clock menu more-arrow no yes

Filed under:

Taking the Next Base

New, 4 comments

The reason that I like wOBA so much is that it's so fantastically intuitive. It's easy to get annoyed with WAR and other high level statistics because you don't really know where the numbers are coming from and why they're there. You can trust that some pretty smart people have done a pretty good idea of slogging through the numbers, but it's a little bit harder to wrap your arms around. In contrast, wOBA, as a pure measure of offense, is incredibly gratifying.

Whatever your view of statistics and the measurement of offensive contributions, you would be hard pressed to find anyone who disagrees with the following: in order of usefulness, batters can make an out, draw a walk or get hit by a pitch, hit a single, hit a double, hit a triple, or hit a home run. Each outcome is better than the one before it and on average will provide more value to the team.

So all that we really need to establish is how much more valuable each outcome is. How much better is it to hit a single than draw a walk? Is a single and a triple better or worse than two doubles? The answer to those questions might not be intuitively obvious, but the only real unknowns are the coefficients.

Now is when we still have to put a bit of trust in the smart guys slogging through the numbers. If you have thousands and thousands of baseball innings to go over, you can determine that a single added X runs to the average inning and a double added Y runs to the average inning. This is done every season and with the results of each trip to the plate, we can calculate wOBA.

As baseball can range from The Year of the Pitcher to the steroid era, this is not an entirely static process. Yearly data is used to tweak the coefficients from year to year to reflect the current run environment (link).

On that list, you'll also notice run values given for stealing bases, with an average of 0.247 runs being added from a successful steal and 0.493 runs being lost as a result of a caught stealing. The same methodology was followed to get to these numbers, and they give you a ballpark idea of the value of what can be done on the bases.

If we accept these numbers, here is the value contributed by Yankee base stealers:

Alex Rodriguez: +2.224 runs
DeWayne Wise: +1.729 runs
Eduardo Nunez: +0.989 runs
Derek Jeter: +0.743 runs
Raul Ibanez: +0.741 runs
Brett Gardner: +0.494 runs
Russell Martin: +0.494 runs
Ichiro Suzuki: +0.247 runs
Curtis Granderson: +0.003 runs
Mark Teixeira: +0.001 runs
Jayson Nix: -0.738 runs
Nick Swisher: -0.739 runs
Robison Cano: -0.739 runs

As a team, the Yankees have combined to net +5.449 runs stealing bases, putting them fourth in the American League, behind only the Angles (+10.145), the Athletics (+7.185), and the Rays (6.203). Despite their fifty-two stolen bases being only eleventh most in the American League, the Yankees have benefited from an excellent 77.6% success rate.

Several things about these numbers are interesting to me. Again referencing the constants page on Fangraphs, you'll notice that the margins for stealing bases expand when the run environment shrinks. Stealing a base in 2000 -- the height of a steroid fueled boom of offense and home runs -- was a less valuable act than stealing a base in 2012. Furthermore, getting caught stealing also provided more negative value. So, in order to be a positive play, you had to be successful a lot more often in 2000 than in 2012. That makes sense. You're might risk making an out on the bases with Russell Martin at the plate. You probably wouldn't if it was Robinson Cano.

It follows directly from that that -- as a general rule -- teams with better batters should be more conservative on the bases. Getting caught stealing in front of Granderson, Rodriguez, and Cano probably takes runs off the board. Getting caught stealing in front of Martin, Nix, and Stewart probably doesn't.

As an above average offensive team (first in the AL with .344 wOBA), the Yankees should be a bit more selective in when they take chances on the bases. A lower total of stolen bases but a high success rate is a good reflection of that. They've done exactly what they should do. Pick their spots and try to avoid running themselves out of innings.

Having said all of that, I don't know how much I buy the talk of the Yankees being held back by a lack of speed. It was definitely something that was brought up as a result of bringing Ichiro Suzuki on board. While there's definitely more to speed on the bases -- going first to third, infield hits, forcing defensive errors -- than stealing bases, I have to conclude that the general idea is overvalued.

While every team could benefit from a bench player that can pinch run and steal a base in a close game, stealing bases seems to be a very small portion of the offensive picture. As stated above, the Angels have gained just over ten runs from their Mike Trout base stealing. The AL average is 4.33 runs. So the top base stealing team has accumulated less than six runs above average over the course of nearly 100 games.

I can live with Alex Rodriguez being our top base stealing threat.