»
S
I
D
E
B
A
R
«
Updated NCAA Plus-Minus 01-27-10
Jan 27th, 2010 by Jon Nichols

I have updated college basketball plus-minus for today:

Further Explanation of NCAA Plus-Minus
Jan 24th, 2010 by Jon Nichols

I think there’s been some confusion about the college basketball plus-minus I’ve been presenting here lately, so I’d like to provide some more information explaining what the numbers mean exactly.

The most important thing to realize is that the offensive, defensive, and overall plus-minus numbers are “net.”  This means every player is essentially compared to only his teammates and not the rest of Division I.  Therefore, the worst team in the country can have just as many positive plus-minuses as the best team.

Why does it work this way?  The plus-minus I use is on/off.  That means that each rating is a comparison of how the team does when a given player is on the court versus how they do when he is off the court.  For example, if Kentucky scores 80 points per game while John Wall is on the court and 75 points per game while he is off the court, his net offensive plus-minus is 5 (80-75=5).  If they allow 60 points per game while he is on the court and 65 while he is off the court, his net defensive plus-minus is -5 (60-65=-5).  As you can see, it’s better to have a positive offensive plus-minus and a negative defensive plus-minus.  Overall net plus-minus is simply the combination of offensive and defensive plus-minus.

Hopefully it is clear now why really good teams have a lot of players with negative plus-minuses.  If not, here is a real example.  Eric Bledsoe of Kentucky has a -10 net overall plus-minus.  Is he a bad player?  Absolutely not.  Remember that his plus-minus is simply a comparison of Kentucky’s production while he is playing to Kentucky’s (and only Kentucky’s) production while he is not playing.  It is safe to say that Kentucky is quite good even without Eric Bledsoe.  So he’s being compared to a very high standard.

This leads me to my next point.  Plus-minus is very unreliable without a large data set (in this case, games played) to work with.  I have made this point many times, but it bears repeating: take these early plus-minus numbers with a grain of salt.  Things can and likely will change as the season progresses.

Two more clarifications: the first three columns (net overall, net offense, and net defense) are per an average-paced game, which is estimated to be 66.5 possessions.  Finally, the column “plus-minus total” is simply the sum of the points a player’s team scored while he was in the game minus the points they allowed while he was in the game.  For those unfamiliar with on/off ratings and unsure why good teams have players with negative net plus-minuses, this column may be what you thought you were looking at.  If you notice, just about every player on good teams has a positive plus-minus total.  Again, this is a measure of how that player’s team outscored their opponents over the whole season while that player was in the game.

Updated NCAA Plus-Minus 01-15-10
Jan 15th, 2010 by Jon Nichols

I have updated college basketball plus-minus!  See below.  Remember you can click the arrow in the top left to view the document full screen or download it as an Excel spreadsheet.

Updated NCAA Plus-Minus (as of 1/8/10)
Jan 8th, 2010 by Jon Nichols

I have updated college basketball plus-minus numbers today, along with a few helpful improvements. If you’re wondering why the numbers are so much different than last time, be sure to check the first bullet:

• The biggest and most significant change is the way in which plus-minus is now calculated.  Previously, I compared the team’s production with the given player on the court to the team’s production as a whole.  Now, I’m comparing the team’s production with the given player on the court, as before, to the team’s production with the player OFF the court.  For example, John Wall’s plus-minus is the difference between Kentucky’s differential when he’s playing and Kentucky’s differential when he’s not playing.  This puts my plus-minus in the more traditional on/off form that most people are used to seeing.
• Every team and player in Division I is now available.
• Along with each player’s net overall, offensive, and defensive plus-minuses, the spreadsheet now includes the player’s total plus-minus and the estimated amount of possessions they were on the court for.   The plus-minus total is simply the total amount that player’s team outscored or was outscored by their opponent while the player was in the game.

The numbers are embedded in the table below.  To download the spreadsheet as an Excel document or view the online spreadsheet in full-screen mode, click the arrow in the top left.  You can also use the filter to sort by team.  Enjoy!

NCAA Plus-Minus 01-01-10
Jan 1st, 2010 by Jon Nichols

Just in time for the new year, net plus-minus for the current NCAA Division I college basketball season is available! By net plus-minus, I mean that the team’s performance with that player on the floor is compared to the team’s performance overall. It is measured on a per possession basis. For example, if a player has an overall plus-minus of +5, it means that his team is expected to outscore the opponent by five points over an average-paced 40 minute game. For overall and offensive plus-minus, a more positive number is better, while for defensive plus-minus a more negative plus-minus is better.

However, you must approach these numbers with caution. First of all, it is very early in the season. Most teams have only played 10 or 12 games, so plus-minus with such a small sample is very unreliable. As the season progresses the numbers should become more accurate. Secondly, as I’ve mentioned before, the official play-by-play data is filled with errors, so for some players the numbers may be slightly off. Also, some players and teams aren’t available, so if there’s a team you’d like to see that’s not already there, let me know. Finally, we must remember that all plus-minus stats have their own strengths and weaknesses. This is not adjusted plus-minus, so a player’s teammates are not taken into account. Therefore, if a poor player is often paired with an excellent player, the poor player may look better than he actually is.

With all of that being said, below is a spreadsheet containing all of the data. To sort by team, click on the filter at the top and choose which team you’d like to see. Also, if you click the arrow at the top left, you can download the file as an Excel spreadsheet. If you notice any errors, please let me know.

»  Substance: WordPress   »  Style: Ahren Ahimsa