October 8, 2009

The Amazing Shotgun

Filed under: Uncategorized — wwinston @ 9:29 pm

The shotgun formation is amazingly effective.  As explained in my new book Mathletics you can easily assign a point value to each NFL play. For example, a 3 yard gain on first and 10 is usually worth around 0 points. For the 2008 NFL season I found that when teams were in the shotgun they generated  an average if.214 points per play and when not in the shotgun they generated an average of only .163 points per play.

   More amazingly, on rushing plays from the shotgun teams generated  on average .249 points per play while on passing plays fro  the shotgun teams generated an average of only .208 points per play. It looks like the element of surprise really helps the rushing plays succeed from the shotgun formation.

October 4, 2009

Who will win the World Series?

Filed under: Uncategorized — wwinston @ 10:17 pm

Although we do not know who will be the final playoff team (Minnesota or Detroit) these two teams are pretty equal so I went ahead and played out the baseball playoffs and World Series 10,000 times. I began by using logistic regression (se chapter 40 of  Matheletics) to obtain chess-like ratings for each playoff team. I then adjusted these ratings based on the relative strength of each team’s playoff rotaiton vis a vis their regular season pitching rotation. After factoring in home edge here is the fraction of the time each team won the World Series.

  • Yankees 26%
  • Angels 19%
  • Red Sox 17%
  • Dodgers 12%
  • Phillies 10%
  • Rockies 7%
  • Cardinals 5%
  • Tigers or Twins 4%

The reason for the Cardinal’s low chance is that they played the easiest schedule in baseball. Toronto, by the way played baseball’s toughest schedule.

Top Players of the Decade?

Filed under: Uncategorized — wwinston @ 7:49 am

                I am getting tons of flack about saying that last season Lamar Odom, per minute was more valuable to the Lakers than Kobe. This gets to the heart of how you evaluate basketball players (see my book Mathletics for more details).  There are two schools of thought on evaluating NBA players. One school says make the most of the box score data. For example, NBA efficiency measures a player based on the  following metric (evaluated on a per minute basis).

((Points + Rebounds + Assists + Steals + Blocks) – ((Field Goals Att. – Field Goals Made) + (Free Throws Att. – Free Throws Made) + Turnovers)).

       There are two problems with this metric: The box score stats are weighted towards offense (and basketball is half defense!) and the “weights” that multiply each box score stat are fairly arbitrary. In fact if a player shoots 35% (which is poor) taking more shots increases his NBA efficiency. I believe that David Berri and his colleagues (see http://dberri.wordpress.com/) have found the best set of linear weights based on box scores stats. By PER and NBA efficiency  Kobe was much better than Odom last season. By Win Score Kobe was only slightly better than Odonm and was inferior to Odom during the 2007-2008 season.

        My partner Jeff Sagarin and I believe in evaluating players based on  how they impact the score of the game and the chance of the team winning We find the set of player ratings that best fit the scores of the game segments which hare defined by who is on the floor. We call this Adjusted +/-, since we modify the old hockey measure to adjust for who you play with and who you play against. On this metric Lamar comes out much better than Kobe last season (but not in other seasons).  The nice thing about Adjusted +/- is that it measures defense as accurately as offense.

   Based on our ratings from 2000-2009 here are the top 10 players of the decade.           


Top 10 Players of the Decade

1.KG +11.7

2.Duncan +11.2

3. LeBron James +10.9

4. Dirk +10.4

5. Wade +8.9

6. Chris Paul +8.6

7. Kobe Bryant +8.4

 8. Stockton +8.2

9. Manu Ginobili +7.5

10. Rasheed Wallace +7.5

        Of course, you say, why is Rasheed Wallace here? In his prime (2001-2006), Rasheed averaged a +10 rating. How can he be better than Shaq?  First of all Shaq’s best years were in the 1990’s. Remember Rasheed’s team beat Kobe and Shaq in the championship. This only happened after the Pistons traded for Rasheed.  If Kobe and Shaq were so great how could this happen? By the way, Paul Pierce and Ray Allen are #17 and #18 for the decade, so it looks like this year’s Celtics have 4 legitimate Hall of Fame Candidates (although they are all probably past their prime).  Rasheed’s technical fouls propensity may prevent him from being realized for the great player that he was (is?), but the numbers speak for themselves.

