Cuse Country

Home Away From Dome for Syracuse Orange Fans

Orange Plus (or Minus)

Posted by andrewgp on March 5, 2007

I don’t take credit for this idea, it was fed to me by the Cuse Country brain trust. But I found it so interesting that I spent the better part of the week trying to make it happen.

The point is this: Hockey players are ranked by a simple statistic that counts up all the goals scored by their team while they were on the ice and subtracts all the goals scored against their team while they’re on the ice. This way you can get a picture of what they bring to the game — maybe they help you offensively, but they take chances and hurt you defensively. By simply looking at their Plus / Minus rating, you can see whether their risks are worth their rewards.

So why can’t we have that for basketball? Well, at Cuse Country, you can.

I calculated the Plus / Minus rating for each Syracuse player this season, in each game of the season. Here’s the summary:

Player Season total First 5 gms. Next 5 Next 5 Next 5 Next 5 Last 4
ROBERTS, Terrence +198 +65 +46 +21 +5 +32 +29
NICHOLS, Demetris +217 +86 +54 +15 +5 +24 +33
WATKINS, Darryl +116 +39 +16 +12 +21 +11 +17
HARRIS, Paul +116 +90 +19 +5 +4 -5 +3
WRIGHT, Josh +80 +31 +39 +2 +7 -2 +3
THOMAS, Justin -3 0 -3 0 0 0 0
RAUTINS, Andy +171 +40 +22 +22 +30 +17 +40
BURACH, Todd -9 0 -3 0 0 0 -6
DEVENDORF, Eric +116 +75 +1 -10 +1 +29 +20
GORMAN, Matt +83 +52 +34 -8 +4 +3 -2
BRENNAN-McBride 0 0 0 0 0 0 0
PRESUTTI, Jake -3 0 -3 0 0 0 0
DiLIEGRO, Ross -3 0 0 0 0 0 -3
TEAM MARGIN +215 +99 +38 +34 +13 +6 +25

Even better is the game-by-game analysis (click here). Among the trends you might notice: a great second half by Andy Rautins and a signifcant drop-off for Paul Harris.

It’s interesting to compare it to the team’s margin of victory (or defeat). A player who stayed on the court for every minute would end up with that same rating. (A couple statistical caveats here, for the nerds: substitution data was not available for a handful of games, which may throw off this summary comparison a bit. In addition, in some games it was not possible to determine the second-half starters. Best approximations were made.)


11 Responses to “Orange Plus (or Minus)”

  1. Josh said

    Wow, cool man.

    To me the most interesting trend is Devendorf; even allowing for his illness, he has had quite a “U-shaped” season. Perhaps this represents three phases of his season: (1) early domination of weak opponents with superior skills; (2) poor production as he tried to pull the same stuff he’d done vs. the weaker opponents, but this time against better teams; (3) maturation over the last group of games, coinciding with increased time at PG and improving decision-making. That kind of sums up his year, I think.

  2. OrangeRay said

    Thanks for the statistical breakdown. This plus/minus has been something I’ve been clamouring about. It has some flaws, like any other statistic (such as you may benefit by being in a rotation with four great players) but its a good indicator of your scoring, your defense, and your intangibles.

    Good work.

  3. Tim said

    Maybe this is just my English degree showing, but shouldn’t the sum of the players +/- net out for every 5 game period? For example, for the first five game of the season, the players add up to +480ish, but the team was only +99 in net scoring margin during that stretch.

    Can anyone explain how that works? Or is my math really just that bad?

  4. Tom said

    This is a great stat when averaged out over a season, because no one can hide behind good teammates for every minute of a year as they could in a single game. And it pretty much confirms what we’ve suspected, that Demetris Nichols is really good, and some players (I’m looking at you, Todd Burach) are not pulling their weight around here.

    The Rautins trend is surprising, as it shows that he’s been our most effective player over the last 14 games.

    Great work!

  5. andrewgp said

    Tim — I spent some time wondering about that, too. But I don’t think it works that way. For instance, if five players played an entire game, each of them would equal the team’s margin of victory, and in total they would have five times the team’s margin. So the sum of the players should be five times the team’s margin — and you’ll see they’re very close.

    Why not perfect? Because there are some games we don’t have data for, and there is a little bit of built-in statistical error. But I don’t think those numbers change any of the trends you’ll see. (I’ll iron those kinks out next time.)

  6. syracusan said

    This is amazing. I’m going to choose to take it at face value, because working my way through the potential logical flaws makes my head hurt. I wonder if Jimmy would be interested in something like this? (I’m thinking not, but maybe Hopkins?)

    What jumped out at me:

    1. Terrence is the 2nd most valuable overall contributor to team success, by a wide margin. Three words to explain that: defense, defense, defense. Also a clear trend: his absense (due to injury or foul trouble) hurts our defense and thus brings down everyone else’s +/- while he’s stuck on the bench, preserving his own +/-.
    2. Josh Wright is even more useless than Matt Gorman (a contest which otherwise would have been tough to determine).
    3. Paul Harris’ decline in midseason matches our observations at the time: there was a long stretch where he was not fitting into the flow of the offense at all, and when being on the court was a little disruptive to the team; however, I’m shocked that those numbers haven’t bounced back up over the last 6-8 games.
    4. Andy Rautins numbers are inflated because he has the benefit of being on the court at the same time as the hot shooting offensive spark, Andy Rautins.
    5. Mookie is a hard guy to figure out.

  7. Tim said

    Aha! Thanks, Andy. That makes sense.

  8. Big props on the legwork. Pomeroy has been hustling the plus/minus stat for a while now, but I’m not yet sold.

    Value, at least inherently, lies not simply on having the opportunity to contribute, but rather actually contributing. The plus/minus statistic seems to reward (or demerit) individuals that simply have court time during vital team moments (field goal opportunities, etc.). So, while helpful as a production indicator, it is not a clear illustration.

    As a result, I’m still a Dean Oliver disciple, happily supporting the possession-based mentality. However, should the plus/minus methodology progress, I could be moved a bit.

  9. Jer said

    This is really great work, Andy. Does get penalized on his overall plus/minus because he missed a couple of games due to injury (and a couple where he missed most of the game)? It would seem if we factored this in, it would show him being even more valuable than D-Nic.

    A related question I have is whether this statistic would be even more valuable if it considered how many minutes the player played in a certain game? Since Syracuse as a whole and every individual player (who is not a walk-on) has a positive plus minus, one would assume that if players like Paul Harris and Matt Gorman were allowed to play more, their plus/minus would only go up because Syracuse would be scoring more points than they are giving up (on average). This would seem to show an increase in their contributions to the team effort, but really it just proves that Syracuse (usually) scores more points than their opponents.

    If you divide the player’s plus/minus for a game into the number of minutes they play in that game, you will get their plus/minus for every minute they are in the game. Am I wrong in thinking this would be even more valuable if we are trying to determine a player’s worth while they are on the court?

  10. […] by andrewgp on March 7th, 2007 By request, here’s an extension of our earlier post that showed each player’s plus / minus value over the course of the season. This table shows […]

  11. […] first of the season, though not of his career. (That’ll at least improve the walk-ons’ plus-minus for the season.) Hey Tim, can we get a “Todd Burach” […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: