Win Share Predictions: Celtics Among Elite

Posted by Zach Lowe on Sep 24, 2009 33165 commentshttp%3A%2F%2Fcelticshub.com%2F2009%2F09%2F24%2Fwin-share-predictions-celtics-among-elite%2FWin+Share+Predictions%3A+Celtics+Among+Elite2009-09-24+12%3A47%3A20Zach+Lowehttp%3A%2F%2Fcelticshub.com%2F%3Fp%3D3316

I’m a little late on this, but I wanted to make sure to pass along Basketball Reference’s Win Shares-based projections for next season. Quick explainer: BR’s Win Shares stats act like Bill James’ Win Shares stat in baseball—and a little like VORP, or Value Over Replacement Player, as well—in that they attempt to calculate how many wins an individual player is worth over the course of a season. As explained here and here, Win Shares attempts to factor out the quality of a player’s team, so that players don’t get artificial credit for playing on a good team (as happens with offensive and defensive ratings).

The stat also factors in minutes played in addition to per-minute production. So a role player who is productive in limited minutes doesn’t score as well in WS as he would in John Hollinger’s PER stat. (Hollinger also has a Win Shares formula at ESPN.com.

In any case, here’s the top of the Eastern Conference standings, per BR’s Win Shares:

Obviously, this is no major surprise to us. We consider the Celtics contenders on par with Cleveland, Orlando and the Lakers, and BR’s formula reflects that. But it’s a relief anyway. I have to say—there is a growing perception out there that the Celtics are not really contenders this season. You occasionally see it in non-Boston papers, and I hear about it everywhere I go. I was at a wedding last weekend, and a friend who knows I write about the Celtics pulled me aside and asked me something like: “Ok, Zach. Basketball fan to basketball fan—Are the Celtics really contenders this season?” He doesn’t think so, for a litany of reasons—age, health, a general belief that certain Celtics, especially Glen Davis, are overrated, etc.

I’m surprised how much I hear this kind of thing at social gatherings. A lot of it obviously has to do with the fact that the Big Three are all over 30, and that KG’s knee broke down last season.

The Win Shares stat gives me another argument to use the next time I come across a doubter. Of course, Win Shares cannot calculate the likelihood that Kevin Garnett’s popliteus tendon stays healthy all season. I hope those lazy bums at BR are getting to work on a Win Shares formula that can factor in KG’s popliteus. Come on, guys.

Out West, BR’s Win Shares formula has the Lakers and Utah neck and neck for the top spot, with six teams bunched below them.

Oh—go ahead and ignore the low win totals at the top of the standings and the high-ish ones at the bottom. The formula shoves everything toward the mean, including the number of minutes each player is projected to play. It errs on the ultra-safe side. The best players get credit for fewer minutes than they’ll actually play, and the formula squeezes totals toward the middle. Obviously, BR is well aware that Boston, Cleveland and Orlando will all likely win far more than 50 games.

Below is the Celtics player-by-player breakdown. As you can see, the formula projects Rajon Rondo and Paul Pierce as the most productive players on the team, just ahead of KG and Ray Allen. But Kevin Garnett is really the most productive player of the bunch, because the formula, perhaps considering his recent injury history, gives him just 2,167 minutes, fewer than he has played in any season other than 1999 (the lockout) and last season.