will also help with the number of Alt+F4’s, so that is great as well.

3 Likes

According to that post, then a 3300 team can face a 2500 team and still winning 16 points, inflation for stackers will sky rocket.

How? Its the average. One teams average will be 3300 and the other’s 2500.

I don’t think thats how the example was working. You avarge the team elos together and then it makes the call in the case of 3300 vs 2500 the 3300 team would earn 1 elo each for the win. In the examole in the blog both teams elos average was the same so the loss or gain was on equal grounds hence the 16.

2 Likes

Yeah it was bad explained i guess, i tested it out, everyone gets the same points as the higher player of the team, meaning if you stack with 3200 you only get like 4 points even if you have a way lower rank than him, good change then.

The announced change is a huge improvement. Still not perfect (no system is perfect) but we are likely to see win rates much closer to 50% in some time, and so games will generally be much more balanced.

1 Like

Hey guys! So what is your impression with the new tg rating system?

My tg rating does up a lot, bc my teammates are lower rated, but now we all get the same points, so i am happy.

I think excessive smurfing should be the next thing to be addressed.

1 Like

I am glad that a fix to the rating system has been applied. However there are a couple of key points that the new system doesn’t address. They have been discussed in this thread previously but I will restate them here as a recap.

  1. The mean of ratings may not be the true skill level of a team. In games with disbalanced players, the result is more dependent on the performance of the higher rated players in both teams. This suggests that some form of weighted average is a better indicator of a teams win chance. Here Mercy9545 suggests one way to do this that I like. gnarfk also suggests an improvement in this direction. The old system gave a weight 1 to the top player and weight 0 to everyone else. The new system gives weight 0.25 to everyone in a team of 4. The balanced answer may lie somewhere in between these extremes.
  2. Suppose we have a team of 4 players with 1v1 rating 2000 playing against a team of 4 players with 2050 1v1 rating. A 50 elo advantage is equivalent to 57% chance to win in 1v1. But when they are put together, the chance to win will be even higher. (just like a bundle of sticks is stronger together). To calculate this exact advantage we may need data but for now I will assume that as a team they have 60% chance to win. This is a 72 elo advantage. So their team game ratings will drift till they are 72 elo apart. This phenomenon doesn’t cause inflation in the TG ratings like before but it will cause it to be more spread out than the 1v1 ladder. This means we will still not be able to make a direct comparison between 1v1 and TG elos. A possible solution to this problem is to multiply the difference in mean elo with a factor dependent on the number of players to get the true chance to win.
2 Likes

this is what I have raised out before few months ago but not many people stand by my side.
probably most of those are those who take premade teams who want to enjoy the high rate winning.

Havent seen much difference yet. I guess in a few weeks the 39% win rate players will finally make their way down and I should see more players with 50% win rate on my games.

1 Like

So the devs fixed the inflation about 3 weeks ago. Lets us have a look back to the last period.

1. The team game ladder is still a mess.

There are still lots of players with an inflated elo. Just an average 1v1 player will still make it to the top of the RM board. Just to give an example:

No name, because it isnt about this specific player. It is about the stats. This player is still somewhere in the top, while for 1v1 it is just and average player. It will really take a long time to clean up the ladder without interference of the devs. I would say it will takes years and that isnt acceptable.

The best way to fix this mess is a full reset of the ladder.

2. The new system is exploitable

Just look at the number 1 on the ladder:

He is about 600 points clear of everyone and the first one who hits 4k. How did he do this? He just played many games with smurfs. Have a friend who makes a smurf account. Just Alt+F4 when the game starts, so you drop you rating even below 1k. And 800 + 4000 = 2400 on average. So the TG elo is calculated based on 2.4k strength, so he still gets more and more elo. Smurfing is already pretty toxic, but it looks like it is a valid strategy at DE atm. He wont get banned for this behaviour. This point definitely needs a fix.

I suggest a small adjustment for premades to deal with big elo differences. For calculation the strength for a premade you dont just look at the average, but there is also a minimum floor based on the maximum elo of the premades. I suggest upping the elo of every player in a premade compared to the maximum elo of the premade.

For example: If the elo gap is more then 500 (note: This number is just an example), then the lower rated players will get the fictive elo of Max(elo) - 500. Based on this elo the elo the team strength is calculated. To illustrate this with the example from above: 4k + 800 will end up as 4k + 3.5k. As result the average will be 3.75k for the elo calculation. So previously they got matched against 2.4k players, but now they are 3.5k and got matched against 3.5k players. Also in the calculation for the update of the elo, this TG elo will be used. This will greatly reduce the impact of smurf on the ladder. You cant really boost your elo with smurfs any more.

3. Will team games ever be balanced?

Balancing TGs is much harder then balancing 1v1s. And even for a 1v1 a game might feel extreme easy of difficult sometimes. It pretty much already depends on how you play that day, you civ match up, the map, your strategy vs opponent strategy.

For team games this is even worse. You have the same variables for every player on both teams. It not is only about your strategy vs your opponent strategy, but also your team strategy vs you opponents team strategy. Its your team work vs their team work. Your teams consistency vs your opponents team consistency. It is not only about the civs picks, but also about the position of these civ picks. Like pocket cav civs, flanks archer civs. All these factors will add much more variables and makes the outcome much less predictable. As result team games will always be much less balanced then 1v1s. I cant really see any fix for this point. There are some slight improvements. You can find examples of games of only solo players which can be much more balanced based on their elo.

1 Like

I have already finished my 50 games record,
when I vs solo I can get around 60% win rate, while I only get 30% win rate vs solo.
It’s not all about the inflated elo man.

Even there’s no any inflation or deflation of elo, solo vs premade is just a mistake especially AOE2 is kind of different than the other games.

1 Like

Perhaps TG Elo shouldnt be a thing and matchmaking should use the 1v1 Elo as the parameter for TG Matchmaking.

2 Likes

You can’t possibly believe this would be a good idea.

1 Like