There’s but a week left in the season and there’s still a chance to win a free BounceX3 T-Shirt by predicting the best individual performance left in the season. This post should provide ample data for those of you looking to make a guess (and for those of you that have tried and failed, I’ve given you a second shot to try and win the shirt.)
Do not go gentle into that good night,
Old age should burn and rave at close of day;
Rage, rage against the dying of the light.
We all have favorite posts. For me, It’s a particular bit of crazy I came up with back in my old website. It was a series of posts (see part 1, part 2 and part 3) built around the concept of a really good night, the kind of night that carries your team to victory, and how the very best players should have a series of very good nights.
I loved it for it’s simplicity and for how it gave me a completely different perspective on a well known story.
Let’s start with the very best nights of the season (thru April 15th):
That’s the top 200 games of the year in fact (Powered by Nerd Numbers, I really missed typing that) . The stats used are Wins Produced (WP) and my own points based version of WP, Points over Par (PoP) , which is a metric that tells the expected number of points a player adds to their team while they are on the court. As per usual, I’ve provided per 48 minutes, per games and totals. Given that I’m trying for awesome here let me throw in some fanservice.
That’s all the stats you could possibly want for those 200 really good nights (a good place for detail on these is here just scroll to the text in gray).
|Rank||Player||Games||% Of Total|
A quick table reveals 39 players feature more than once on the list and that Mr. James and Mr. Paul are simply at another level (something which should surprise no regular reader of this blog).
But wait there’s more. Remember this is not about just the best nights but about the probability of having a good night (or a bad night for that matter). I want to know what I can expect from each player. I want to characterize them and understand what to expect. First we need to tabulate some more numbers.
As of 4/15/2012 we have 16105 Games played by players at greater than 10 minutes with a spread of
|Wins Produced||Points over Par per Game||Wins Produced per 48 minutes||Points over Par per 48 minutes|
Now the conceit is that I’m going to use this distribution to set my levels of performance. I’m choosing Points over Par per game as the metric and I’m setting my levels based on the population distribution and standard deviations much like those standardized test everyone took as kids. You get points for being progressively better than everyone else on a given night and lose points for being progressively worse. The end result looks like so.
So we’ve characterized the data, picked our metric, confirmed normal distribution and set out our scoring system all that’s missing is to set up our ranking.
It’s go time. Here’s your top 53 (a lot of people were tied at 46 ).
It really is the King’s year. Chris Paul and Tyson Chandler make it a race but Lebron carries the day. Those three, Kevin Love and Paul George (no, really) reach the top level on the list. Kevin Love also carries the distinction of reaching the lowest level.
I’m betting people would like to see their own teams right about now.
I’m all about putting that cherry and whipped cream on top.