RE: [APBR_analysis] Re: Dilution, balance, and Bob-bashing
- -----Original Message-----
From: Mike G <msg_53@...> [mailto:msg_53@...]
Sent: Wednesday, January 01, 2003 4:50 AM
Subject: [APBR_analysis] Re: Dilution, balance, and Bob-bashing
--- In APBR_analysis@yahoogroups.com, "Dean Oliver <deano@r...>"
>... I believe MikeG had some stat a long time agoshows
> showing how much net gain in minutes played per player there was,
> which was a weird stat (one I'm not sure about), but certainly
> expansion years and possibly other things. I thinkMessage 1578 (way back on Dec. 17) recaps this study.
> turnovers/possession would show expansion years, too.
I would have thought Minutes Played would be the least-weird stat
one could find. (Everyone knows the talent-rich teams have players
that would be getting more minutes on an average team.)
Yes, I think the minutes played comparisons that MikeG has done are
a potentially fruitful line of inquiry.
The questions I have about the significance of Minutes are:
1) What's the mathematical relationship between an individual's
productivity, and his minutes?
2) How can a decades-long (expansion years aside) trend toward
diminishing minutes be interpreted?
I think these are two of the biggest reasons why MikeG's year-by-year minutes-
played ratios have to be interpreted with caution. I don't know the answer
to question 1, but it surely will be heavily context-dependent, in particular
dependent on the quality of the players' team. Rashard Lewis is the Sonics'
2nd-best player and except for his early season injury would be the Sonics'
2nd-highest minutes-played player, next to The Glove. But (as several Sonics
fans argued when Cuban was aggressively recruiting him, and Rashard was considering
jumping to Dallas), on a team such as the Mavs, he'd be only about their 3rd or 4th
best player, and probably about 3rd or 4th in minutes played (after Nowitzki, Nash,
and maybe Finley). Same player, but very different minutes played.
As for question 2, the caution that I mentioned in commenting on MikeG's post
still stands: the diminishing minutes may be largely due to the (necessary)
omission of players' rookie seasons. Overall, players' minutes decline
throughout their career -- but part of this is because MikeG's stats ignore the
huge (infinite, in percentage terms) increase in their minutes in their rookie
season, compared to their pre-rookie season.
What kind of correlation is there between TO/Poss and W/L record?
I don't know if this is the right question; it can only be answered in
a cross-sectional context, i.e. in a given league in a given season, teams
with lower TO/Poss might (or might not) have better winning percentages.
But that's not the use of TO/Poss that DeanO is proposing; rather he wants
to compare different leagues; or the same league across different seasons.
These various leagues will have different TO/Poss stats: high in grade school,
low in the NBA. But there will be NO correlation between TO/Poss in these
leagues and the leagues' W/L record, because they all average .500.
That is, one can answer the question with respect to individual teams, but
DeanO is proposing to compare leagues, not teams. And leagues always have
W/L records which average out to .500.
> I particularly enjoyed Rosen's article on the Sonics where hewatched
> one game (one of the worst of the year, for what it's worth) andPut me firmly in the "pompous windbag" camp. The best thing about the
> acted like he knew something about the team. Apparently, Desmond
> Mason can't make a jumper because he had one bad night.
Mason comment, for instance, was that Rosen said the reason Mason
can't make a jumper was his "low release point", which was hilarious
on several levels:
1) Apparently he's never watched Steve Kerr. Or Andrew Toney. Or
Bryce Drew. Or about a hundred other guys who shoot from under their
chin but make everything.
2) Mason's release point isn't low, especially given that he's about
20 feet off the ground when he shoots it.
3) Mason's problem isn't the release point, it's the lack of arc on