Joe Kelly, velocity, and strikeout rate

Joe Kelly throws hard, but doesn't generate a lot of strikeouts. Just how rare is that? - Jeff Curry

Many have considered the connection between strikeout rate and fastball velocity to be obvious, but Joe Kelly's prominence this October causes us to pause and reconsider.

No one would argue that there is a positive relationship between strikeout rate and average fastball velocity. The connection is obvious. It is easier to make contact against slower pitches than faster pitchers and we assume there is a velocity at which a batter could simply not make contact. Many people have demonstrated the relationship, like Dave Cameron back in 2009 and Blake Murphy earlier this year, and pretty much everyone has come to the same conclusion. Fastball velocity matters, but on its own it can only tell you so much. Joe Kelly seems to be the perfect example of that fact.

To get a sense of what we're dealing with let's take a look at Pitch F/X fastball velocities from 2008-2013 and compare them to K% for every season of 40 innings or more. The following graphs include the same data and consider a few different potential types of relationships. The first is a simple linear relationship, the second is an exponential fit, and the third is a theoretical exponential relationship I drew in imagining that once a pitcher routinely threw 104 or 105 mph, batters would have almost no chance to catch up.

Pic1_medium

Pic2_medium

Pic3_medium

The graphs demonstrate that the relationship is positive, just as the previous writers have done and also communicates a few other things. First, strikeout rate is obviously bounded at 0% and 100%, meaning that we'd expect asymptotic behavior at the ends. If you throw either 200 mph or 250 mph, you're probably going to strike out whatever percentage of batters you don't walk. Second, it's also important to consider that relationship is likely not linear. Moving from 90 to 91 mph generates a different effect than moving from 97 to 98. Finally, it's clear from the previous work and from these images that we're talking about an R squared between .18 and .23. Fastball velocity explains something like 20% of a pitcher's K%, but it doesn't explain everything.

That makes perfect sense. At the extremes, velocity matters a great deal. If you could throw 110 mph, the only other variable that is likely to matter is whether or not you can throw it near the zone. But if your velocity is more typical, other factors play a bigger role. The quality of your command, the quality of your breaking ball and offspeed pitch, the amount of movement you generate, your ability to hide the ball, and many other factors contribute to your strikeout rate in conjunction with velocity.

I'm often interested in the outliers more so than the typical performers. It doesn't surprise me that some pitchers do better than other with similar velocity numbers, but those differences can probably tell us something valuable about the entire endeavor. Joe Kelly is where we started, so Joe Kelly is where we'll return. He got me thinking about this idea because I couldn't think of a lot of pitchers who threw as hard as he did (93.3 vFA) while generating such a low strikeout rate (14.9%). It turns out he isn't the biggest outlier.

You can see from the graphs that everyone who averages 97+ has a strikeout rate above 20% and that there are very few who average less than 90 mph and can reach a 30% strikeout rate. If you're interested, one of the points around 90 mph that has a K% near 40 is Koji Uehara this season. We have an easy time imagining those who use deception and good secondary stuff having a better than expected strikeout rate so the ones I'm interested in are those who have a lower than expected rate.

There are many ways to present those who throw harder than most and also strikeout fewer than expected, but let's take the set of those pitchers who were in the top 25% in fastball velocity and the bottom 25% in strikeout percentage. Of the 2,158 pitcher seasons in the sample, 48 met the criteria. Here they are (sorted by lowest K%):

