The new ELO-based ranking system

Stucifer

@AndrewAAGamer Yes, ELO will not award 0 points for winning against an evenly matched player. The points awarded for winning against an evenly matched player are more than for a player ranked below you, and less than for a player ranked above you.

For example if I played you in OOB, your ELO is 1774. It would be my first game in OOB, and so my ELO is 1500, and my K value would be 140. Your K would be 70 as you have already completed at least 10 games.

If I won, I would receive 109 points,

And you would lose 55 points-

Stucifer

For evenly matched players, if the ELO is exactly the same, it should award/deduct whatever the K value in points is for the winning/losing players.

pacifiersboard

@mr_stucifer said in Proposal for a new, ELO-based, ranking system:

So in the case of exactly even ratings, E(a) = 1.

However I could see this being an issue as that would make the winner get 0 points if I’m understanding correctly:

I read it as a multiplication but the formula is actually an Exponent, so having exactly even rated opponents results in an E(a) of 2, not 1.

That means that the points awarded to the winner will be significant. If they are in the 70-point stage for K, it will award the full 70 points to the winner and deduct from the loser’s ELO.

Oh, I had misread it as a multiplication, too!

AndrewAAGamer

Okay - so all this math is WAY over my head. As long as all you math nerds know what you are doing, I feel fine in your hands.

MrRoboto

Okay so you guys figured it out already ;-)

There is no mistake in the formula and divide by zero is never possible.
1+10^(x) would have to be zero for that.
but 10^x is always positive.

.
.

@AndrewAAGamer said in Proposal for a new, ELO-based, ranking system:

BUT, there has to be a standard amount for winning/losing a game against an equal opponent. It is just as difficult for a Tier 2 player to beat another Tier 2 player as it is for a M player to beat another M player. Both winners should go up while both losers should go down by a standard amount. Any system that does not award points for beating an qual opponent makes no sense at all.

And this is the case.
Have a look at some theoretical results I just entered in (I was too lazy to search for actual ones, but I’m sure there are some)

If equal players of any level play against each other, the result is always the same.
@farmboy and @trulpen are basically the same skill level so Ea=1/2 and K-factor with these number of games is 70.
Same with @aagamerz13 and @Odonis
Or with @Sovietishcat and @oysteilo

If the best player against the worst player wins, barely anything will happen though:

@pacifiersboard
Thank you for digging in and trying to find flaws in my system. Highly appreciated! This whole project can only benefit when people are trying to optimize it.
We already cleared your misconception of the current formula. Additionally, your proposal would also be dependent on the highest and lowest rating. This would mean that past results would retroactively change ELO change WHENEVER the highest or lowest ELO changes. Highly problematic! This also creates circular references again.

MrRoboto

New feature just added:

You can filter the Results tab for a specific player! Just type in the name of the player at the top, in the highlighted cell.

You can also filter for:

Type (BM4, OOB or PtV)
Wins
Losses
Axis
Allies

gamerman01

@MrRoboto And that feature takes the place of the PVP matrix I’ve been doing, so that particular information is not lost -

Great!!

Gorshak

This new system and calculation sheet looks great!

I am new to the league, but I could already see a few disadvantages of the old system - that have now been addressed by the new system.

Great work!
💪

But, unfortunately, I cannot support any system that changes my league standing from #1 to #17
😅😜🤪😈

farmboy

@MrRoboto I was enjoying seeing standings change as the results come in.

I can’t really expect any system to get things exactly right. As long as players of comparable level are ranked close to each other that would be enough. And that certainly seems to me to be happening here. That being said, I suspect that more data will still make some significant changes to what the standings are. I expect that players like Adam and Axis-D (and JDOW when he appears) and a couple of others will move up relative to the rest of us, and a few others (like me) will consequently move down a bit.

I’m happy to add some data when I have the time. But right now this is only read only for me so not sure how to do that.

The other thing I still have a concern around is the yearly playoffs. If we want a playoffs based on the top 8 (or top 16 as ghostglider suggested) in a given year, I’m not sure how we pull that data from this (such that one’s lifetime ranking doesn’t impact it). I’m sure there is a way but am curious what the thinking is there.

MrRoboto

Thanks for the input @farmboy

Axis-Dominion is already #2 so when you say you expect him to move up, I’m a bit scared ;-)
JDOW has only 1 recorded game since February 2021 (where the data stops right now), I’m also curious to see where he stands when we add earlier data.

The file is read only, besides the “Results” sheet, where all of you can filter for names, but also for sides (axis or allies), for winners and losers or for type (oob, bm4 or ptv).

If you want to contribute by adding more data, mr_stucifer has created a new spreadsheet and shared it with me, for example

MrRoboto

Actually, I just added a new sheet, you SHOULD all be able to edit it.
That way we don’t have duplicate data in case more than 1 person is adding in data at the same time.

There are only 6 cells per result to enter:

Type (BM4, OOB or PtV)
Date
Axis
Allies
Winner
Bid

farmboy

@MrRoboto I’m not sure why but I still am unable to put info in. I can’t use the filter either.

