Ongoing Value of Celtics' Player Contributions

As I've stated consistently, I'm not as much a fan of individual players' statistics as I am of statistics relating to player combinations.  So I developed my own set of stats as follows, based on 83Games data on the 20 Celtics' player combinations that have played the most minutes together.  (It must be noted that the 82Games stats sometimes run a game or two behind).  It's a bit complex, and I can elaborate on the components of the stats if anyone's interested.

• First, I identified those Celtics player combinations that have produced net "plus" scoring figures over the course of the "season" (as depicted by 82Games).  There were only six such combinations, ranging from +2 to +7.  And none of those six combinations had played together more than 15 minutes.  (That reinforces my suspicion that 82Games is a little behind.)

• Next, I counted the number of times a given Celtic has appeared in any of those six "plus" player combinations.  The results:

4 of the 6 player combinations apiece for Bradley, Wallace, Green and Crawford
3 of the 6 player combinations apiece for Sully and Bass
2 of the 6 player combinations apiece for Humphries, Lee and Olynyk
1 of the 6 player combinations apiece for Pressey and Faverani

• Next, in order not to penalize players who had played fewer minutes overall, I calculated the total +/- PER MINUTE for each of the six player combinations.  

• Next, I ranked the player combinations 1 through 6 in terms of PER MINUTE +/- "plus" totals.

• Next, I added up the total "per minute plus" rankings for all combinations in which each given player participated.  But I reversed the numbers so that, for instance, the player combination that ranked number 1 among the six player combinations would be awarded 6 of my "points;" and a player combination that ranked number 6 would be awarded only 1 "point."  That way, the greater the "reversed" point total for a given player combination, the higher that player combination was ranked in my "per minute" calculations.  I had a reason for doing the rankings this way, and I'll explain it if anyone wishes.

The "points" in my final player rankings reflected both the number of "plus" player combinations a player was part of and the "per minute plus" score of each of those six player combinations.  The results, based on my "point" totals:

1. Crawford (17 "points")
2. (tie) Green (14 "points")
2. (tie) Sullinger (14 "points")
4. Bradley (12 "points")
5. Wallace (10 "points")
6. Lee (9 "points")
7. Olynyk (7 "points")
8. (tie) Bass (6 "points")
8. (tie) Humphries (6 "points")
10. (tie) Pressey (3 "points")
11. (tie) Faverani (3 "points")

Just based on how the various players have been performing lately, the one ranking I'd most quarrel with subjectively would be that of Bass (tied for only 8th).  Mathematically, the reason Brandon appeared in as many "plus" combinations (three) as Sully but ranked lower in total number of my rating "points" is that the combinations Brandon was part of had lower "per minute plus" scores (#4, #5 and #6 out of 6 combinations) than the combinations Sully was part of (#1, #2 and #5 out of 6 combinations).  Judgmentally, I can only assume that the 82Games figures are a game or two behind and Brandon's recent contributions would elevate his ranking.  (I bet the same thing would happen in the case of Humphries too.)

I'm going to conduct this exercise every couple of weeks-or-so to track rankings changes over an increasing number of games and any personnel developments with the team.  Just like a baseball player's batting average, these ratings can fluctuate early in the season much more than later.

Go Jordan!  Given the recent reports on Rondo's situation, we'll be needing your best more than ever.  Anyone want to debate the fact that the Celtics MVP so far (to the extent that an MVP award is a really accurate portrayal of anything) has been either Sully or Jordan?  Let's see now.  Did Danny take considerable risk in obtaining both of these two guys?  Go Danny!  Damn the Tankards accurately defined as "large drinking vessels") and full speed ahead!

(Late note: I checked 82Games, and their stats include only the first 16 Celtics games although they've actually played 20.  The rankings of guys like Bass and Humphries have probably been understated accordingly.)

Sam