Riley Scouts Eric Gordon?

Posted on March 9, 2008 by


Pat Riley – head coach of the Miami Heat – was watching basketball on Sunday.  This is not unusual for an NBA coach.  What is unusual is that he was personally watching a college game.  Apparently Riley has decided that coaching an NBA team is not enough of a challenge, so he has taken it upon himself to go scout players for the upcoming NBA draft.

Although we can laugh about this course of action, it does make some sense.  The Heat are finishing a disastrous season.  And at this point, Riley coaching the team to more wins (not sure that’s actually possible, but that’s a story for another day) is not going to help.  Miami is currently in the hot pursuit of lottery balls, and if the lottery gods smile upon this team, the Heat will land a top choice in the draft.  Such a choice could produce many wins, both next season and for years in the future.

Obviously the more balls, the better the chance at a great pick.  And equally obvious, getting the great pick is only half the battle.  Once you secure the choice, you need to pick the right player.

Scouting Eric Gordon

Hence we come to the truly odd part about yesterday.  Riley was watching the Indiana Hoosiers lose to Penn State.  The only top NBA prospect in Sunday’s game was Eric Gordon.  Gordon plays shooting guard, the same position as Dwyane Wade.  Wade is the only truly great player the Heat are sure to bring back next season.  So it seems odd that Riley would be interested in Gordon.

And that’s not the only odd part of this story.  The truly strange twist in this tale is the play of Eric Gordon and how this was covered in the media. On Sunday the Indiana Hoosiers lost to Penn State.  The Associated Press article included the following quote in the recap of the game:

Gordon nearly single-handedly got the Hoosiers back in the game after Indiana went the first 5:29 of the second half without a field goal to trail 36-30. A 3-point play and three 3-pointers by Gordon during a 14-0 run put Indiana ahead 46-38 with 11 minutes left in regulation.

What’s missing from this take on the game is what happened outside this 14-0 run.  For the game Gordon only hit 8 of 24 shots.  So outside this run, he was only successful on 5 of 20 shots from the field.  Plus he committed five turnovers.  His Win Score – which takes into account all the stats in the box score – was -5.0 in the game.  In other words, one could argue that Gordon single-handedly took the Hoosiers out of this contest.

To see this argument, imagine if this was a football game and Gordon was the quarterback.  In one drive, Gordon manages to move the team the length of the field completing all of his passes.  But if Gordon was throwing interceptions and incomplete passes before and after that drive, the story of the game would be that the quarterback cost the team the victory.

In basketball, though, such analysis is rarely seen.  Gordon was the leading scorer for the Hoosiers.  Therefore, he must have been a positive for his team.

29 Games of Gordon

Gordon played 29 games before facing Penn State on Sunday.  When we look at Gordon across these games, we do see that his woeful shooting on Sunday was an anomaly.  Gordon’s adjusted field goal percentage is 54.2%, which is quite good.

Let me put Gordon’s shooting in perspective. The other top shooting guard who is expected to be available in the NBA draft is O.J. Mayo of USC.  Mayo’s adjusted field goal percentage is 53.0%, and again, that’s quite good.  Despite this performance, though, here is how Chad Ford describes Mayo at

It’s clear Mayo’s going to be a great NBA scorer … but will he give a team anything else?  

When we compare Gordon and Mayo we see that the latter produces more – on a per-minute basis – rebounds, assists, and steals.  Mayo also commits fewer turnovers.  In sum, if we are concerned that Mayo isn’t going to give anything beyond scoring, shouldn’t we have the same concern about Gordon? And yet, in the mock drafts I have seen, Gordon is consistently rated as the better prospect (and Ford describes Gordon as a “do-it-all combo guard that is already lighting up the box score”)

The Win Score story

When we turn to Win Score – a metric that can tell us something about lighting up a box score — we see little reason to be optimistic about Gordon or Mayo.  Entering Sunday’s action, Gordon had posted a 6.7 Win Score per 40 minutes (WS40) played.  Mayo has a mark of 6.0.  When we look at all shooting guards drafted out of college between 1991 and 2006 (and who played at least some significant minutes in the NBA), we see an average WS40 of 8.7 the last year the player played college basketball.  In other words, Gordon and Mayo are far below average.

To put “far below average” in perspective, here are the shooting guards with a WS40 below 7.0 (again from 1991 to 2006): Kareem Rush, Kirk Snyder, Toby Bailey, Tony Dumas, Lawrence Moten, Courtney Alexander, Ed Gray, DeJuan Wheat, Chris Smith, Jimmy King, Felipe Lopez, Larry Hughes, and Jamal Crawford.

Now that is not an impressive list of NBA talent.  I didn’t bother to look, but I am guessing that none of these players have a career Win Score that’s above average.  And I know some of these players are very far below average.

When we look over this list, one has to wonder why Riley was scouting Eric Gordon. Miami already has Dwyane Wade (college WS40 = 11.9).  And it doesn’t look like Gordon -despite his ability to score in college – is going to help any NBA team win many NBA games next season. 

And yes, the same story can be told about O.J. Mayo.  Given what each of these players are doing in college, it seems likely that any team drafting these players in the lottery in 2008 will be printing a ticket to the lottery in 2009.  In other words, if it’s Miami that takes Gordon or Mayo, Riley might just have to plan on scouting college talent next spring as well. 


Our research on the NBA was summarized HERE.

The Technical Notes at provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.