“Reverse Boxscores”: pinning my 95 theses

Here’s what I think has deterred the basketball community from embracing some form of “reverse” or “defensive” box scores:

1. Basketball has no positions: 

Basketball is not like football or baseball.  You cannot state with certainty who is playing what position.  Yes, they “announce” the positions, but that doesn’t mean the players will defend those positions.  You can say who you “think” is playing a position based onwhat he’s doing, but you don’t know.

But its pretty easy to overcome if you simply focus on defense.  Yes, sometimes you will get oddball matchups, but generally, if you walk me into any YMCA at noon, or any NBA court at 7:00 PM, and you give me ten guys and tell me their heights and weights, I can match them up with pretty good accuracy.  Its a left right spectrum.  Height or weight moves you to the right.  Lack of either moves you left.  Where its close, eg Salmons-Stackhouse, its a balancing act.  BTW, my formula accurately placed Stackhouse and Salmons defensively.  Stackhouse guards the 3, Salmons the 2 whenever they are on the court together.  82games had it backwards.

2) “Fairness”

There’s this notion that “we can only credit guys with things they had an influence over” when it comes to rev erse statistics. 

This is the biggest “non-issue” and its whats holding back guys like the Basketball Geek and others from providing simple, straightforward “what did his opposite produce?” statistics.

YES, there are some things the guy can influence and somethings he can’t.  But guess what?  We are humans.  We can discern that a basketball player has “defensive influence” over some things, and he doesn’t over other things.  We can see that by the statistics themselves.  Those that are constant, we can assume he might have some influence over.  (And you’d be surprised at what a player can influence — counterpart opponent rebound average I found to be very consistent… some guys box out, some guys consistently drag their defenders to the perimeter — these are pieces of information we ought to know).

The larger point, and the one I”ve been trying to make for years is this.  There are five guys on the court for each side at any one time.  Somehow, someway, each of the five has a responsibility to “outproduce” one of the opposition’s five players in order for his team to win the game. 

Yeah, some of the things he can influence.  That’s why we match him up against the guy he is most likely to be able to influence.  But, guess what?  Some of the things HE CANT influence… but he STILL has to outproduce the guy. 

Its just like any sport.  In baseball we say, “You have to get such and such numbers out of your first baseman”.  Why?  Because in order to win, he must outplay the counterpart first baseman, and those are the kind of numbers the usual first baseman produces.  Same deal with quarterbacks, etc.

The idea of “reverse statistics” was embraced by hockey analysts years ago.  In basketball we have to get past the notion of “defense” and get to the notion of “comparative production”  with the understanding that the counterpart players do indeed have “defensive influence” over certain statistics.

Join the Reformation.


One last reason to start producing official, reliable, real, reverse statistics.  The clear division of labor in basketball.  Berri and Schmidt spoke of it in The Wages of Wins and it exists.  If you go back through basketball history and take players by height and weight, you will find that the smaller guys produce about 15% of the team’s overall Win Score (each), the middle sized guys about 20%, and the big combine for about 50% (the middle guys are constant at 20%, the little guys were less prior to the three point shot… since the advent of the 3 they’ve stolen some percentages from the bigger guys).  But otherwise these numbers are constant. 

Because this division of labor exists, its fair to match guys with their “height/weight” counterparts because their duties are likely similar.


Leave a Reply

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

WordPress.com Logo

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

%d bloggers like this: