F400 Clan Ranking - Needs Repair

[IcePack]

Archive

[Fewnix]

EXTRA !!! EXTRA!!!
READ ALL ABOUT IT!!!


Legion is 15th in the two year rank, 8th in the one year rank and on the way up



WOO GOO!! WOO GOO!!

[sdh]

How comes HDN are on the rankings with a record of 0/3 but TWD are not with a record of 3/3?

[IcePack]

How comes HDN are on the rankings with a record of 0/3 but TWD are not with a record of 3/3?

HDN had a 100+/- game war mixed in there. 180 +/- total
TWD had a 60+/- game war with 140+/- games total.

They took a heavier load, and showed up earlier. The win / loss record doesn't help or hinder getting on the board.

[Emmdizzle]

EXTRA !!! EXTRA!!!
READ ALL ABOUT IT!!!


Legion is 15th in the two year rank, 8th in the one year rank and on the way up



WOO GOO!! WOO GOO!!

Congrats I'm pulling for you with all of those ex PACK stars in your ranks

[agentcom]

Hey Ice,

Since cheme just posted it on his leap ladder, I was wondering if you have a readily available statistic of how often the F400 predicts the winner of a clan war.

[IcePack]

Hey Ice,

Since cheme just posted it on his leap ladder, I was wondering if you have a readily available statistic of how often the F400 predicts the winner of a clan war.

I don't have any stats like that, though I suppose I could take a sample time frame and look into it.

[Leehar]

Hey Ice,

Since cheme just posted it on his leap ladder, I was wondering if you have a readily available statistic of how often the F400 predicts the winner of a clan war.

I don't have any stats like that, though I suppose I could take a sample time frame and look into it.

I think FD did have something similar

viewtopic.php?f=438&t=140929&start=15#p3088157

• Compute accuracy of the model, in terms of how often the higher ranked clan actually won
  
  
  Now you can know how accurate the model is in real terms, measured by how often the higher ranked clan wins for various rating difference intervals
  
  Min Weight to include in Accuracy = 100
  
  
  This means, that given a match between two clans, where one clan is 109 points above the other, it is 82% likely to end in a non-upset. Btw, this 109 point value was auto-calculated as 10% of the range between min and max ratings (after excluding clans below weight threshold).

[IcePack]

Yeah, but that's not quite the same as chemes stat as his was an overall.

[agentcom]

Thanks Leehar. Very interesting and (fortunately) not at all surprising.

Ice, might be an interesting stat to keep even if you started it from here forward. You could likely automate something similar using the current F400 table and the historical results rather easily. That wouldn't exactly tell you whether the top ranked team at that time won, but it would probably be close. Since you keep of what place each clan is in at each update, you could go a step farther and calculate the actual win percentage based on ranking at the time of the war.

[Auger13]

how is ACE ahead of TOFU in the 2 year rankings, but behind them in the 1 year rankings? ACE isn't yet a year old, so I'm not sure how their wars from a year or more ago push them over TOFU in the 2 year rankings.. and it's not like TOFU had a good past year, but a bad year before that. I'm just a bit confuzzelled as to how that works?

[betiko]

31-18 LHDD beating FOED, last game ended on the 6th of may

[benga]

AFOS-ACE 7-8 2014-05-10

[IcePack]

Archive

[benga]

2014-05-17 ATL - OSA 19-22

[betiko]

+3 positions for lhdd and 2 war wins coming up! Top 10 here we come!

[uckuki]

Random Cup

FAD vs G1 [21-40]
Game 14291006 2014-05-09 19:23:44 - ProsthoDoc, Joey3GA, ADAIR410 won the game

VVV vs AFOS [14-47]
Game 14288278 2014-05-17 00:22:15 - lanyards, Eldritch, Most_Wanted, bahrain won the game

[The_Samurai]

(RC-R1) RET Vs GR (43-18) RET WINS - FINAL 05/25

[betiko]

(RC-R1) LHDD Vs GON (39-22) LHDD WINS - FINAL 05/29

[uckuki]

[RC-R1] FOED vs HH [34-27]
Game 14191109 2014-06-01 06:19:42 - badmunkay, Jurasu, JCsolidrock won the game

[benga]

2014-06-03
[RC-R1] ATL - OSA 26-35

[nicestash]

how is ACE ahead of TOFU in the 2 year rankings, but behind them in the 1 year rankings? ACE isn't yet a year old, so I'm not sure how their wars from a year or more ago push them over TOFU in the 2 year rankings.. and it's not like TOFU had a good past year, but a bad year before that. I'm just a bit confuzzelled as to how that works?

My guess is that during the previous year, TOFU had not played as well as during the most recent year so it evens out. I'll show an example-year two is the most recent year

Tofu
Year 1: TOFU rating: 1350
Year 2: TOFU rating: 1390
Overall: TOFU rating: 1382
(the most recent year is weighted more heavily than the previous one)

ACE
Year 1: ACE rating: N/A
Year 2: ACE rating: 1384
Overall: ACE rating: 1384

This are hypothetical numbers, but you can see how it is possible for TOFU to have done better during the most recent year than during the last 2 years, yet still be behind ACE is the last two years.

[IcePack]

I've been a bit backed up due to all those Whac games but I'll update Wednesday .
Sorry for the delay!

[betiko]

you'll do a little update for the 15th ice?

[IcePack]

I'll do an update for the first tomorrow, and another for 15th. Yes