Honor Among Thieves

It has become unfashionable to steal things in baseball. The game for a very long time was a place where tricks and traps and grit and hustle were the pieces that formed the core of baseball's offense. Back in baseball's ancient wet-ball-of-twine days, the way you scored runs was to hit the ball where they ain't, jump back and forth in the baseline, bunt the ball around, and steal bases on empty pitches. Turning a single into a double through larceny and generating runs on two singles was the meat and potatoes of scoring runs.

These days people don't run. They don't steal. They take turns standing tall in the batter's box and try to score from home plate by crushing the living daylights out of the ball. Naturally, if you have obscene power and you have a decent chance of putting wood on the ball there's a lot of times you will end up not seeing anything good to hit and they'll just walk your butt to first base where these mountainous sluggers become basepath cholesterol and seize up the offense, unless another slugger can empty the arteries with a big blast.

Speed may be very unfashionable, but it does still exist. And in the pursuit of market inefficiencies it might be a cheap way to generate runs. However, as the people who carry sharp pencils will tell you, stealing bases is only a benefit to the team if you are successful a majority of the time. Fangraphs did a wonderful job of breaking this down, showing the success rate needed in different situations in order to break even as an offensive tactic. In order to have a perpetual green light, you need to have a 75-80% success rate, more or less. If you get caught stealing more than a quarter of the time you should probably sit tight and wait for a ball in play, because the truth is that about 80% of stolen bases don't directly lead to a run being scored.

There are a lot of arguments to be made that adding aggressive baserunning to our offense creates opportunity and disrupts the pitcher's rhythm, but the math people will be quick to point out the obvious; these factors are already included in the results, and the numbers don't lie. You need to be proficient or you shouldn't run, unless maybe you're the back-end of a double steal. So first of all let's look at the most proficient baserunners. These guys should have the green light a lot:

Name SB%
Chris Owings 91%
Andrelton Simmons 91%
Brian Dozier 90%
Billy Hamilton# 88%
Rajai Davis 88%
Kevin Kiermaier* 88%
Mookie Betts 87%
Paul Goldschmidt 86%
Terrance Gore 85%
Keon Broxton 85%
Trea Turner 85%
Hernan Perez 83%
Josh Harrison 83%
Byron Buxton 83%
Tim Anderson 83%
Jason Kipnis* 83%
Marcus Semien 83%
Ichiro Suzuki* 83%
Howie Kendrick 83%
Michael Taylor 82%
Wil Myers 82%
Paulo Orlando 82%
Jarrod Dyson* 81%
Dee Gordon* 81%
Mike Trout 81%
Alcides Escobar 81%
Carlos Correa 81%
Starling Marte 80%
Eduardo Nunez 80%
Leonys Martin* 80%
Javier Baez 80%
Brett Gardner* 80%

Not everyone on this list is equal of course. If you have a huge number of attempts then people are going to be better prepared for you to steal. For example, Billy Hamilton is ridiculously fast. When he gets on first there's a pretty good chance he's going to run, and a very good chance you're not going to stop him unless you cheat the odds a little. You've got to keep him from getting a big lead, you want to keep your delivery short, and call pitches in situations that make it easier for your catcher to deliver a throw to second. When the defense does this it changes the situation for the batter.

Now let's take a look at the second list, the players who are stealing bases as a percentage of stolen base opportunities they see. These guys are getting the green light a lot, and taking it.

Terrance Gore 1.000
Billy Hamilton# 0.392
Keon Broxton 0.377
Rajai Davis 0.287
Trea Turner 0.258
Jarrod Dyson* 0.254
Jonathan Villar# 0.246
Jose Peraza 0.241
Starling Marte 0.233
Mallex Smith* 0.225
Hernan Perez 0.224
Dee Gordon* 0.216
Eduardo Nunez 0.207
Melvin Upton 0.189
Travis Jankowski* 0.178
Michael Taylor 0.167
Kevin Kiermaier* 0.153
Carlos Gomez 0.151

I'll be honest, I don't know who Terrance Gore of the Royals is. B-Ref tells me that he's basically just a pinch runner (This is probably a good thing to have if you had a guy like Billy Butler in your lineup). In his three years in MLB he's got just 9 plate appearances in 37 games, with 19 stolen bases and 2 times caught stealing. He's a designated base stealer (DBS).

After that anomaly you have a guy who you would expect to see at the top: Billy Hamilton. If there's a base open in front of him he's going to take it from you 40% of the time. This is his major method of production. For the 2016 season Hamilton only had a .321/.343/.664 slash, but if you add in his stolen bases and caught stealing his OB% drops to .303 and his OPS climbs to .768. A full 13.5% of all of his offensive production is tied up in stealing extra bases. And he didn't just steal second, he nabbed third base 20 times. That's a real weapon.

Circling back to the beginning of the discussion, we get the last group of guys I want to talk about. If you're a skinny thief they might give you pitches to hit because you can't do that much damage, and if you're a slugging tub of lard they might just walk you because you're not going anywhere when you get to first base. But what if you're both? We know that Mike Trout is definitely both, a slugger with great speed, and he's a perennial MVP candidate. And Bill James created a measure for this combination called PwrSpd, which is calculated as 2(HR*SB)/(HR*SB). It is described by B-Ref as "the harmonic mean of SB and HR."

PwrSpd is meant to be considered based on season totals, but if you adjust the totals to reflect the number of plate appearances the player had (PwrSpd/PA*502) you can see the part-time players who are also balanced threats. In these you might see the next generation of PwrSpd leaders, as long as you don't take sample sizes that are too small (minimum 200 PA). The results might surprise you:

Name P/S per 502PA
Trea Turner 28.97
Keon Broxton 26.54
Hernan Perez 21.95
Mike Trout 21.75
Jonathan Villar# 21.51
Melvin Upton 21.42
Wil Myers 20.79
Michael Taylor 19.70
Paul Goldschmidt 19.51
Mookie Betts 19.46
Eduardo Nunez 19.32
Rajai Davis 19.07
Jose Altuve 18.69
Ryan Braun 18.60
Kevin Kiermaier* 18.55
Brian Dozier 18.31
Jean Segura 18.01
Bryce Harper* 17.93
Charlie Blackmon* 16.76
Carlos Gomez 16.73

This is obviously exciting if you're a Brewer fan, because the best thing you can have in your lineup is a guy who's a threat to hit dingers but can't be given a free base because he'll turn it into two bases and put a runner in scoring position all the time, and there are four current Brewers on that list, and three of the top five.

Moreover, of the numbers I posted - stealing proficiency, stealing prolificity, and power-speed-per-PA - there are only seven names on all three lists (with age):

Name Proficiency Prolificity PS/PA
Trea Turner (23)
85% 0.258 28.97
Keon Broxton (26)
85% 0.377 26.54
Hernan Perez (25)
83% 0.224 21.95
Michael Taylor (25)
82% 0.167 19.70
Eduardo Nunez (29)
80% 0.207 19.32
Rajai Davis (35)
88% 0.287 19.07
Kevin Kiermaier (26)
88% 0.153 18.55

And that's why we should be excited about Keon Broxton and Hernan Perez.