Connect with us

Fantasy Baseball

A Good Way to Tell If a Player is Fast: Another Stat Mike Trout Is the Best At

Week 9 Injury Report

Over the last couple of weeks I’ve done research on infield hits and what, if any, importance they have.

While I’ve struggled to come away with directional action items to take away from this research, here is what I have learned so far:

  1. Infield hits have a high year-to-year correlation.
  2. Infield hits have a high correlation with stolen basses.
  3. Players that get infield hits on 10% or more of their ground balls are able to steal bases at a much higher rate than players that get infield hits on less than 10% of their ground balls.

The Research

Infield hits have a high (.51r) year-to-year correlation, which suggests that a player’s ability to get infield hit is a skill. Of course a players ability to get infield hits, like all ground balls that go for hits, is BABIP/luck dependent to some extent, but the .51r expresses that there is some innate ability when it comes to  a player’s IFH%.

And when I say infield hits, I mean infield hits per ground ball. We have to look at this on a per ground ball basis, as opposed to a per plate appearance rate, because some players hit fewer ground balls than others. What we really want to know is what players convert a higher rate of their ground balls into hits. The logic would be that players that turn a greater percentage of their ground balls into hits do so because of their speed.

When I originally started down this path, I assumed that that this could explain some of a player’s ability to sustain higher BABIPs than others, and it does explain some of BABIP—infield hits per ground ball (i.e. infield hit percentage (IFH%)) has a .17r with BABIP—but, for now, I went in another direction.

What I found is that IFH% has a very high correlation with stolen bases. For this study I looked at all qualified hitter seasons from 2002 (infield hits were not kept track of before 2002) to 2014 and calculated the correlation between IFH% and, in order to account for the fact that IFH% is a rate stat, stolen bases per plate appearance. This revealed that the two scores have a .42r with each other.

When you think about this it kind of makes sense. The process that occurs when a player hits a ground ball, repositions his feet, takes his first step out of the batters box, and sprints down the first base line is very similar to the process a runner on first base takes when he attempts to steal second. The only difference is that when you steal second base you try to beat a throw from the catcher followed by a tag, instead of a throw from and infielder, which requires no tag.

What Does This Mean?

In order to understand to understand how IFH% correlates with stolen bases better, because one would assume that the correlation isn’t a strictly linear path, but rather a stream with many ebbs and flows, I conducted what was dubbed by Mitchel Lichtman a “poor man’s regression.” This allows us to see if there are certain thresholds that exist that for IFH% that make it more likely that someone is a better base stealer.

I put players into buckets by the rate at which they produced infield hits (i.e. all the players that got infield hits on 9% of their ground balls were put into one bucket, all players that had a IFH% of 10% were placed into another, and so on) and then averaged the stolen base per plate appearance rates together for all of the players in each bucket.

The graph below is what I found:

Screen Shot 2015-04-05 at 9.37.46 PM

When we interpret this graph we can see that the correlation between stolen bases and infield hits is a gradual linear climb until a player gets an infield hit on 9% of his ground balls. Once a player is able to get a hit on 10% or more of his ground balls, he is much more likely to be an elite stolen base threat. The jump from 9% to 10% is followed by a comparable jump from 10% to 11%

Here is a chart of all of the players that were able to produce an infield hit rate of over 10% last year:

Mike Trout15.9%2.27%16
Starling Marte14.5%5.50%30
Lorenzo Cain12.9%5.58%28
Carlos Gomez12.7%5.28%34
Andrew McCutchen12.6%2.78%18
Adam Eaton12.2%2.79%15
Yoenis Cespedes11.8%1.09%7
Jose Reyes11.8%4.58%30
Dexter Fowler11.3%2.18%11
Dee Gordon11.3%9.85%64
Billy Hamilton11.2%9.17%56
Yasiel Puig11.2%1.72%11
Trevor Plouffe10.8%0.34%2
Jose Altuve10.7%7.92%56
Josh Harrison10.4%3.27%18
Xander Bogaerts10.3%0.34%2
Brett Gardner10.2%3.30%21
Helpful Articles for Week 2
[su_list icon=”icon: chevron-right” icon_color=”#160a47″]


One of the implications of this is that now that we have identified IFH% as a proxy for stolen base ability, we cannot only identify which players are elite stolen base threats early on in their careers, but we can also use this as a means to identify which players have lost a step or two before it shows up in their stolen base totals.

More importantly, the next step is to adjust a players IFH% by their handedness.

I haven’t done the math yet, but one would think that on average, because they are closer to first base and the natural momentum of their swing leads them out of the box, left-handed hitters would have a higher IFH% than right-handers.

Once this adjustment can be made, this adjusted IFH% will lead to a higher correlation and predictive way to look at stolen bases.

Mike Trout Photo Credit: Keith Allison

Follow Devin

More in Fantasy Baseball