Why St. Louis Cardinals Manager Mike Matheny's Decision to Start Joe Kelly in NLCS Game 1 Is Worrisome

Jeff Curry

On Thursday, Cardinals manager Mike Matheny made the curious announcement that Joe Kelly deserved to start NLCS Game 1.

Before we get to the substantive analysis, I would like to do a thought exercise. You, dear reader, are to put on your manager cap. Your baseball club has made it to a best-of-seven League Championship Series. It is up to you as a leader of men to select a pitcher to start the first game of this series. The pitcher you choose will likely start two games in the series.

Unfortunately, you don't have in front of you the pitchers' ERA, LOB%, BABIP, any of that. Heck, you don't even know the names of the pitchers. Instead, your general manager has given you six stats upon which to base your decision: (1) Strikeout Rate (K%), which measures the percentage of opposing batters a pitcher strikes out; (2) Walk Rate (BB%), which measures the percentage of opposing batters a pitcher walks; (3) Home Run to Fly Ball Rate (HR/FB), which measures the percentage of fly balls allowed by a pitcher that are home runs; (4) Fielding Independent Pitching (FIP), which is based on a pitcher's strikeouts, walks, home runs, and innings pitched; (5) Expected Fielding Independent PItching (xFIP), which is based on a pitcher's strikeouts, walks, and innings pitched as well as the league-average HR/FB rate; and (6) Swinging Strike Percentage (SwStr%), which is simply the share of pitches thrown by the pitcher that opposing batters swing at and miss.

Here are your Game 1 starting pitcher options:

Starter

K%

BB%

HR/FB

FIP

xFIP

SwStr%

Pitcher A

14.1%

6.9%

10.0%

3.80

3.92

7.7%

Pitcher B

14.9%

8.3%

8.9%

4.01

4.19

6.7%

Would you choose Pitcher A or Pitcher B?

If you chose Pitcher A, you chose 2012 Jake Westbrook. If you chose Pitcher B, you chose 2013 Joe Kelly. The two pitchers have similar numbers, so why did 2013 Joe Kelly have a 2.69 regular season ERA while 2012 Jake Westbrook posted a 3.97 ERA?

Here are the two pitchers' more complete stat lines:

Starter

K%

BB%

HR/FB

LOB%

BABIP

ERA

FIP

xFIP

Westbrook ‘12

14.1%

6.9%

10.0%

70.9%

.312

3.97

3.80

3.92

Kelly ‘13

14.9%

8.3%

8.9%

82.4%

.289

2.69

4.01

4.19

The National League average Left On Base Percentage (LOB%) for a starting pitcher in 2013 is 72.6%. In 2012, it was 71.9%. There is a strong correlation between striking batters out and LOB%. While Kelly strikes out opposing batters at an ever so slightly higher rate than Westbrook, his K% is still well below average. This gives a strong indication that his strand rate is likely to be lower than the league average going forward. This means Kelly is likely to give up more runs.

I've also highlight Batting Average on Balls In Play (BABIP). Rather than simply give the inch-deep analysis that 2012 Westbrook was slightly unlucky and 2013 Kelly slightly lucky, I want to display for you the two pitchers' respective batted ball profiles for the seasons in question.

Starter

GB%

LD%

FB%

IFFB%

HR/FB

IFH%

BUH%

Westbrook ‘12

58.1%

20.8%

21.0%

8.3%

10.0%

7.8%

14.3%

Kelly ‘13

51.1%

20.7%

28.2%

8.0%

8.9%

5.4%

28.6%

Putting this batted ball data, as well as Fangraphs National League batter "Speed" score, opponent batter Isolated Power into PaulSpike's Expected BABIP (xBABIP) calculator, we find out that Westbrook's xBABIP for 2012 is .329. Kelly's for 2013 is .314. Both pitchers got luck on balls in play, but 2013 Kelly has been much luckier than 2012 Westbrook.

The bottom line is that 2012 Westbrook was a mediocre pitcher. The "Minus" stats reflect this reality: 104 ERA-, 101 FIP-, and 100 xFIP-. 2013 Kelly has pitched in a mediocre fashion but somehow generated very good run suppression: 74 ERA-, 110 FIP-, and 111 xFIP-.

The precarious nature of Kelly's 2013 run suppression is made that much clearer when comparing Kelly to the average National League starter in 2013.

Starter

K%

BB%

HR/FB

LOB%

BABIP

ERA

FIP

xFIP

SwStr%

NL ‘13

19.0%

7.4%

10.4%

72.6%

.292

3.86

3.82

3.83

8.8%

Kelly ‘13

14.9%

8.3%

8.9%

82.4%

.289

2.69

4.01

4.19

6.7%

Kelly induces swinging strikes at a below average rate. Kelly strikes out opposing batters at a below average rate. Kelly walks opposing batters at an above average rate. Yet somehow he has managed to strand opposing runners at a rate nearly ten percentage points higher than average and his ERA is 1.17 below average. The last two are inextricably intertwined. Neither are likely to continue. The question is not if Kelly can keep up his current levels of run suppression but for much longer can it continue. Three innings? Five? Seven? Twenty?

All of this is why I don't understand why St. Louis Cardinals manager Mike Matheny elected to start Joe Kelly in NLCS Game 1. On Thursday, Matheny said Kelly "earned" the spot by making the most out of his opportunities. While the clichés of coachspeak are often nothing more than bull roar, Matheny's claims that Kelly's lucky run to date "earned" him a start in NLCS Game 1 rings particularly hollow, for the big leagues are not a luckocracy.

Matheny has already shown a penchant for placing unmerited trust in pitch-to-contact hurlers who are able to defy the gravity of their peripherals. Last year there was Mitchell Boggs and to a lesser extent, Kelly. This season Matheny has Seth Maness and Kelly, to a greater extent--an NLCS Game 1 extent. Matheny's faith in unsustainable luck is foolish and troubling. It is only a matter of time before it bites the Cardinals. We can only hope it happens next April as opposed to this October.

Don't get me wrong. I hope that Kelly's run of good fortune continues. I pray the bespectacled sinkerballer does not allow another run this October. But I know that Kelly is not a skillful enough pitcher to maintain a 2.69 ERA or even finish his career with an ERA of 3.30. And that makes me rather worried about Game 1 and any other start this postseason Kelly might make.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Viva El Birdos

You must be a member of Viva El Birdos to participate.

We have our own Community Guidelines at Viva El Birdos. You should read them.

Join Viva El Birdos

You must be a member of Viva El Birdos to participate.

We have our own Community Guidelines at Viva El Birdos. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker