What's new
By:
Filters

The Utah Jazz: the Hilary Clinton of the NBA

I

idiot

Well-Known Member
Basketball-reference.com calculates a Pythagorean Win stat which is expected wins based on points scored and allowed.

Since Quin Snyder took over (2014-5 season), the Jazz "lead" the NBA in number of Pythagorean wins versus actual wins (in other words, according to point-differential margin they should have won more games than they actually won) at 12. Snyder would have a .547 winning percentage as a head coach if the Jazz had been merely average in his actual vs. Pythagorean win luck (rather than his current .481).

The others who lost more than five games more than they should have during this time period:
MIN 8
OKC 6
DET 6

The Donald Trumps of the NBA (more actual wins than expected wins during the same period)?:
MEM 14
GSW 11

Don't know what this means. Maybe that the Jazz continue to play harder throughout the game: keeping losses closer while having more big wins. For what it's worth the pattern was opposite under Ty Corbin: the Jazz won 7 more games during his three full years than they should have been expected to win based on point differential.

Oh, and the Jazz so far this year are close to league bottom again, losing more than expected based on point differential (-2), having better luck than only the T-Wolves (-3). Some of our close likely competitors are among those tied for the top, having won two more games than expected thus far: MEM, POR, SAS, ORL, NYK.
 
And refereeing hack job of yesteryear. It's nice to see that we are getting more respect this year and refs are swallowing their whistles instead of automatically blowing phantom fouls anticipating contact that isn't there.
 
ONE LOVE said:
And lack of a 'closer'.
Click to expand...

Our offense was fine last year in the closing minutes. We lacked defense, rebounding and calls to end games. Injuries and our youthfulness did hurt us last year. This year injuries are hurting us although we still are losing more games than we should this year based on the formula in the OP and our difference in points per game. But I blame that on injuries, rough schedule getting this team down and not playing as hard early in a few games that we let get away early and had to put too much effort to get back into.
 
The Jazz Wonderbra said:
And refereeing hack job of yesteryear. It's nice to see that we are getting more respect this year and refs are swallowing their whistles instead of automatically blowing phantom fouls anticipating contact that isn't there.
Click to expand...

Has anyone seen a compilation of the 2 min report and where we rank this year so far?
 
Rather than bore you with the details, I'll just report that I made a quick perusal of whether this Pythagorean vs. actual wins seems to correspond with several other stats that the Jazz are known to have fared "poorly" at recently:

  • experience (average age)
  • unfavorable calls (based on Andy Larsen's last two minute research)
  • clutch play (last-five minutes)
  • games lost to injuries

The short answer is that these factors do not seem to correlate much. For every team that looks like a good correlation on these measures there are two or three that are not. So I think small correlations exist on these generally, but the strength of those correlations is not strong. (Experience and clutch play seem to have the closest thing to a solid, if modest, correlation to the Pythagorean vs. actual wins; games lost to injury looks like it has the least.)

For the Jazz to fare so poorly on all of these measures recently is rather remarkable. So if you think these "expected wins" are a significant measure of team quality, it's hard to conclude anything except that this is one more measure showing that the Jazz have been especially snakebitten during the Snyder era.
 
Last edited:
Ended up running some correlation coefficients for the NBA season in 2015-6 . Clutch-time performance and experience did indeed have some effect on the difference between Pythagorean and actual wins; injuries and last-two minute refereeing had very little impact and are probably not related:

Clutch-time differential: r=.652 (so r-squared, which is theoretically the percentage of variance in one variable that explains the variance in another = 42.5%)

Age: r=.534; r-squared=29.0%

Last-two minute report: r=.286; r-squared=8.2%

Injuries: r=-.222 (which means that the more injuries teams had, the better they did in getting actual wins compared to Pythagorean wins; though at a very low correlation level, so it's probably just a completely random relationship); r-squared=4.9%
 
You must log in or register to reply here.

Share this page

Top