RealGM

I've spent the last few hours creating All-Defensive Teams for 5-year periods of time based on their defensive win shares (if you're unsure what defensive win shares are, then please read the article).

The results are interesting.

Check out which players dominated defensively during each era of NBA basketball.

http://basketballjournalist.blogspot.co ... e-win.html

Interesting. Its tough though to do super long periods of time. It would be interesting to see the best 5 winshare seasos of top players. I think the teams are pretty good just some odd stuff for some 5 year periods

I quite liked using 5-year periods. It gives a good reflection of who was dominating defensively during each era.

There are one or two oddities amongst the results but, on the whole, they're about what I expected.

As I said in WillC's excellent thread about this on the PC Board, it does show how much defensive win shares depend on teams.

Look at the period from 55-65 and you will find that in addition to Russell, over half the other players are Celtics with such notoriously weak defenders (according to his own coach/GM) Bob Cousy having very high numbers. Going forward, you see Greg Ballard (good but not very good) and Truck Robinson (not very good) on my own Bullets teams showing up, and for that matter Allen Iverson and Paul Westphal, neither of whom were good individual defenders.

Interesting but take it with a grain of salt.

It is interesting, but I also think it demonstrates why win shares aren't considered the be-all in terms of player analysis.

Sure, on average a player with a lot of defensive win shares will be better defensively than one with fewer, but its possible to be a great defender without being elite in that stat and also possible to be a lousy defender with great defensive win share stats.

Yeah I think you're right, Snakebites. The results are fairly accurate but it just goes to show that most elite players happen to be good defensive players (which makes sense, since they wouldn't be 'elite' otherwise).

What stat is the best to measure defensive ability? Adjusted +/-?

any time larry bird/john stockton is on a def team, you know something is broken

Larry Bird might not have been a great individual (man-to-man) defender, but he was a tremendous team defender.

Meanwhile, John Stockton has the most steals in NBA history....

WillC wrote:Larry Bird might not have been a great individual (man-to-man) defender, but he was a tremendous team defender.

Meanwhile, John Stockton has the most steals in NBA history....

Bird couldnt defend lets leave it at that. If you think steals= good def then go cry in a corner

primecougar wrote:Bird couldnt defend lets leave it at that. If you think steals= good def then go cry in a corner

The most important aspect of defense is team defense. That means making the right decisions, not getting out of place, and knowing how to guide the guy you're guarding into the help. Add to that being able to anticipate passing lanes, and defensive rebounds well, and that's a lot of good.

While it's certainly true that to be an elite defender you must also be agile enough to be a strong "shut down" kind of man defender, this notion that that constitutes all defense is wrongheaded. There are quite a few extremely athletic guys in the league who are poor defender simply because they don't understand how to function on defense within the complexities of the team game at the NBA level.

primecougar wrote:

WillC wrote:Larry Bird might not have been a great individual (man-to-man) defender, but he was a tremendous team defender.

Meanwhile, John Stockton has the most steals in NBA history....

Bird couldnt defend lets leave it at that. If you think steals= good def then go cry in a corner

You're right, steals aren't everything.

However, Larry Bird was certainly a great team defender. If you can't accept that (and want to start being childish about it), then that's your problem.