Wednesday, March 5, 2014

What Happens Next?

The 2014 MIT Sloan Sports Analytics Conference is over, and it’s time to get down to work. So what comes next? While this post is more for people looking to break in or break out in sports, it could prompt ideas for insiders as well.

The best way to reach your dreams and goals is to help somebody else reach theirs. So here’s how to take concepts from the conference and use them to build value for others.

Choose. Sports analytics can do much more than people realize, not less. There are many organizations, companies, and individuals in the sports industry who can gain a competitive advantage with quantitative information. Focus on areas where you can show indisputable proof of its value.  They may not be at the highest level of sports, but they certainly exist.

Think beyond the playing surface. The best opportunities may lie in sports business, where innovation can move as slowly as on the field. Identify things that don’t make sense. For example, why do teams spend millions on a gigantic new video board, and then fill it with per game stats, batting averages, and videos of two hands clapping together?

Create. One way to mine innovative – and valuable – research is to merge two or more different types of data. Combining park effect statistics and weather data is one example, and this impressive study is another.

Compel. Some sports insiders will ignore hard data and evidence no matter what. Will anybody ever convince Jim Boeheim to play his bench more? Fortunately, there are college teams led by coaches that think like Brad Stevens, who are more likely to consider new concepts and welcome analysis from outside sources. The level of acceptance will vary widely at all levels, so find the best targets and demonstrate how analytics will make an impact, either in the win column or bottom line.

Sports may take time to embrace change. But once a team or organization succeeds with a fresh approach, others rush to follow.

Thursday, January 16, 2014

Adjusting for Context and Arbitration


One of the keys to solving the value puzzle in sports is separating what the player controls and what he doesn’t. Sounds simple, but it’s not.

In baseball arbitration, context can be a huge factor. In extreme examples – such as pitcher who plays home games in Coors Field – the arbitration process makes the proper adjustments to salaries. But there are many others where it doesn’t. For example, factors like run support, defensive play, and bullpen performance all have a massive impact on starting pitcher statistics.

Relievers also face challenges when it comes to context. Holds and saves greatly influence arbitration salaries, and they get impacted by the man calling the shots. Managers vary tremendously in how they handle relievers, and this can be good or bad for pitching stats. Here are just a few examples:

Pitching Changes: Some managers change pitchers far more often than others. Giants Manager Bruce Bochy used 524 relievers in 2013 versus 440 for Washington’s Davey Johnson. The frequency of Bochy’s changes created more opportunities for holds. Giants relievers had 46 holds of less than one inning. The Nationals had just 15.

Free Passes: Seattle’s Eric Wedge ordered 48 intentional walks last season, compared to 10 by John Farrell and the Red Sox. This hits the statistics of relievers especially hard, since they pitch far fewer inning than starters. Intentional walks cause stats like walks per nine innings and WHIP to rise, even though the pitcher did nothing to cause it. Because intentional walks are usually poor strategic moves, they can also increase a reliever’s ERA.

Wearing Them Down: Overuse can also make relief pitchers less effective. The Red Sox and Indians had similar ERA figures – both for their starting rotation and bullpen – yet Terry Francona worked his pen much harder. Cleveland relievers pitched on consecutive days 122 times. Farrell, on the other hand, used Red Sox pitchers only 71 times without a day of rest.

Every player’s situation is different. And The Sports Resource identifies how context impacts all the stats that matter in arbitration.

Tuesday, June 25, 2013

Stepping Up


Who increased their scoring rate more than anybody else during the NBA Playoffs? Was it NBA Finals MVP LeBron James, the hot-shooting Danny Green, or an emerging star like Paul George? Actually, none of them came close. The correct answer is a backup point guard for a club that didn’t survive the first round.

Andre Miller upped his points per 40 minutes rate from 14.6 during the regular season to 21.8 in the playoffs. His 7.2 points increase led all players who saw at least 1000 regular season and 100 postseason minutes.

Since much of the media focuses on misleading per game stats, vital details like this get missed. While points per game is never a good metric, it’s even less effective during the playoffs when rotations shorten. Regulars see more minutes per game which can cause per game stats to increase even when per minute production dips. Our simple adjustment eliminates this bias. The top 10 appears below. Among superstar players, Chris Paul posted the largest increase.




PS P/40
RS P/40
Change
1
Andre Miller
21.8
14.6
+7.2
2
Gerald Green
21.0
15.6
+5.4
3
Carl Landry
23.1
18.6
+4.5
4
Chris Paul
24.5
20.3
+4.1
5
Udonis Haslem
12.3
8.2
+4.1
6
Draymond Green
12.6
8.6
+4.0
7
Gerald Wallace
13.8
10.2
+3.6
8
Francisco Garcia
15.6
12.5
+3.1
9
Lamar Odom
11.2
8.2
+3.0
10
Andre Iguodala
17.8
14.9
+2.8

One more key point: scoring drops during the playoffs because teams play at a slower pace. Among the 105 players who met the criteria, only 38 (36.2 percent) scored more points per 40 minutes in the playoffs. This makes the performances of Miller and Paul even more impressive.