Factors Determining Syracuse Wins

[EastCoast2]

I thought some of you would find this interesting. Attached is a short (4 page) excerpt from a paper I'm writing for a professional credential. I chose to examine Syracuse's win percentage as a function of several variables. The strongest correlation I have found so far, which shouldn't be a surprise to many here, is the number of future NBA draft picks on the roster. Spoiler alert to my fellow math nerds, the R-squared value (RegSS/TSS) of this regression is 0.81.

 

[jncuse]

I think you should consider adding turnovers. On our really good teams in recent years, the guards all had a high level of steals. I think turnovers created might be the best measure of the effectiveness of our zone.

 

[HoustonCuse]

Interesting but I wonder if these are predictive or not. In terms of NBA draft picks for example: when our team is good our players tend to get drafted. Likewise, our better teams have better stats in a lot of areas but you can't necessarily predict that rebounding, for example, will cause more wins in the future because we had successful teams with good rebounding.

 

[stuckinbig11]

Other than draft picks, a lot of noise, but correlation does not mean causation.

 

[EastCoast2]

Interesting but I wonder if these are predictive or not. In terms of NBA draft picks for example: when our team is good our players tend to get drafted. Likewise, our better teams have better stats in a lot of areas but you can't necessarily predict that rebounding, for example, will cause more wins in the future because we had successful teams with good rebounding.

Tough to say, there's a lot of interdependence of course, but our players get drafted even when our team isn't good. Last year, our worst season in ~20 years, McCullough and Christmas got drafted. Our best year was 2011-2012 when we went 34-3 with 5 draftees (Melo, Waiters, MCW, KJo, and Christmas). 2007-2008 wasn't a great year and we had two first rounders in Flynn and Greene.

 

[PoppyHart]

Seems strange to suggest that we were better b/c of Freshmen Xmas, Melo, MCW, etc... who really had very little impact in those years.

 

[EastCoast2]

Seems strange to suggest that we were better b/c of Freshmen Xmas, Melo, MCW, etc... who really had very little impact in those years.

2011-2012 had sophomores Fab and Waiters, who were both studs, and KJo who was our leading scorer.

 

[Scotch]

I think you should consider adding turnovers. On our really good teams in recent years, the guards all had a high level of steals. I think turnovers created might be the best measure of the effectiveness of our zone.

Agreed. Plus they create transition opportunities, which is typically a really efficient way to score.

 

[General20]

I thought some of you would find this interesting. Attached is a short (4 page) excerpt from a paper I'm writing for a professional credential. I chose to examine Syracuse's win percentage as a function of several variables. The strongest correlation I have found so far, which shouldn't be a surprise to many here, is the number of future NBA draft picks on the roster. Spoiler alert to my fellow math nerds, the R-squared value (RegSS/TSS) of this regression is 0.81.

Now the real question is this, does Syracuse get more players drafted when they win more or does having more future NBA players make them win more?

I think its the former.

There is a distinction between a guaranteed NBA player like Carmelo Anthony, and a fringe NBA player like Andy Rautins who gets drafted but is not measurably better than a hundred other guys who could have been drafted instead but weren't. In Andy Rautin's case, I think winning a lot in college absolutely made the difference between being drafted and not being drafted.

 

[Knicks411]

Now the real question is this, does Syracuse get more players drafted when they win more or does having more future NBA players make them win more?

I think its the former.

There is a distinction between a guaranteed NBA player like Carmelo Anthony, and a fringe NBA player like Andy Rautins who gets drafted but is not measurably better than a hundred other guys who could have been drafted instead but weren't. In Andy Rautin's case, I think winning a lot in college absolutely made the difference between being drafted and not being drafted.

Right, easy for me to say this since I'm not doing any of the work, but it would be interesting to figure out a way to weight draft picks; obviously Carmelo should be worth more than Rautins or Rak or whoever goes in the second round.

 

[MSOrange]

Where I'm sitting to watch the game seems to make a big difference also FWIW. We seem to do better in my wheelchair as opposed to sitting on the couch.

But seriously, very interesting analysis.

 

[jncuse]

Perhaps you can consider Dean Oliver's 4 factors of winning basketball (google it) and apply it to Syracuse to see how they hold. Syracuse is unique in that the 2-3 zone has inherent sterngths and weaknesses that may make certain factors less or more meaningful. Maybe the 4 factors and NBA Draft Picks?

I believe KenPom has the historical data on these team stats, so you can just plop them into your regression..

 

[orangefan13]

FG attempts differential adjusted with the freethrow differential is a huge catagory. That's what turnovers, rebounds, offensive rebounds, blocks, loose balls, and out of bounce balls accumulate to. I am going to be watching that stat tightly this year. So far the first three games we have been worse then our opponents (not a good sign), and that is why we blew out the first two exhibitions.

 

[EastCoast2]

Perhaps you can consider Dean Oliver's 4 factors of winning basketball (google it) and apply it to Syracuse to see how they hold. Syracuse is unique in that the 2-3 zone has inherent sterngths and weaknesses that may make certain factors less or more meaningful. Maybe the 4 factors and NBA Draft Picks?

I believe KenPom has the historical data on these team stats, so you can just plop them into your regression..

Awesome idea. When I get some free time I'll do that.

 

[Dave85]

I thought some of you would find this interesting. Attached is a short (4 page) excerpt from a paper I'm writing for a professional credential. I chose to examine Syracuse's win percentage as a function of several variables. The strongest correlation I have found so far, which shouldn't be a surprise to many here, is the number of future NBA draft picks on the roster. Spoiler alert to my fellow math nerds, the R-squared value (RegSS/TSS) of this regression is 0.81.

0.81 because Route 81 passes through Syracuse. God I'm so smart!

 

[Dave85]

I think if you take the difference between our score and our opponents score it might provide you insight into how well our team is doing.