WEBSITE

[Majorcrisis2]

I can probably do gamescore on the google sheets, it depends on how complicated the formulas used are. I am willing to make the attempt if someone is willing to post the formulas used. It would be nice to see how gamescore stacks up against the ratings I have been working on.

In wb/jav they are pretty simple. As in k-d+team kd
The basing ones had a while formula behind it. Pretty sure ogron still has them somewhere, else I gotta dig deep somewhere...

[Nipple Nibbler]

In wb/jav they are pretty simple. As in k-d+team kd
The basing ones had a while formula behind it. Pretty sure ogron still has them somewhere, else I gotta dig deep somewhere...

Claushouse You have them gamescore formulas around somewhere?

Hmmm, if it involves team stats, there is no way to accurately calculate that when people don't play the full game (!sub, !switch, !change).

<PROBLEM>
Let's say for some reason you start the game with steadman in terr. But he fell in his toilet or something during pre-game and is AFK in the ear, now you need a new terr. You sub in major crisis here, he has a major great game for the other 25 minutes of the game. Now he gets 0 gamescore since he missed the first 90 seconds. Obviously this isn't limited to just base.

<POSSIBLE SOLUTION>
I could get a rough estimate by doing something like:

Total team kills / time to get a per min avg and use that to roughly estimate how many kills the team had at the time of the player leaving the game. If that would be acceptable I can do that.



P.S. sorry if we are derailing your thread a bit Wirah.

[Majorcrisis2]

Claushouse You have them gamescore formulas around somewhere?

Hmmm, if it involves team stats, there is no way to accurately calculate that when people don't play the full game (!sub, !switch, !change).

<PROBLEM>
Let's say for some reason you start the game with steadman in terr. But he fell in his toilet or something during pre-game and is AFK in the ear, now you need a new terr. You sub in major crisis here, he has a major great game for the other 25 minutes of the game. Now he gets 0 gamescore since he missed the first 90 seconds. Obviously this isn't limited to just base.

<POSSIBLE SOLUTION>
I could get a rough estimate by doing something like:

Total team kills / time to get a per min avg and use that to roughly estimate how many kills the team had at the time of the player leaving the game. If that would be acceptable I can do that.



P.S. sorry if we are derailing your thread a bit Wirah.

I entered data as time played etc. Cuz someone going 20-1 in a 15 min game isnt as impressive as 40-2 in 30 mins imo. Also switches and subs were able to get calculated that way.. Were I ignored ppl playing less thsn 5 mins.

[Rab]

The schedule needs advancing to show week 1 by default.

The history needs updating, see ?go twdtb

[Wirah]

Nicely spotted. It's still set to week 0 (preseason). Will fix tonight

[Nipple Nibbler]

Another bug, flew's stats are messed up in spider on the website and the API.

[Wirah]

Nipple Nibbler can you explain what's wrong with the stats and I'll investigate and fix for Week 2. I know we get certain issues pop up occaisionally that sometimes automatically resolve themselves after the next set of games. If you let me know which stats are wrong and what they are meant to be, I'll be able to debug

[Nipple Nibbler]

Wirah

<copied from scoreboxes>

70

Flew

3

40

35

0

0

0

7%

49

0

1600x900

70

Flew

3

104

71

0

1

2

10%

121

0

1600x900

<Copied from DT site for spider he has this:>

Flew

Patouf

7

2

1

17

17

1

0.57

0

35

he has 144 kills not 17
he has 106 deaths not 17
he has 2 teks not 0
KPD and KPM are way off as well

<in the api>
in addition to above he somehow he has a tk in spider and 69 bombs.
he has only shot 67 bullets and has short, normal, long, extra long stats (I assume this is the R1 R2 R3 R4 for wb/jav?)
he also has 0 fc here and that should be 1

<theory>
In jav he is 17-17 in with 1tk and the same R1/R2/R3/R4 stats. Maybe it is double counting his jav stats as spider stats.

[Rab]

Wirah When you click on the matches for a future week it's fast, but when you click on a past week it's slow. Probably something to do with reading the results. Can't be that much data so it shouldn't be so slow, and the results don't change so could be cached.