“Three Rings for the Elven-kings under the sky,
Seven for the dwarf-lords in their halls of stone,
Nine for Mortal Men doomed to die,
One for the Dark Lord on his dark throne,
In the Land of Mordor where the Shadows lie.
One Ring to rule them all, One Ring to find them,
One Ring to bring them all and in the darkness bind them
In the Land of Mordor where the Shadows lie.” J.R.R Tolkien The Lord of the Rings
Basketball is not Baseball. In Baseball for the most part, it is easy to separate the effects of an individuals performance (his value) from external factors. Well not really easy, but it’s been done. Brilliant people like Bill James, and Nate Silver have done the work and now fans, and teams can accurately track the real value of a player. This is now part of the culture of baseball and in fact the culture at large.
Thirty three years after Bill James first self-published The Bill James Baseball Abstract because no one would publish him, the world is a very different place. Oakland Athletics general manager Billy Beane began the revolution, famously applying sabermetric principles to running his team (you have read Michael Lewis‘ Moneyball). The seismic shift came in 2003, when James was hired by a former reader, John Henry, the new owner of the Boston Red Sox. Soon after the crazy stats helped bring a title to Beantown. The journey from obscurity to validation took 25 years for James at personal level.
At a cultural level, 2010 was the year of the stat head. The two watershed’s being the Simpsons episode and Felix Hernandez winning the Cy Young. A 13-12 player won the Cy Young based on his WAR. While this was deserved (and right) five years ago this was unthinkable. Somewhere Joe Morgan is naked and telling the stat geeks to get out of his locker room. Nate Silver is a bonafide statistical rock star.
Why the baseball history lesson? We study history to learn lessons and identify patterns. Stories tend to repeat themselves over time and I believe the Baseball story is being repeated in Basketball. John Hollinger, David Berri,and Dean Oliver all started publishing their work in the late 90’s. A decade later, we know that teams like the Mavs, Blazers, Thunder, Rockets and Nuggets are listening . But we’re not even close to a broader understanding in the culture. I still listen to game announcers and have the same WTF reaction that Ken Tremendous had back in the day (seriously Amare for MVP don’t make me hurt you ESPN man). We are partly to blame.
Sometimes friends have to agree to disagree and move on
In baseball the game is over, the runs are counted and the stat geeks won (and they did it going away). In basketball, the game is early in the second quarter and we haven’t quite figured out how to get out of each others ways. We snipe, we argue and we quibble over our turf and our numbers and we lose sight of the goal. In very simple terms we’re looking for the one number (or set of numbers) that will quantify a player’s value and contribution to his team winning.
At least, this is my goal.
But where actually are we in our prospective basketball revolution. What at this point can we say we know? What is still to be determined? What’s the plan going forward? Here’s my take for the holidays.
We’ve always known how to accrue value. It’s point margin and it’s super correlated to wins (see the Basics or here for my work on point margin). The key finding with Wins Produced is how to assign that value to boxscore stats. Again this is highly correlated. The remaining questions all have to do with accountability and our own technical limitations. Wins Produced uses the available boxscore stats and builds a model to assign value to players based on those stats and the know and confirmed weights. The end result is an effective model with some know and listed deficiencies. Because of lack of play by play/game by game information or the technical challenge of getting it, fair assumptions where made about the numbers not available. For each player, the model assumes his opponent is performing at an average level and the results that we get by making this assumption are for the most part good and consistent.
But now the world is different and we have better tools (by the Power of NerdNumbers, thank you Andres). We can now start looking at opponent data at the game level. We can work this into our analysis and make it better.
Tomorrow for christmas, I’m rolling out the MVP/Team rankings for 2011. Perhaps it will make a difference and perhaps not. It’ll be fun at least. Anything is preferable to the ranting I’ll do if Amare wins the MVP.
The road to the one number is getting shorter every day and perhaps in twenty years some of the luminaries mentioned above can have a guest spot on the Simpsons (and we can make Stephen A. Smith’s head explode).
Arturo's Silly Little Stats v2.0 
Man of Steele
12/25/2010
A quick question. You’ve mentioned a couple of times that Andres has play-by-play numbers/info, but I can’t seem to find them on his blog. Am I missing something? Are you talking about the automated wins produced, or something different?
arturogalletti
12/25/2010
MOS,
He does. It’s under development and not publicly available as of yet. The Season information is what’s available right now.
Man of Steele
12/26/2010
I see. Let me be among those to encourage Andres. Once we have play-charting data, we will be able to make several quantum leaps forward.
entityabyss
12/26/2010
Wait, with this in mind. A couple of thing.
1. Will you be accounting for And1s, technicals, and flagrant freethrows to get an accurate value for each player’s freethrows?
2. Will you use percentages for rebounding rather than pace-adjusted rebound numbers (I feel that works better)?
3. Is opponent WP for players consistent over time (the thing with defense was consistency over time)? I doubt it is with switches and all that. Not saying that it’s bad to use though.
4. When does dberri’s faq come out? That’s unrelated but I was just wondering.
arturogalletti
12/26/2010
ea,
1.With the game to game (which is what were working on) we’re sticking with the available stats. Play by Play (which is next on queue) has more intriguing possibilities.
2. The point margin correlations are strong. I feel if we can get at actual rebs for player and opponent that’ll be better. The issue isn’t the value of a rebound rather the allocation.
3. I need to do the equations to isolate a player’s individual WP from the position then we can look at consistency over time.
4. I’ve read chunks of it. I figure it’ll be out by january.
reservoirgod
12/29/2010
I’m actually very cynical about the “advanced stats movement” in basketball. I don’t think there will be a consensus on 1 number anytime soon because if each person has their own model, then each person can make their own claim to fame and garner that illusive position w/ a team or media outlet. Assuming those are the end goals for every blogger banging out 500-word posts w/ advanced stats, I don’t see how they achieve those goals w/ uniformity. In the land of the blind, the one-eyed man is king and that’s what we have in the “advanced stats community” – a bunch of one-eyed kings. Everybody says they’re Frodo but they’re really Sauron.