Valuing Jason Kendall

Use your ← → (arrow) keys to browse more stories
Valuing Jason Kendall

Balking Traditionalism

Personally, when it was announced that we had signed Jason Kendall to a major league deal last season, I was worried. Kendall was coming off of a 2007 season where he hit near the mendoza line, putting up a wOBA of .272 adding up to a combined batting runs above average of -25.6 between his time in both Oakland and with the Cubs. This put Kendall dangerously close to replacement level at catcher. Not only was Kendall terrible at the plate, he was questionable behind it as well, allowing 111 stolen bases while catching only 20 runners, for a runner success rate of 84.7%, compared to a league average rate of 74.4%. To sum it up, although he was a liability at the plate, he was also a liability in the field.

Kendall had a very hot start to the 2008 season, hitting over .500 for much of the first month of the season. However, Jason was due for the mother of all regressions, finishing with a batting average of .238. Kendall’s poor batting average, combined with very little power, resulted in -17.0 batting runs above average. Neutral to position, this is barely above a replacement player, who is assumed to acquire -20 batting runs over the course of a season. So there’s one fear that was realized.

The other fear was that he would continue to be the defensive liability that he was in 2007. In his stint in Chicago, 52 of 57 runners who attempted a steal against him were successful, for a success rate of 91.2%. Runners were only successful against him 69.6% of the time in ‘06, but in ‘05 he posted a rate of 82.1%. The question was whether or not Kendall could return to his pre-2004 form, back when he was with Pittsburgh, when his worst runner success rate was 73.2%.

As anybody who watched the Brewer season knows, Kendall smashed all expectations behind the plate. He was brilliant at fielding balls in front of him, rating at +4 runs by David Pinto’s probabilistic model of range. This number is not necessarily predictive, due to a small sample size, but it still speaks volumes for the results from 2008. He also threw out a whopping 41 runners out of 96 attempted steals, for a runner success rate of only 57.3%, the best number from any full season for Kendall (he had a 56.5% runner success rate in 1999, but he only started 75 games at C that year).

So let’s see if we can put these numbers to some use. Really, the fact that 57.3% of runners against him were successful is meaningless without some sort of way to translate that into the “currency of baseball,” namely, wins. And the easiest way to do that is using the run values derived from linear weights. Although the value of the stolen base is highly dependent on context (stealing 2nd or 3rd in the top of the 9th in a tie game is much more valuable in terms of wins then stealing 2nd down by 4 in the bottom of the 3rd, for instance), we want to eliminate context for simplicity’s sake. The run value of the average stolen base is .18 runs, and the run value of the average caught stealing is -.47 runs. Note that the average run value of the caught stealing is much higher than that of the average out (which is approx. -.3 runs) because not only is the runner making an out, he is also removing himself from the basepaths. So in order to find the amount of runs saved by a catcher, we apply the following formula.

Runs Saved = -.18*SB + .47*CS

Thanks to this formula, we can find the “break-even point” for the stolen base – at which runner success rate does Runs Saved = 0 - The answer is roughly 72% - that is, any catcher who is allowing less than 72% of the runners against him to successfully steal is saving runs for his team, or conversely, any runner who successfully steals more than 72% of the time is creating runs for his team.

Now let’s get back to Kendall. We know that he allowed 55 stolen bases and caught 41 runners stealing. Applying our formula, we find that Kendall saved.(-.18*55 ) + (.47*41) = 9.37 runs. That’s nearly one whole win.

This is all very nice and uses lots of pretty numbers, but it still doesn’t tell us quite exactly what we need to know. What we really want to find out how many runs better he was than the average catcher. So we need to find the average rate of success for base stealers in 2008. A quick use of the Baseball Databank spits out 2799 SBs and 1035 CSs for the 2008 season. That gives an average runner success rate of 73.0%. Here, let’s apply this success rate to the number of stolen base attempts Kendall faced (96). This gives us, that if Kendall were an average 2008 catcher, he would’ve allowed 70.08 SBs and 25.92 CSs. Applying our runs saved formula again, this gives

AverageRunsSaved = (-.18*70.08) + (.47*25.92) = -.43 runs

Now to see how much above average Kendall was as a catcher, we take Kendall’s runs saved (9.37) subtracted by the average runs saved (-.43), and we get 9.80 runs saved above average for Jason.

Of course, this analysis isn’t perfect. There are many factors that come into why a runner gets caught stealing. It’s much harder to steal against left-handed pitchers, for instance. So really, I’m not entirely sure how many of the 9.80 runs should be credited to Jason – to a point, they deserve to be spread between Jason and the pitching staff. For now, let’s just go with it and say that it was all Jason, hypothetically.

Then Jason was worth

-17.0 Batting Runs

4.0 runs by PMR

9.80 runs by SB/CS

10.61 runs by positional adjustment (that’s 12.5*(594/700), since it’s the 12.5 run catcher positional adjustment pro-rated by plate appearances).

16.97 runs by replacement adjustment (pro-rated, similarly to above)

Adding that all together gives us that Kendall was worth 24.38 runs, for a WAR of 2.4, which makes Kendall  an above-average player at a premium position, which is a very valuable piece going forward.

Follow Milwaukee Brewers from B/R on Facebook

Follow Milwaukee Brewers from B/R on Facebook and get the latest updates straight to your newsfeed!

Out of Bounds

Milwaukee Brewers

Subscribe Now

By signing up for our newsletter, you agree to our Terms and Privacy Policy.

Thanks for signing up.