BuzzerBeater Forums

BB Global (English) > Accuracy of Box Score

Accuracy of Box Score

Set priority
Show messages by
This Post:
11
243936.5 in reply to 243936.1
Date: 6/16/2013 8:17:00 PM
Overall Posts Rated:
3131
Thanks for taking the time to research this.

The thing about using the box scores is that we should expect them to be based on the underlying skills, but still random. Players have good games, average games, bad games... and if we're tracking one game in a whole season, the only way we know if it was good, bad, or normal is by comparing to other games that player has played. We can't do that, so we're stuck with a bit of guessing.

Altogether, the box scores should provide reasonable indications of each prospect's abilities. However, every so often, we should expect one draftee or another will have something in their box score that is completely "off", simply because they had that kind of game.

This Post:
00
243936.6 in reply to 243936.1
Date: 6/17/2013 7:50:31 PM
Overall Posts Rated:
164164
Am I the only one shocked that a Center with atrocious JS hit a 3-pointer while we have guys in BB that attempt 1000 FTs and never make one due to atrocious FT skill?


This Post:
00
243936.7 in reply to 243936.6
Date: 6/17/2013 8:17:53 PM
Overall Posts Rated:
774774
C with a 3: 0.1-0.9 sublevel of JS guaranteed

any player with 0 FT: 0.0 sublevel of FT guaranteed

even atrocious guys can hit FT with 0.1-0.9 sublevel of it (usually under 40% FT)

Last edited by Kumiko (CAN U21) at 6/17/2013 10:16:19 PM

If you remember me, then I don't care if everyone else forgets.
This Post:
00
243936.8 in reply to 243936.1
Date: 6/27/2013 5:47:11 AM
Woodbridge Wreckers
DBA Pro A
Overall Posts Rated:
13911391
It's statistics, the sample size is just way too small to draw any conclusions. You can check in hindsight and things might check out partially, but that doesn't mean you can use it to predict anything. If on average 50% of the skills check out with the box score, it's still a complete crapshoot because you don't know which 50% check out beforehand.

This Post:
00
243936.9 in reply to 243936.8
Date: 6/27/2013 3:41:09 PM
Overall Posts Rated:
4848
As I said in my original post, not everything is going to be completely accurate, but it can be an okay way to predict skills. It definitely isn't perfect, but it can be a good indication.

This Post:
00
243936.10 in reply to 243936.9
Date: 6/27/2013 5:59:32 PM
Woodbridge Wreckers
DBA Pro A
Overall Posts Rated:
13911391
Alright, next time try to actually predict what skills your draftee will have. That's a lot harder than to look at it afterwards and say "this is actually pretty close". I'm curious how close you can guess the skills with the box score ;)

This Post:
00
243936.11 in reply to 243936.10
Date: 6/27/2013 6:03:41 PM
Overall Posts Rated:
4848
I'll accept your challenge, for this season's draft, I'll try to predict my draftees' starting skills. However, just as I said before, it isn't going to be 100% accurate, however, I bet I can get pretty close. Now, all we have to do is wait over a month for the All Star Break ;)

Last edited by #ThugLife at 6/27/2013 6:05:42 PM

This Post:
00
243936.12 in reply to 243936.11
Date: 6/27/2013 6:06:33 PM
Woodbridge Wreckers
DBA Pro A
Overall Posts Rated:
13911391
Of course, if you want you can set a more specific goal. Maybe something like at least 1 skill correctly guessed within 1 level, at least 5 skills within 2 levels and 4 skills within 3 or something. Then you can say wether you succeeded or failed the challenge afterwards. Of course you can set the bar higher if you're feeling confident ;)

This Post:
00
243936.13 in reply to 243936.12
Date: 6/27/2013 6:16:57 PM
Overall Posts Rated:
4848
Well, I'll be trying to draft guards this season, since I already have a few guard trainees, so I'll tell you what. I bet I can guess at least 4 out of 6 guard skills within one level.

This Post:
00
243936.14 in reply to 243936.13
Date: 6/27/2013 7:20:55 PM
Woodbridge Wreckers
DBA Pro A
Overall Posts Rated:
13911391
Alright put it down and let us know before the draft what you're guessing, and show the results afterwards! Nice challenge.

This Post:
00
243936.15 in reply to 243936.14
Date: 6/29/2013 12:46:33 AM
Overall Posts Rated:
3434
I'm going to try this as well I think, I think if more people try it and we can get more data we might be able to correlate the box scores to better predict the prospects we are potentially getting.

Advertisement