Monday, January 28, 2013

Putting Base Running into WAR

As the great American League MVP debate raged last summer, one of the reasons some supported Angels outfielder Michael Trout over Tigers third baseman Miguel Cabrera was his speed.  Cabrera backers insisted that speed was far less important than hitting and that Trout's base running contribution should not have been enough to overcome Cabrera's hitting advantage.  They were correct that hitting plays a substantially larger role in winning games than base running but it can make a difference 

Looking at the spread in Batting Runs versus Base Running Runs (BRR) in 2012 shows us how much of an impact hitting had in comparison to base running.  Batting Runs ranged from 114 above average for the Yankees to -109 below average for the Cubs.  On the other hand, Base Running Runs went from 18 for the Angels to -18 for the Nationals.  So, the best and worst hitting teams added/cost six to seven times more runs with their bats than the best and worst base running teams added with their feet.

However, there are some players that create enough runs with their base running where it can not be ignored in calculating a player's involvement in team wins.  According to the FanGraphs base running statistics, Trout added 12 runs or just over one win with his base running, making him the most most valuable base runner in baseball in 2012.  His base running contribution was quite a bit less than either his or Cabrera's or any other big hitter's batting contribution, but it was not insignificant which is why base running is part of Wins Above Replacement (WAR) 

During the 2012 season, FanGraphs decided to separate hitting and base running as components of WAR .  The base running metrics are Weighted Stolen Base Runs (wSB) and Ultimate Base Running (UBR).  I'll use Tigers second baseman Omar Infante to illustrate how wSB is calculated. His combined statistics for the Marlins and Tigers in 2012 were as follows:

PA 554
SB 17
CS   3

By the linear weights theory, Infante gets credit for 0.2 runs for each stolen base and loses 0.4 runs for each caught stealing.  So, that's 17 * 0.2 - 3 * 0.4 = 2.2 runs.  That number then needs to be compared to league average.  Based on MLB totals for plate appearances, stolen bases and caught stealing, the average base runner adds about one run every 1,000 plate appearances.  So, the average player would contribute 554 * (1/1,000) =   0.6 stolen base runs in the same number of plate appearances as Infante.  Finally, subtracting 0.6 from 2.2 yields 1.6 wSB.

There is more to base running than stolen bases though.  For example, a player can use his base running ability to go from first to third on a single or to score on a fly ball.  He can also hurt his team by failing to move up an extra base or being thrown out trying.  Thus, Mitchel Lichtman developed the UBR statistic as a way to account for a runners base running beyond stolen bases and caught stealing.   It is determined using linear weights where each base running event takes on a value according to what bases are occupied, the number of outs and the result of the at bat.  More specifics can be found in the FanGraphs glossary and in Lichtman's primer

Infante had a UBR of 1.1 in 2012 meaning that he added one run with his base running other than stolen base and caught stealing over what you would expect from the average runner.  Adding wSB and UBR gives us Base Running Runs.  In Infante's case that's 1.6 wSB + 1.1 UBR = 2.7 BRR.  Since 10 runs run is worth approximately one win, Infante gets about 0.3 wins added to WAR for his base running.

Other Tigers (including newly acquired Torii Hunter)  can be seen in Table 1 below.   The leaders were speedy Quintin Berry (5.9 BRR), Hunter (4.7) and Infante.  The worst base runners were Prince Fielder (-6.6 BRR), Jhonny Peralta (-3.5) and Delmon Young (-3.0).  There are no surprises there.  The most noteworthy result is probably Austin Jackson's -0.7 BRR.  You would think he would take more advantage of his speed which might explain why they recently hired base running coach Jeff Cox.  

Table 1: Base Running Runs for Past and Current Tigers, 2012

Player
wSB
UBR
BRR
Wins
Berry
3.8
2.1
5.9
0.6
Hunter
0.6
4.1
4.7
0.5
Infante
1.6
1.1
2.7
0.3
Boesch
-0.5
1.4
0.9
0.1
Raburn
-0.4
0.4
0.0
0.0
Dirks
-0.6
0.6
0.0
0.0
Santiago
-0.1
-0.2
-0.3
0.0
Jackson
-2.0
1.2
-0.7
-0.1
Avila
-0.1
-2.4
-2.5
-0.2
Cabrera
-0.4
-2.4
-2.8
-0.3
Young
-1.4
-1.6
-3.0
-0.3
Peralta
-1.2
-2.3
-3.5
-0.4
Fielder
-0.7
-5.9
-6.6
-0.7

Data source: FanGraphs.com

No comments:

Post a Comment

Twitter

Blog Archive

Subscribe

My Sabermetrics Book

My Sabermetrics Book
One of Baseball America's top ten books of 2010

Other Sabermetrics Books

Stat Counter