Season Name Team IP vFA K%
2009 Sean White Mariners 64.1 93.4 10.70%
2013 Brandon League Dodgers 54.1 94.1 11.20%
2010 Blaine Boyer Diamondbacks 57 94.6 11.60%
2008 Brad Penny Dodgers 94.2 93.1 12.00%
2009 Blaine Boyer - - - 54.2 93.6 12.00%
2011 Anthony Bass Padres 48.1 93.5 12.10%
2011 Kyle Kendrick Phillies 114.2 93.1 12.30%
2010 Tony Pena White Sox 100.2 94.0 12.60%
2008 Juan Oviedo Royals 48.1 94.2 12.70%
2010 Frank Herrmann Indians 44.2 94.2 12.70%
2010 Danys Baez Phillies 47.2 93.6 13.00%
2009 Bobby Keppel Twins 54 93.2 13.20%
2009 Roman Colon Royals 50.1 94.2 13.20%
2010 Chris Ray - - - 55.2 93.8 13.30%
2008 Freddy Dolsi Tigers 47.2 93.9 13.30%
2013 Matt Albers Indians 63 93.4 13.40%
2011 Frank Herrmann Indians 56.1 93.9 13.40%
2013 Jarred Cosart Astros 60 94.5 13.40%
2008 Jim Johnson Orioles 68.2 93.8 13.50%
2013 Henderson Alvarez Marlins 102.2 93.3 13.60%
2008 Edwin Jackson Rays 183.1 94.2 13.60%
2012 Daniel Bard Red Sox 59.1 93.2 13.70%
2013 Bryan Morris Pirates 65 93.9 13.70%
2008 Manny Acosta Braves 53 93.9 13.70%
2013 Carlos Carrasco Indians 46.2 95.1 13.80%
2009 Mike MacDougal - - - 54.1 95.7 13.80%
2009 Vin Mazzaro Athletics 91.1 93.2 14.00%
2011 Kyle Drabek Blue Jays 78.2 93.4 14.00%
2012 Tyler Chatwood Rockies 64.2 94.1 14.00%
2010 Blake Wood Royals 49.2 95.2 14.10%
2009 Brian Stokes Mets 70.1 94.7 14.20%
2009 Guillermo Mota Dodgers 65.1 94.0 14.30%
2008 Craig Hansen - - - 46.1 95.1 14.30%
2009 Seth McClung Brewers 62 93.2 14.40%
2013 Chad Bettis Rockies 44.2 93.3 14.40%
2011 Scott Proctor - - - 40.1 93.7 14.40%
2008 Radhames Liz Orioles 84.1 93.7 14.50%
2011 Ryan Webb Marlins 50.2 93.9 14.50%
2009 Brad Penny - - - 173.1 94.0 14.50%
2011 Alex Burnett Twins 50.2 93.3 14.70%
2013 Luis Avilan Braves 65 93.5 14.80%
2012 Nathan Eovaldi - - - 119.1 94.1 14.80%
2012 Garrett Richards Angels 71 94.9 14.80%
2013 Joe Kelly Cardinals 124 93.3 14.90%
2008 Gary Glover - - - 54.1 93.1 15.00%
2009 Santiago Casilla Athletics 48.1 94.7 15.00%
2008 Jeremy Guthrie Orioles 190.2 93.4 15.10%
2010 Brad Penny Cardinals 55.2 94.1 15.10%

Joe Kelly's 2013 season shows up five from the bottom of this list. He's out of the ordinary in that only about 2% of pitchers fall into this category, but he just barely falls into this arbitrary category. I don't think any of these pitchers are terribly surprising. Most of the time high velocity pitchers who don't generate strikeouts either lack command or good secondary stuff. The results here aren't surprising, but they can prove valuable.

During the offseason, I'll pursue a more granular analysis of this population to see if we can better understand the network of causes behind hard throwers failing to generate strikeouts. How do command and secondary stuff factor into the equation? Is this a repeatable "skill?" This should prove useful from a scouting and development standpoint as we might be able to use that information to better predict future performance. This is the first step in a longer process, so for now, just sit back and peruse the list. Perhaps we'll call them the Brad Penny All-Stars.

. . .

All statistics courtesy of FanGraphs.

Neil Weinberg is a Staff Writer at Beyond The Box Score, contributor to Gammons Daily, and can also be found writing enthusiastically about the Detroit Tigers at New English D. You can follow and interact with him on Twitter at @NeilWeinberg44.

More from Beyond the Box Score:

Btbs-twitter-insert_medium

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 Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. You should read them.

Join Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. 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