BuzzerBeater Forums

BB Global (English) > The BB Training Crowd-Sourcing Project

The BB Training Crowd-Sourcing Project

Set priority
Show messages by
From: wozzvt

This Post:
22
203921.78 in reply to 203921.74
Date: 1/8/2012 11:46:51 AM
Overall Posts Rated:
228228
I'm wondering how it handles the partial minutes in the data...

First of all, everything that's posted uses only instances of 48 minutes, so not a concern that it could be skewing the data.

As for the effect of minutes, yes, showing it like age is the goal. The problem is controlling for everything. E.g., you could show it for 18yo's getting Pressure PG training, but that's so specific that we won't have enough data. You can't really group too many things, because the basic rates change a lot as a function of age, skill and training type.

The real solution, which I've started playing with, is to create a single model that can explain all the data, then doing parameter fitting to figure out exactly how much things matter. The problem with this is that you need to have a decent guess for what the equation should be. I mean, it's easy if things are linear, but minutes are almost definitely not, so you need a way to figure out whether it's best modeled as a decaying exponential, a power law, or something else you haven't even thought of. Which is why I've been trying to start with plots like the age one, it gives a really good sense of what that component should look like.

My current idea is to guess something close and do an analysis of residuals (looking at the difference between the actual and expected data) and seeing if you can use that to find a better model. Unfortunately, this is a bit of a computationally intensive process, so I haven't found a way to do it "live" in php yet (and the method is still *very* rough). And it's tough to combine across a lot of training types. But that's a sneak peak at the long term aim.

Some other good stuff in there that i'll think about wolph.

This Post:
00
203921.80 in reply to 203921.78
Date: 1/10/2012 2:38:29 AM
Kitakyushu
ASL
Overall Posts Rated:
12341234
I found a small problem...It has to do with inherited minutes..
It has to do with this player I bought..(21305614)
I bought him on a Friday morning(Dec. 16th, 9 hours before training) and he came with 48 minutes at C(really 56 but you know what I mean).
so when I look at the training history in the BB training project it says he was not trained. It only list the players that got minutes for my club. After I bought him my front page listed him at 48+, so I know he got the training. Is there anyway that the INHERITED MINUTES bug(don't know if it is a bug) could be fixed. Now my player is listed wrong in ID,IS and SB due to the not read 1 position ID training.
Thanks

Last edited by Yellow cake at 1/10/2012 2:39:12 AM

This Post:
00
203921.81 in reply to 203921.80
Date: 1/10/2012 8:00:45 AM
Overall Posts Rated:
246246
It might have been bug, because when I bought players I tried to train 'em using the minutes they played in their former team.

From: Yellow cake

To: red
This Post:
00
203921.83 in reply to 203921.82
Date: 1/10/2012 10:07:14 AM
Kitakyushu
ASL
Overall Posts Rated:
12341234
Nope...not from a bot and not a FA...it was a planned purchase..he played him at C and I gave him an extra 200K....after the buy it listed his 48 minutes in the training box on the manage my team page.
So it must be a bug with the BB training page program.

This Post:
00
203921.84 in reply to 203921.83
Date: 1/10/2012 10:09:10 AM
Overall Posts Rated:
228228
Nope...not from a bot and not a FA...it was a planned purchase..he played him at C and I gave him an extra 200K....after the buy it listed his 48 minutes in the training box on the manage my team page.
So it must be a bug with the BB training page program.

This wouldn't surprise me at all... I'll take a look into it as soon as I have some time. Might take a little while, but it'll be easy to fix retroactively, and in the meantime it's probably a small enough number of cases that it's not really skewing things (other than that player's own info, of course).

This Post:
00
203921.85 in reply to 203921.84
Date: 1/10/2012 10:12:45 AM
Kitakyushu
ASL
Overall Posts Rated:
12341234
Thanks for the reply...I will just add the ID 1 position to it in my head...but I have to say....very nice program...

From: Calum
This Post:
00
203921.86 in reply to 203921.85
Date: 1/11/2012 9:48:46 PM
Overall Posts Rated:
394394
Damn! The age drop off is even bigger than I expected!

From: Koperboy

This Post:
00
203921.87 in reply to 203921.86
Date: 1/12/2012 8:23:27 AM
Overall Posts Rated:
952952
I agree, big dropoff. From experience I already knew that 18 and 19 years trains almost the same, 20 has a bit of dropoff and 21 has a significant dropoff. But I didn't know it slows so much after 22...

Advertisement