Basketball Analytics

We live in a world of data and sports are no exception.  The first introduction into sport analytics for most people was "Moneyball," which profiled the Oakland A's general manager Billy Beane and his approach for team success through innovative ways.  Basketball analytics is a newer concept that more and more information is being put out there. From the NBA to high school basketball, stats are forcing coaches to adjust.

According to analytic data there are two different shots that are considered good shots for your team. Those two shots are 3's and lay-ups/dunks.  3's are considered a good because of the stat that gives you eFG%, which is effective field percentage.  The formula is eFG% = (FGM + 0.5 x 3PTM) / FGA. So, in not math genius terms.  Player A is 2-6 from 3 for 6 points. Player B is 3-6 from 3 for 6 points. Both point totals are the same thus having the same eFG which is 50%. In the NBA, the metric is slightly different because corner 3's are a little bit shorter than anywhere else on the floor, therefore that is considered a high percentage shot.

One of the known leaders in basketball analytics is Daryl Morey, the general manager for the Houston Rockets.  He has come up with innovative ways to implement to what data suggests is the right way to play.  His biggest experiment is the Rockets' D-League the Rio Grande Vipers.  They are coached by Nevada Smith who is bound to be the old guy at the gym that gets a rebound and sprints out to the three point line to shoot a three (He actually would do that in the North Country coaches' noon time hoops.).  Nevada is a former D3 player and coach (imagine that, back to the D3 love) that loves to play fast and shoot threes.  Here is a great Grantland article on his team from this past season. If you don't want to read, here is a video example of the way they play.



The Portland Trailblazers are also another example of a team that has completely bought in to the analytics craze.  They have coach and general manager that are on the same page, which is the key. Everyone needs to "buy in" in order for it to work. Here is an article regarding their approach from ESPN Insider. Trailblazers.

If you look at the college ranks, the only team that really uses analytics that I can think of is Grinnell and "their system."  Their coach has created his own formula to order to win games.  While, I'm not a huge fan of the way they do things (that's a whole different entry), he has created a buzz about the program and they've won more games than they have before.  Here is a inside look at what they do.



Analytics at the high school and college level are a little different. Why? Zone defense.  It forces teams to attempt mid-range shots which, according to analytic data, is a bad shot.  Now, I know that certain coaches have their own metrics which could include getting 15 offensive boards, holding team to under 40%FG, etc.  It's just really hard to get 15-22 year olds to be consistent enough to get the same results all the time.

I really like analytic data and I think over the course of time you can come up with some great data.  My biggest issue is that none of the teams that I mentioned above have actually won a championship. Then the question becomes, why haven't they won more?  I think the answer to that is analytic data doesn't always carry over in a one game series.  That's why "the best team" doesn't always win the NCAA tournament or the State Basketball Tournament.

You would think that a seven game series like there is in the NBA and MLB would give you enough time for the analytic stuff to work but things are just a little different in the post-season.  What it boils down to is, you have to put the ball in the basket or have guys cross the plate.  In the playoffs, scouting is enhanced and adjustments have to be made.  And it's tough to make adjustments to the way you have played for the whole season for a short period of time to get wins.