Automatch statistics: Map vetoes and win percentages
Posts: 4301 | Subs: 2
Posts: 493
The least vetoed maps in 4v4 are ... Port of Hamburg
U WOT
games seem to happen on Steppes more than twice as often since then
Because vetoing it doesn't work.
Posts: 1138 | Subs: 2
Because vetoing it doesn't work.
Yes, there were reports of people getting Steppes even though they vetoed it. However, it seems unlikely that that's the (main) cause for Steppes appearing more than twice as often (a little more than 20% recently, vs. 8.6% before).
Edit: I base that on the fact that the least vetoed maps end up being played in around 10% of the games and I suspect that simply not vetoeing a map can't get it up to 20%. For example, Rzhev Summer in the 3v3 pool is definitely not vetoable, yet still it is played only a little more often than Across the Rhine (8.2% vs. 7.3%).
My guess is that the place of Hill 400 on the server list of maps now went to another copy of Steppes, so that the chance of getting Steppes is twice as big.
I don't think the numbers are accurate enough to conclude how vetoes are currently working regarding Steppes (given that it wasn't vetoed as often before). It might be that the veto indeed doesn't work for Steppes at all. Or it still works for the original Steppes, but not for the "Hill 400" copy of Steppes.
Amended the section accordingly...
Posts: 493
Posts: 2066
Posts: 1295 | Subs: 1
I genuinely wish, as a mapper, there was some way to get live data from relic on all of these statistics to better develop and understand map dynamics, balance, likability and quality. Relic should seriously consider this when they design their next COH game...
_________________________________________________________________
These two diagrams in particular were great for me. They validate Crossroads design. This has made my day. Thanks again for putting this data together gentlmen.
Posts: 851 | Subs: 1
Suddenly I come to understand why Lorch is so god damned hard as axis
Posts: 2458 | Subs: 1
Posts: 2458 | Subs: 1
These two diagrams in particular were great for me. They validate Crossroads design. This has made my day. Thanks again for putting this data together gentlmen.
Crossroads is probably the best 1on1 map in the game. Fair and fun to play. With lots of different strategy options. You did very well.
Posts: 449
Wish we had this in the first year of release when Relic still had a team on payroll doing balance.
Posts: 2181
Posts: 1295 | Subs: 1
thanks blvckdream
Posts: 612 | Subs: 1
Posts: 10665 | Subs: 9
Posts: 96 | Subs: 1
Posts: 1138 | Subs: 2
I genuinely wish, as a mapper, there was some way to get live data from relic on all of these statistics to better develop and understand map dynamics, balance, likability and quality. Relic should seriously consider this when they design their next COH game...
Well, the data is kind of from the perspective of the players; I guess as mapper you'd like to have some more numbers.
I would be interested to see in particular how the win percentages for the starting points differ, but those don't seem to be in the data packages that I retrieve (at least I couldn't find it and I tested some options...).
I could do win percentages for different matchups, though...
Suddenly I come to understand why Lorch is so god damned hard as axis
Unfortunately, the post doesn't tell you the "why", but yeah, it's not you
Nice article, how did you get this data exactly, is it publicly available somewhere?
Sort of. A bit simplified, I have a script that retrieves the game history for each player on the ladders twice a day (it uses the API that the official leaderboards use to generate the history when you click on the "recent matches" tab there). The data packages that you get contain somewhat more information than what is displayed under that tab, for example the map name.
Posts: 1295 | Subs: 1
Thanks again for the great writeup
Posts: 3602 | Subs: 1
Posts: 5441 | Subs: 36
Posts: 1467 | Subs: 4
Thats excellent man. Relic should formalize this kind of software and use it for ANY RTS game they create. This data could be used to watch the maps from a higher perspective and refine ladder map lists. This could be a hugely powerful tool if you focused on the right data and had goals that utilized the data. Make it public and then everyone would benifit.
Thanks again for the great writeup
Had this in the works for a while, will be testing the waters with Relic on it soon. Have a meeting with them tomorrow or Friday to discuss this, and many other things.
Livestreams
40 | |||||
22 | |||||
1 | |||||
3 | |||||
3 | |||||
1 | |||||
1 | |||||
1 |
Ladders Top 10
-
#Steam AliasWL%Streak
- 1.831222.789+37
- 2.610220.735+4
- 3.34957.860+14
- 4.1110614.644+11
- 5.276108.719+27
- 6.306114.729+2
- 7.916405.693-2
- 8.722440.621+4
- 9.261137.656+2
- 10.1041674.607-2
Replay highlight
- cblanco ★
- 보드카 중대
- VonManteuffel
- Heartless Jäger
Board Info
8 posts in the last week
34 posts in the last month
Welcome our newest member, Pound309
Most online: 2043 users on 29 Oct 2023, 01:04 AM