Author Message
Mike
Joined: 01 Jul 2008
Posts: 2
PostPosted: Wed Jul 02, 2008 12:41 pm Post subject: Alternate Win Score Reply with quote
Followed a link from True Hoop related to Wins Produced. Dan Rosenbaum said he came up with a formula, which he called alternate win score, that had better predictive value than wins produced. What is the formula for alternate win score?
Back to top
View user's profile Send private message Send e-mail
NickS
Joined: 30 Dec 2004
Posts: 384
PostPosted: Wed Jul 02, 2008 7:45 pm Post subject: Reply with quote
I don't have a link, but I believe it was, essentially, Win Scores except with a weight of .7/.3 for DR/OR rather than 1.0/1.0
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3596
Location: Hendersonville, NC
PostPosted: Thu Jul 03, 2008 9:50 am Post subject: Reply with quote
DefReb = .70, and OffReb = .30 ?
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
kjb
Joined: 03 Jan 2005
Posts: 865
Location: Washington, DC
PostPosted: Thu Jul 03, 2008 9:52 am Post subject: Reply with quote
I thought the only change Dan made was on FGA. Dan multiplied missed field goal attempts by .7 to reflect the reality that not every missed shot results in the end of a possession. Recall that Wins Produced subtracts for every field goal attempt regardless of whether they're rebounded by the offense.
Back to top
View user's profile Send private message AIM Address Yahoo Messenger
NickS
Joined: 30 Dec 2004
Posts: 384
PostPosted: Thu Jul 03, 2008 11:25 am Post subject: Reply with quote
Mike G wrote:
DefReb = .70, and OffReb = .30 ?
That was me being sloppy. Reverse those.
Back to top
View user's profile Send private message
Mike
Joined: 01 Jul 2008
Posts: 2
PostPosted: Thu Jul 03, 2008 8:48 pm Post subject: Reply with quote
Thanks for the responses. I guess that little change made the difference, and wins produced overrates defensive rebounds.
Back to top
View user's profile Send private message Send e-mail
anarcholis
Joined: 12 Jun 2007
Posts: 19
PostPosted: Fri Jul 04, 2008 10:33 pm Post subject: Reply with quote
This is an important change. I've been messing around with Wins Produced trying to re-weight things to reflect reality a bit better. The best change I could do was this rebounding change to reflect the logic that the offense generally gets 26% of all rebounds and the defense 74%. Therefore an offensive rebound is worth .74 or a possession, because there was a .26 chance that the offensive team would rebound any shot. Similarly, the defensive rebound is worth .26 of a possession.
At the same time, reflecting this change, a missed field goal would really only count as losing .74 of a possession, since there is a .26 chance that the missed shot will be rebounded by the offense.
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3596
Location: Hendersonville, NC
PostPosted: Sat Jul 05, 2008 7:02 am Post subject: Reply with quote
Missed FG are rebounded by the offense closer to 30% of the time. Missed FT, much less. Together, about 26% of reboundable missed shots become OReb. These numbers, of course, vary by team, by shooter, by type of shot, etc.
However, the likelihood of an event doesn't have any relationship to its worth. A missed FT may be 10 times as likely to be rebounded by the defense, but that doesn't mean the rebound is only worth 1/10 as much to the one team as it is to the other. In both cases (FGx or FTx), the ensuing possession has equal value to either team.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Harold Almonte
Joined: 04 Aug 2006
Posts: 616
PostPosted: Sun Jul 06, 2008 5:31 pm Post subject: Reply with quote
Quote:
However, the likelihood of an event doesn't have any relationship to its worth. A missed FT may be 10 times as likely to go to the defense, but that doesn't mean the rebound is only worth 1/10 as much to the one team as it is to the other. In both cases, possession is equally valuable to either team.
That's true. That likelihood is already implicitlly inside the relationship between the player's total rebounding and the player's rebounding usage, with variation by rebounder's skills of course. But, unlike Scoring, we (neither the boxscore) don't know the player's rebounding usage, just because we don't know how many rebounds a player really conceded to his real opponent when he's on the floor, not to mention how many he could concede to a sidekick teammate. We have an overall aproximation from the usual rebounding rating formula, since we are taking every missed FG, as an everybody's possible rebounding attempt (far from reality, since every position have a different rebounding quota).
The rebounding shortage weighting is arbitrary although it seems logical, as it is the FGA penalization applied to an individual: the change of possession between teams after the FG Made is not the scorer's fault, and the debate about the value of a missedFG at the individual shooter level, when teammates's Off. Rebs. are in the mix, is for the ages.