Saturday, August 2, 2008

Extremistan



It was Q over at Rethinking Basketball who introduced me to the work of The Arbitarian, who is one of a small but dedicated network of NBA (and by extension, WNBA) stat grinders.

He introduced something called the "playing style spectrum". The formulas are at the link above. They are simple formulas -- nothing more than simple division -- but the explanation is a bit long.

What's the idea? The idea is to assign three numbers to every player....

Scorer index: Determines how much a player specializes in points and field goal attempts
Perimeter index: Determines how much a player specializes in assists and steals
Interior index: Determines how much a player specializes in blocked shots and rebounds

I then applied the score to all players with more than 10 minutes per game playing time in the WNBA. Let's take a look at the players on the Dream. An important note is that a high number for a player does not mean that they are the "best in the WNBA", but rather that a high portion of their individual game rests in one of the three areas above.

Betty Lennox (0.91, 0.45, 0.2)

It's what we think. A high degree of Lennox's value comes from scoring, but the middle number is good - she can also handle the ball.

Ivory Latta (0.78, 0.84, 0.03)

Latta is not just a shooter, but a point guard who can work the perimeter and handle the ball. Her interior game, however, is nonexistent...but we don't expect Ivory to get rebounds.

Tamera Young (0.59, 0.4, 0.59)

Young is the type of player who can do several things well in the time she gets.

Iziane Castro Marques (0.99, 0.46, 0.09)

Izi's game is more heavily related to scoring even than Betty Lennox's is. Like Betty, she's also good at handling the ball and lousy at the boards.

Jennifer Lacy (0.79, 0.21, 0.50)

For someone who is a forward-center, Jennifer seems to concentrate on scoring points more. My guess is that either a) Jennifer has been forced to rely more on shooting due to the need to come back from losses, or b) the presence of other shot blockers has allowed Lacy to be more agressive with scoring. Then again, her numbers might have always been like this.

Katie Feenstra (0.22, 0.01, 0.98)

Not much of a scorer, and as that 0.01 indicates, a lousy ball handler with only 10 assists and 3 steals all year. Her game is a big woman's game. But does that 0.98 mean that her game depends more on blocked shots and rebounds than anyone else on the Atlanta Dream?

Kasha Terry (0.31, 0.16, 0.88)

A bit more of a scorer than Kit is.

Alison Bales (0.05, 0.11, 0.99)

Holy moley. Bales personal game comes almost entirely from her ability to block shots and get rebounds.

Erika DeSouza (0.13, 0.04, 0.97)

Again, another pure shot blocker and rebounder.

(* * *)

So what does this mean? It means we have a lot of players on the Dream that have "extreme" games. Lennox and Castro Marques are extreme scorers. You give them the ball, and they're going to take a shot with it, offense be damned. They like to create their own offenses. How good they are at doing that is a matter of debate, given Castro Marques's streakiness and Lennox's tendency to turn over the ball. If you have a pure scorer, you want someone who can hit what she's aiming at and not turn over the ball.

Let's look at Cappie Pondexter's triple of (0.96, 0.6, 0.06). She's sort of like Betty Lennox and Izi Castro Marques. Cappie is also a "shoot first" kind of player. However, she hits 41 percent of her shots, compared to the 35 percent that Izi hits. She's also only turned the ball over 62 times, compared to Betty Lennox's 100.

Let's also look at Feenstra, Terry, Bales and DeSouza. They rebound. They block shots. But for the most part, that's all they do, particularly Alison Bales. Granted, Bales does it very well - in Wins Score, Bales has the highest value of any player on the Dream, even Ivory Latta or Betty Lennox! However, aside from Terry, Bales, Feenstra and DeSouza are all one-dimensional. On the offensive end, they can set screens or get the offensive rebound, but other teams know that Bales or Feenstra won't be given the ball to score (heck, if Izi or Lennox is on the floor, it's a much better idea to double team the lead shooters). Compare their triples with Lisa Leslie's (0.33, 0.26, 0.82). Leslie's game depends a lot on getting those rebounds and shot blocking, but not to the extent of the big women on the dream. Defenses still have to worry that Leslie will score with the ball.

So why is this post called "Extremistan"? If you look at other teams than the Dream, players don't seem to be clustered up at one end or the other of the spectrum as much as the players on the Dream. The Dream is a bit more predictable in their individual games than players from other teams. That makes it easy to build defenses around what the Dream can do offensively, and vice versa.

1 comment:

Q said...

Great analysis of the team and an interesting point about them being "clustered".

It would be interesting to use these same numbers with versatility rating to see if there are trends across different teams.

And for those interested, the player styles graphic for the WNBA can be found here: http://gmapuploader.com/iframe/sMe3egnJXB