Westbrook defensive lapse analysis

[Crow]

This is a suggestion for anyone looking for a potentially high profile project:

Look at the video and video generated data at the plays where Russell Westbrook quits or fails to offer a vigorous / acceptable defensive effort. Or someone else if they offend as much. Are there trends of when it happens or happens more often? Is it fatigue late in stints or late in games? More common after offensive usage on last play or after usage of x frequency in last 5-10 plays on offense? after multiple drives? Made or missed shots? Fouls not called? Ahead or behind? Against bad, mediocre and good teams? When with starting lineup, mostly starter lineups or mostly bench? After previous win or loss patterns? After being attacked on defense at a certain frequency? Against vets, rookie, stars, role players? Faster guys, bigger guys, guys with driving ability? There is so much that could be checked beyond acknowledging that it happens and being bugged by it or accepting it. Perhaps knowing the tendencies you could try to get in front of or minimize the occurrences. I hear there is some Vantage Sports data on pick n roll defense (after a successful pick?) that might suggest it is not as big an issue as it may appear. Is his behavior, because of the defensive often around him, actually a trap? Anyone want to do it? I'd like to but can't see getting to the full load of it any time soon.

[bondom343]

I don't think I would, but he's interesting. Every defensive metric likes him, yet he's got a rep as a bad defender, which I can't understand. He's a slight plus in last RAPM I saw this season from JE, good in BPM, RPM, on/off, etc. It seems to me he's doing something at a team level thats making up for the lapses.

[bchaikin]

Every defensive metric likes him...

the player tracking data at http://www.nba.com shows on westbrook's defensive dashboard FG%s allowed of 51% (119/234) on 2s and 41% (46/111) on 3s...

compare the same data for stephen curry FG%s allowed of 46% (137/296) on 2s and 28% (48/173) on 3s...

[Crow]

Steals are so valuable. Riding alongside a guy or getting a hand up on a mid-range jumper far less. Perhaps gambling is sensible though alarming and quitting not that bad as long as the guy doesn't get all the way to an open rim very often.

[bondom343]

Every defensive metric likes him...

the player tracking data at http://www.nba.com shows on westbrook's defensive dashboard FG%s allowed of 51% (119/234) on 2s and 41% (46/111) on 3s...

compare the same data for stephen curry FG%s allowed of 46% (137/296) on 2s and 28% (48/173) on 3s...

True, one doesn't. But its the one I trust least by far (the Vantage article convinced me, and there are too many outliers).

By the same metric, Jimmy Butler is a poor defender. I should say all team metrics.

[Mike G]

Don't you have to look at the player FG% against the rest of the league, and compare to FG% vs the defender you're analyzing?
Normally your best outside defender will cover the opponent's best outside scorer, yes?

OKC opponents' 3fg% is .351, exactly league avg.
Opp 2fg% is .462, behind only the Bulls and Spurs, .024 better than avg.

Somehow, their Opp 3fg/fg ratio is 7th lowest -- .244 vs NBA avg .282
Do they have the strategy to chase shooters off the perimeter, and toward their interior defenders?

[Crow]

"Do they have the strategy to chase shooters off the perimeter, and toward their interior defenders?"

Yes, Donovan has mentioned that as a priority, including yesterday.