A Six Sigma Analysis of the Wolves: Player Heat Map

I want to start this post with an apology.  A couple of weeks ago, I proposed doing a full-on 6 Sigma analysis of the Timberwolves to Canis Hoopus, and promised to get going on it right away.  As it turns out, making promises like that at the end of the year is a REALLY, REALLY stupid idea.  Between traveling, hosting guests, shopping, shoveling snow, and various other holiday craziness, my personal time has been basically zero.  Work is just as bad as we’re trying to wrap up a bunch of initiatives before the New Year hits.  Not to make excuses, but it was really crappy timing on my part…sorry about that to everyone that voted to go forward.

I still think it’s worth doing, and I definitely plan on diving in sometime in January.  In the meantime though, there are a few exercises we can do now that will lay the groundwork for the overall project, while also being fun and interesting in their own right.  At some point, I’m sure we’ll end up doing a formal root cause analysis of what makes the Wolves defective, and we’ll have to take a look at the players.  What I’d like to accomplish in this post is to understand how we, as a Canis Hoopus community, view and evaluate our players in an objective fashion.  What are the things we as a group find important when forming our opinions about players?  In order to accomplish this, let’s create a “Heat Map” that charts our collective opinion regarding each player’s effectiveness.

A Heat Map is basically a two-dimensional, XY Scatter Chart where each player will be a data point.  Where the players fall within the chart should be eye opening, and help us identify the studs vs. the duds and the solid role players vs. the mediocre.  Once everyone’s data point is charted, we divide the chart into quadrants and examine which ones rate highly and which ones are an “area of opportunity” for improvement.  I think it makes sense to have one axis be for Offensive Effectiveness and the other for Defensive Effectiveness, however that’s based on my personal opinion they are of equal importance.  If anyone thinks they should be weighted differently, or has a different idea altogether of what the axes should be, let me know.  I’m open to any suggestions you have.

So here’s what I need from you guys.  As we are compiling these ratings, what are the data points we should be using?  When we say Offensive Effectiveness, what does that mean?  As a starting point, I’ll state my personal opinion.  As already mentioned, I think the axes should account for Offensive and Defensive effectiveness.  I support using both current year and historical stats as factors (we’ll talk about weighting factors later).  I think using raw stats would be very complex and overly complicated, so I recommend using a combination of existing metrics like PER, Win Shares, +/-, efficiency ratings, etc… 

I’m also curious about what other factors you feel should be used.  Should age and/or experience be a consideration?  How about usage rates, durability, consistency, or college performance?  Anything and everything is on the table, so if you feel strongly about certain characteristics being a key indicator of effectiveness, throw it out there for discussion.  We can use as many data points as you feel are necessary.  I’ll leave it up to the community to decide which ones they are.  Once we’ve reached agreement on what to use and how to use it, I’ll pull the data, crunch the numbers, and share the results.  As you are leaving your comments, here are the key things I’ll be looking for:

What should the two axes represent?

What offensive and defensive metrics should be used as data points in the analysis?

What ranges of time should be used (such as current year, last 3 years, career stats, or some or all of the above)?

What other factors should be considered?

Should be a fun project.  Looking forward to hearing your feedback…