Some questions about Murphy

by Ian Levy on Feb 19, 2010 10:13 AM EST

Now that the hollow excitement of the trade deadline has passed, I am left with some questions about our roster, specifically Troy Murphy. I obviously don't think his production matches the huge amount of money we will be paying him the rest of this year, and probably most of next year. I guess I am a little confused though about what his actual value is. I admittedly have missed a lot of games this year and am still learning about some of the advanced statistical systems, so that may factor into my confusion.

(I posted this questions on a few other more general nba and statistical blogs but nobody seemed interested enough to tackle it. I wrote it up about 2 weeks ago so some of the numbers may have changed a little. I would love some thoughts from some more advanced basketball minds than my own.)

     Everytime I hear discussions about Troy I get more and more confused. His traditional box score statistics seem respectable, not as great as last year, but respectable. From the data I have looked at on 82games.com it would appear that he is absolutely KILLING the pacers. His simple rating is a -5.8, only Ford and Solomon Jones are worse. His Net Plus/Minus is a -291, by far the worst on the team. Even per 48 min his plus/minus is a -11.5, and again only Ford and Solomon Jones are worse. He is also featured in 4 of the 5 worst 5-man units that have played over 25 minutes together this year.
       However, if you look at some of the other advanced statistics they tell a completely different story. According to Hoopdata.com he has a PER of 18.38, the best on the team and above average compared to the rest of the league. According to Dave Berri at Wages of Wins, he has a WP48 of .315. This is the highest on the team, ranks significantly above average, and would indicate that on a per minute basis he is producing more than Josh Smith (.272), Dirk Nowitzki, (.150), Carmelo Anthony (.160), Chauncey Billups (.214) and several other All-Stars.
      I understand that each system emphasizes certain statistical factors, but the fact that they paint such polar opposites pictures of his performance is really confusing. I have seen instances where these systems disagree about the relative effectiveness of a certain player (One system says a player is good, another says he is great). I haven't run across a case before where one player looks like he is carrying his team according to one system, and absolutely killing his team according to another system.

I have read all the opinions about how he affects Hibbert's production when they play together, but I guess I am asking about his performance on a larger scale. Is he really hurting the team more than he is helping us? If we removed him from the team maybe Hibbert's production would increase, but what about the loss of Murphy's production? Would that slack be picked up by Hibbert and others? When this contract expires he is going to have to take a huge a paycut. If he really is producing as much as his PER and WinScore indicate, should we bring him back at a much more reasonable price?

Any thoughts?

Tweet Comment 5 comments | 0 recs |

Do you like this story?

Comments

wow, a lot of numbers there.....

you will never have good team numbers though if you play absolutely no defense and do not help others score on offense.

If Youre not First, Youre Last

by drsuessrunner07 on Feb 19, 2010 1:09 PM EST reply actions

I hate these advanced stats in basketball, theyre getting too much.

by captain flitzy on Feb 19, 2010 2:15 PM EST up reply actions

Clarification

I Think I understand the systems enough to get why they have different opinions of Murphy. I think the crux of my question is “What am I supposed to take from this information?”

These advanced statisitical systems seemed designed to deepen our understanding of what has happened on the basketball court and make predictions for what will happen in the future.

Are the plus/minus numbers really saying that the Pacers would be winning more games if Hansbrough or someone else were taking Murphy’s minute? I have to admit this line of reasoning doesn’t totally hold up to me.

Is the problem that he doesn’t work well with his teammates, or that they don’t work well with him?

by Ian Levy on Feb 19, 2010 3:04 PM EST reply actions

The Differences

I believe the +/- numbers are a scoring-based, team statistic. It shows how a team excels (scoring) when a certain player is on the floor.

A PER (Player Efficiency Rating) is more of an individual statistic based upon a player’s per-minute statistical production. The league average is always 15.00. So, to have a high PER means you put up a bunch of statistics while you’re on the court. Most teams’ best player has their highest PER. According to John Hollinger, the creator of the PER statistical measurement, Danny Granger has the highest PER for any Pacers player at 18.36. Murphy isn’t far behind at 17.56. Hibbert is third at 16.35.

What’s interesting to me is another advanced statistic called usage rate – the average number of possessions a player uses per 40 minutes. This would indicate which player is involved most in a team’s offense. The player with the highest USG number in the NBA is Dwayne Wade at 33.4. Murphy’s USG number is 16.9 which puts him ranked at 187 in the league, and 9th on the team behind Granger, Hansbrough, Hibbert, Price, Ford, Dunleavy, Head, and D. Jones.

A player like Murphy who has a high PER, low USG, and low +/- number tells me two things: his skills are not being used often enough, and when he is on the floor the team’s point production goes down. These three ratings equate to a negative value. He’s efficient at what he does, but his skill set doesn’t fit into what the Pacers do offensively.

The FAULK

by incredibleFAULK on Feb 20, 2010 9:57 AM EST reply actions