FAF Leaderboards 4.0!
-
This is awesome. Great job.
-
Looks good, a couple of suggestions:
- Add custom ranks as an option. While you may not see it as of interest, some will. If the win rate is bugged, then just hide this column only
- Allow the data to be sorted by clicking on a column header
- I assume it'd be a lot more work but just in case it wouldn't be hard to add, add a filter option to allow you to display only players between a certain rating, or with a certain win rate, or with a certain number of games
M27AI and M28AI developer; Devlogs and more general AI development guide:
https://forum.faforever.com/topic/2373/ai-development-guide-and-m27ai-v71-devlog
https://forum.faforever.com/topic/5331/m28ai-devlog-v130 -
@Javi I agree that global should 100% be there. Not that its super important or that its the best indicator of skill, but many people dont play ladder/tmm and want to be able to find themselves on the leaderboard or just want to see the current highest global player.
-
Don’t see the point in win rate existing when the whole point of trueskill is converging players to a 50% win rate alongside the bugs you mentioned. You’d be better off with a streak value or something.
If we have a short term leaderboard like a chess blitz event then win rate serves a better function imo
-
@maudlin27 These are some great features that could be added in the future. I'll try applying some of your feedback in the future to come.
Not so sure about global thought, could just erase the winrates when displaying it but even then
-
@ftxcommando average user understands winrate better than true skill xd?
FAF Website Developer
-
Looks great Javi!
I can understand not wanting the win rate up, though its good for showing people that losing is just as much a part of the game as winning (statistically).
-
@javi said in FAF Leaderboards 4.0!:
@ftxcommando average user understands winrate better than true skill xd?
Average users understand bigger number = better, having a leaderboard that makes you think bigger win rate = better is going to lead to frustration in a skill based matchmaking system. Like pepsi has a high 30% win rate in that picture. That probably puts him in like the bottom quartile of FAF. Is that a reasonable conclusion to make?
-
That is a great background. Top/bottom fade ins remind me of the scroll effect on this website monokai.nl — which, pretty sweet imo.
Can we consider the placement of "Next"/"Previous" buttons? Both on the top and the bottom of the page. I'd like to switch around both the buttons and the position on the page.
Current
To enable easy navigation, position the navigation centered and consider the following layout:
"Previous"/"<Page Number>"/"Next"And... in an ideal world <page number> obviously can recieve user-input
.+ for sticking with Electrolize.
-
Something really wrong with placement of this buttons
-
@Javi
Re-add global rating leaderboard data. It's useful/nice to be able to look at and search a global rating leaderboard. Far more custom games get played than any matchmaker queue type, and many people play mostly or entirely custom games. If win rates are too inaccurate, they don't need to be displayed; they're not so informative in our TrueSkill-based system anyway. -
I feel like winrate is broken, is that possible to be 2400 with lower then <40% winrate xD? its should be like 1 win vs 2300 when I was 1700 with 3-4 loses. And like 2k games with it. Or I always played versus higher rated as like 400 more rated playes. Or was like lose first 1k games after I got my first win XD.
UPD: Or maybe thats stuff for real showing broken rating drops by massive players of our community when I was lower ranked. But thats for realy strange to have much rating and have lower then 50,0001% winrate, it means, I lose half of games, and still keep at same place, or idk xD wtf, how to analyse thats.
UPD2: Also Kazbek stats showing 52,28% winrate, Yea seems like its broken anyway.
-
Also all time winrate dont showing anything, the good idea its winrate of 3 last month, or 6 month.
-
My winrate has consistently gotten worse as I've gotten better at the game
-
https://aoe2.net/#aoe2de-leaderboard-rm-1v1
Maybe can take some things from this
-
There is no point in having a custom game leaderboard. PLayers have so much control what maps and positions and with whom they play, that a rating comparison of different players makes no sense, which is the whole purpose of a leaderboard.
In this community we are already an outlier by providing rated custom games. Many other games don't even do that. -
There absolutely is a point. We have already established that although global rating is flawed, a very large portion of the community cares about global rating and considers it useful/desirable to see. If you can't remember that, you're welcome to reread the global rating-related discussions circa last year's PC election.
-
Not only is this a terrible argument that has nothing to do with the purpose of a leaderboard but you also attempted to move the goalpost from “the need for a global leaderboard” to “people like global rating” so nice one on that.
-
If we could include accurate win rates that would be cool, but the current win rates used have been broken for some time. The streaks from AoE seem cool. It would also be interesting to have longer timespans as an option, but not default, perhaps all time, 5 years, and 1 year.