AD is ranked fractionally better than me right now. And Adam is fractionally worse But based on their record over the past few years (they and JDOW are regularly ranked 1,2 and 3 of playoff eligible players) I expect we are going to see them move further ahead and put some distance with the rest of us once we input more data. It would be deserved.

farmboy

I did do feb 2021 and sent it to Roboto in case anyone else was going to work on this

Stucifer

@farmboy Thanks, I expect to have some time tonight to keep working on it so I’ll plan on Oct-Dec 2020 and January 2021

Stucifer

October 2020 - January 2021 is now complete, @MrRoboto can update when he has time.

gamerman01

@Gorshak said in Proposal for a new, ELO-based, ranking system:

But, unfortunately, I cannot support any system that changes my league standing from #1 to #17
😅😜🤪😈

There’s one really effective way to remedy that…

Wait, how did you do emojis? As images?

Stucifer

@gamerman01 you can use them if your keyboard supports it (like from mobile) 🤓

I use emojidb.org when on desktop to copy (it seems like the most user-friendly one I’ve found so far)

gamerman01

🔥🔥🔥🔥

Got it, thanks, just what I needed, you are fast

gamerman01

@farmboy said in Proposal for a new, ELO-based, ranking system:

I can’t really expect any system to get things exactly right. As long as players of comparable level are ranked close to each other that would be enough.

Right… and I would add, high level of predictability as to who will win. Of course, the greater the difference, the higher the predictability should be

And that certainly seems to me to be happening here. That being said, I suspect that more data will still make some significant changes to what the standings are. I expect that players like Adam and Axis-D (and JDOW when he appears) and a couple of others will move up relative to the rest of us, and a few others (like me) will consequently move down a bit.

The ratings crossing years is new and exciting, and like you, I can’t wait to see how Babe Ruth matches up against Derek Jeter, though arguments can still remain 😆

The other thing I still have a concern around is the yearly playoffs. If we want a playoffs based on the top 8 (or top 16 as ghostglider suggested) in a given year, I’m not sure how we pull that data from this (such that one’s lifetime ranking doesn’t impact it). I’m sure there is a way but am curious what the thinking is there.

Some of the top goals of the system we’ve been using for many years has been

Clean 0-0 record every year, while allowing past years’ results to have some impact.
Annual playoff focus - the system was designed as an improvement of simple win/loss % while maintaining a focus on post-season playoffs.

The following is not to be down on the new ELO system that has many big improvements, but is a demonstration of the quandary that farmboy poses. How do you get a good playoff matchup each year and get good annual ELO results without taking into account any information that occurred before January 1 of this year?

I resisted ELO for years because the rating only goes up or down a limited amount per game, and everyone starts in the same place. If God plays against a tomato, His rating goes from 1500 to 1560, and the tomato’s goes from 1500 to 1440. They would have to play a lot of games before God gets to the perfect 3,000 (indeed it may even be impossible for Him if there is not strong enough competition to play against) and before the tomato goes down to its rightful 0000.

ELO does not do well in our league within a single year and creating good playoff matchups based on that year, and one reason is one player is playing 6 games (the minimum for playoff) and another is playing 50, and everyone else is in between.

However, like farmboy I think there’s some kind of solution.

Without even thinking very long and hard about it, I think it involves factoring in lifetime results, or results over the past few years.

gamerman01

In response to the complaints/criticisms about “circular references”, there is a very good reason for why points are adjusted after a past opponent changes tiers.
Just one example is when we get a new-comer to the league who’s actually a veteran of A&A and online Triple A. They come in as a completely unknown quantity, one of our guys graciously takes him on, and the new-comer blindsides him. All we know is he’s 1-0. He continues to win, and at the end of the year he has a real nice 5.67 average.

Under the current system the first guy and the second guy, and the third guy who faced him and got ambushed are adjusted up as this new guy rises tiers, and rightly so. They played against strong competition and lost.

With ELO, there is no correction whatsoever - if you lose against a newcomer, it makes no difference whatsoever how good he actually is.

In the past, the complaint has been that your average points per game will lower even if you win. This presupposes that the goal of the ranking system is to earn points and see how high you can get. If that is your goal, then you won’t take on a player more than a tier or two below you because you dont’ want your average to drop. But not everyone is like that. Maximum points has never been the #1 goal of the ranking system. Sizing up the strength of your potential opponent, having a clean slate each year while also taking into account historical strength in making good playoff matchups have been higher goals. The misconception is quite understandable, but players with this complaint do not understand the ordering of these priorities.

Now doesn’t this mean it’s much riskier to play a newcomer under ELO? Under the current system, if it turns out he’s a wiz, you get credit for that if he plays a few games before year-end. Under ELO, your score will go up when you beat him in his first game here, but there is absolutely no account for how good that player actually was or is. He’s just another 1500.

The new ELO-based ranking system

Featured Topics

T-shirts, Hats, and More

Suggested Topics

L25 OOB Donutgold (X) v wizmark (L+53)

L25 OOB elche (X) vs. Martin (L+35)

Wiz (X) vs Sunkhareb (L+21) BM4 G1 L25

Wiz (X+13) vs Stucifer (L) PtV L25

L25 PTV Stucifer (X+11) vs Axis-Dominion (L)

L25 PTV ArtofWar (X+10) v Stucifer (L)

L24 PtV Gorshak (Axis+8) vs 666 (Allies) #1

League General Discussion Thread

33

17.8k

40.4k

1.8m