You Are Here > Basketball-Reference.com > BBR Blog > NBA and College Basketball Analysis

SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all Basketball-Reference content, so you can quickly and easily find the content you want.

Also, our existing Basketball-Reference blog rss feed will be redirected to the new site's feed.

Basketball-Reference.com // Sports Reference

For more from Neil, check out his new work at BasketballProspectus.com.

When Ray Allen Plays 30 Minutes…

Posted by Neil Paine on June 4, 2010

After last night's Laker rout, in which a foul-troubled Ray Allen only took the floor for 27 minutes, I wondered how important it was for the Celtics to have Allen play a substantial amount of time. Because while he may be the 3rd-best player on the team (4th-best if Garnett is anything like his vintage self), his ability to knock down huge threes has been indispensable to Boston over the past 3 years, especially in these playoffs.

To make a quick-and-dirty test on how important it was for Allen to play 30 or more minutes in a game, I created a sort of bastardized version of Adjusted Plus-Minus (using RS + playoff data in the "Big Three" era -- 2008-10) that predicts a team's homecourt-adjusted efficiency differential in every game based on whether or not any given player played at least 30 minutes or not in the game (a 48 minute game, that is -- the threshold was proportionally higher for OT games).

Controlling for the player's teammates and opponents in the game, we can get a rough estimate of what the impact in efficiency differential is on a given player playing 30 or more MPG. Here are the results for all 450 players who had a 30+ minute game in the past 3 years (except Lorenzen Wright, who had 1 such game and was dropped due to singularity issues).

Obviously, the sample sizes are small and the standard errors are huge there, but if we limit the list just to guys who had 50 or more 30+ minute games (the players with the smallest standard errors), look who leads the way:

Player 30mp Games Estimate Std. Error t value Pr(>|t|)
Ray Allen 254 9.112618 2.003239 4.549 0.000006
LeBron James 263 7.691158 1.848023 4.162 0.000032
Carlos Delfino 62 7.571865 2.356393 3.213 0.001324
Rashard Lewis 244 7.338562 2.009081 3.653 0.000263
Andrew Bogut 123 4.708467 2.043482 2.304 0.021273
Michael Beasley 58 4.407164 2.182843 2.019 0.043563
Chris Bosh 199 4.003952 1.876923 2.133 0.032974
Chauncey Billups 218 3.993828 1.698698 2.351 0.018773
Stephen Jackson 189 3.945586 1.575134 2.505 0.012293
Paul Millsap 82 3.791507 1.865556 2.032 0.042191
...
Antawn Jamison 222 -4.250158 1.752314 -2.425 0.015340
Nate Robinson 75 -4.620498 1.914224 -2.414 0.015840
Martell Webster 62 -4.805516 2.122498 -2.264 0.023630
Jeff Green 179 -4.893888 2.216937 -2.207 0.027344
Ricky Davis 67 -5.025963 2.749007 -1.828 0.067593
Michael Redd 95 -5.326994 2.067929 -2.576 0.010036
Eric Gordon 118 -5.680208 1.947040 -2.917 0.003553
Brook Lopez 119 -5.770483 2.177815 -2.650 0.008094
Kendrick Perkins 93 -5.783729 1.933246 -2.992 0.002794
Brandon Jennings 59 -6.221064 2.759197 -2.255 0.024216

(View the full table here.)
Obviously, Allen's foul trouble was not the Celtics' only problem in last night's game -- on defense, L.A. was getting to the rim with impunity, and there's not much Allen could have done to stop it that his replacements weren't already doing. But he could have spaced the floor better on offense and opened up room for Boston to attack the basket instead of hoisting contested Js. Either way, we'll see what kind of adjustments the coaches make before Sunday's Game 2, and I'm sure keeping Allen out of foul trouble will be high on Doc Rivers' list.

ShareThis

9 Responses to “When Ray Allen Plays 30 Minutes…”

  1. AYC Says:

    Great post; the refs called last night's game way too tight. I'm not saying the Celts would have won, but they didn't have a chance with Ray Ray stuck on the bench

  2. Jason J Says:

    I think that's a major flaw in the offense right now. Ray Allen and Pierce are the only big minute floor spacers on the team, and Pierce is the best one on one scorer, hence he's the one who needs the space. Obviously they've found a way to win, but against a huge front line and disciplined defense, it's tough.

  3. Dwight Howard Says:

    My friend always rolls his eyes when I show him how confusingly high Ray's APM is.

    Looking at general trends in APM, non-volume floor spacers (like Ray, Rashard Lewis a year ago) are generally really underrated.

  4. DSMok1 Says:

    Yeah. Ray Allen shows up as a clear top 30 player in the league.

  5. zebulon Says:

    But ray is traditionally characterized as the fourth best player on the celtics - after Rondo, Pierce, and KG. His skills (making the simple percentage play every time, shooting an eFG% of at least 50% from everywhere on the floor) are hugely underrated.

  6. Mike G Says:

    The question -- "how important it was for the Celtics to have Allen play a substantial amount of time" -- should perhaps be asked as:

    "How much time should Allen get when he's this unproductive?"

    According to DSMok1 in this thread:
    http://sonicscentral.com/apbrmetrics/viewtopic.php?t=2586
    -- Ray had SPM of -8.59 in game 1 .

    If a 'replacement level' player has SPM -3 (or thereabouts) it seems the Celts would have been better served if he had not played at all.
    Giving him a few more minutes, would he have become very much better?

    A good coach can see when a guy isn't doing well, and he goes to the bench. Fouls aren't the only reason.

  7. DSMok1 Says:

    I don't know that it's possible to determine in a given game if a player is doing poorly that he will continue to do poorly--you'd think that the weight of years of production is greater than a bad first quarter in terms of choosing whether to play a player (unless there is an injury/illness or something like that). That's an interesting quandary for the coach.

  8. Gabe Says:

    I would be curious what the correlation is between these numbers and the source overall APM numbers.

  9. AYC Says:

    According to Simmons, Allen had 5 fouls in 14 minutes. Hard to get a rhythm going when you're getting whistled every other minute; it's also hard to play agressively in the 13 remaining minutes when your next foul results in disqualification. The Celts wouldn't have been better off if he hadn't played, they would've been better off if the refs hadn't blown the whistle every time he breathed on Kobe