October 2, 2009

Apology to Troy Murphy

Filed under: Uncategorized — wwinston @ 9:31 am

In an interview with True Hoop I said Troy Murphy had done little to help his team. I apologize. I should have said that with Golden State he had done little to help his team (average rating of -8 points in his last 4 years with Golden State). But in the last two years with Indiana he has improved to the level of an average NBA player(which is very good indeed). Kudos to Troy and the Pacers coaching staff for this improvement. It is rare to see such an amazing improvement during the middle of a player’s career,

More on Cavs Magic Series

Filed under: Uncategorized — wwinston @ 8:08 am

We have a player rating for each player during each game. Ben Wallace finished the Magic series with a -42 points rating and -216 Impact rating. After three games he had a -34 points rating so there were indications of problems. Wally Szcezrbiak  also had a subpar series. He had a -35 points rating for the series. When Wallace and Wally were both out Cavs played 23 points better than an average NBA team. Rest of the series Cavs played 20 points worse than an average team.

October 1, 2009

More on the Bens: Gordon and Wallace

Filed under: Uncategorized — wwinston @ 11:03 pm
  1. Here are some good points posters made about Ben Gordon:
  2. Bulls will play Deng Salmons and Hinrich more to replace Ben’s minutes- Good point but Hinrich and Deng do tend to get injured. Deng Rose Hinrich Tyrus and Noah was a very good lineup but in all other minutes without Ben Gordon last year Bulls played 6 points worse than an average team
  3. I said Deng would take Gordon’s position. I meant to say Deng would take many of Gordon’s minutes in starting lineup. In today’s NBA positions do not have much meaning. Josh Howard is now a #2 guard.
  4. Bulls need cap room for the 2010 savior-From media LeBron stays with Cavs or goes to Knicks. If  Bulls have inside track on Chicago native D Wade then I am wrong, because  Wade was LeBron’s equal last year. Then dumping Gordon was right for the long term.  No other free agent out there is worth the 20 million max contract, with possible exception of  Joe Johnson

         On to Ben  Wallace and the Magic -Cavs series. I forgot to mention that in the  Cavs Magic series minutes when Ben Wallace was in without Joe Smith LeBron was always out there. To go -58 points in 57 minutes with LeBron out there is unreal! Rest of time LeBron was in Cavs played 18 points (per 48 min) better than average.

More on Ben Gordon

Filed under: Uncategorized — wwinston @ 6:50 am

Thanks for all the interest on my Ben Gordon comments. Hopefully this post will clarify things even more.

Here are Ben Gordon’s metrics for each season. First number is Impact Rating, 2nd number is points rating , 3rd number is Offense Rating and last number is defense rating. Thus in 2008-2009 Gordon had a 45% Impact (4th in league) and played 6.94 points better than average per 48 minutes. He was 8.55 points better than average on offense and 1.6 points worse than average on defense.

Ben Gordon’s Adjusted +/- Metrics.

2004-2005 42%  5.37 6.12  0.76

2005-2006  -2%  -1.25 2.95  4.19

2006-2007 23%  6.38  8.67  2.29

2007-2008  -16%  -6.44 1.96  8.4

2008-2009  45% !!  6.94 8.55 1.6

These numbers include playoff performance. Ben Gordon had a 77% Impact rating in Celtics Bulls series!

Throw out 2007-2008 (coaching problems etc.).  Gordon averages 4.5 points above average overall and Impact of 27%.  In a typical season these averages would place Gordon 40th in points and 15th in  impact.   He is young and healthy. This would indicate he deserves one of the Top 40 salaries in the league. Using the methods described on  page 234 of Mathletics (based on Value of Replacement Player from baseball) I get that (assuming salary cap does not move up) Gordon should make around $14 million per year.

« Newer Posts

Powered by WordPress