Hitters Who Should Be Shifted More Often

October 22, 2015

The 2016 Bill James Handbook will be on bookshelves on November 1, but we wanted to give you a bit of a sneak preview. In the Handbook, we go through a detailed Q and A that is all about shifts. What follows is one of those questions and answers.

 

Q: Who are some of the more interesting hitters? Specifically, how do these guys perform when shifted compared to when they are not shifted?

A: There are quite a few players who are almost always shifted now (90% of the time and more): David Ortiz, Chris Davis, Lucas Duda, Ryan Howard, Adam LaRoche. Collectively this group hit .199 on grounders and short liners with the shift in place. The 2015 MLB average was .230 on grounders and short liners with a shift on. That's 33 points lower than the MLB average of .263 with no shift on. Let's put that in a chart:

Batting Average on Grounders and Short Liners
Name Shift Percentage When Shifted No Shift
MLB Overall 18% .230 .263

 

But the more interesting hitters are those that were not shifted as frequently as David Ortiz and clan. Here is a sampling of players who were not shifted all the time but were severely hampered when shifted:

Batting Average on Grounders and Short Liners
Name Shift Percentage When Shifted No Shift
Prince Fielder 65% .250 .337
Adrian Gonzalez 62% .148 .195
Edwin Encarnacion 56% .196 .260
Curtis Granderson 54% .192 .392
Albert Pujols 48% .196 .240

 

Edwin Encarnacion and Albert Pujols are particularly interesting in that they are right-handed hitters. Teams are becoming more and more comfortable shifting on righties, despite the fact that there is a bigger gap on the first base side of second with the first baseman needing to stay closer to the bag.

 

The 2016 Bill James Handbook contains 12 additional questions and answers related to shifts as well as more detailed shift analytics. You can pre-order The 2016 Bill James Handbook now.

 
 

COMMENTS (7 Comments, most recent shown first)

jonathangruber
Does the data that tells us the Shift % also tell us the context of the PA? Could it be that teams are shifting on some hitters more often than the table showed but are not doing so with runners on?
9:31 PM Feb 16th
 
flyingfish
So the question that occurs to me is whether a player who is shifted all the time adapts better to the shift than one who is shifted say only half the time or less.
8:44 PM Oct 23rd
 
chill
If the unshifted averages are scattered between .195 and .392, I think we can conclude we're looking mostly at small sample size artifacts. Maybe there's some signal in the data somewhere, but the numbers presented here look like noise.
1:04 PM Oct 23rd
 
MarisFan61
(BTW I'm going to get it anyway, just sayin'.) :-)
2:29 AM Oct 23rd
 
MarisFan61
Can we be sure that stuff is in there?
If so, is there likely to be some awareness of how it might reflect on the stuff that's presented here?

What we see here, and especially how it's stated, isn't very suggestive in that direction.
10:02 PM Oct 22nd
 
Gfletch
Hey, Maris. I think the answer to your question / suggestion is: buy the handbook.
4:13 PM Oct 22nd
 
MarisFan61
Very interesting!

And (or actually I think this should be "But") :-) -- an additional thing that would be interesting and that would perhaps help us to know how to judge that latter listing, including to get a better feel for how meaningful it is, as opposed to being a 'small-sample-random-variation' kind of thing:

How about a listing of players on whom the data show (or would appear to show) the opposite, i.e. whose batting averages on those kinds of balls were higher when shifted against?

The magnitude of those differentials and the identities of the players on that additional list would be very informative. For example, if no players had 'opposite differentials' nearly as large as the differentials in your latter chart, I think that would give some confirmation to the significance of what is suggested in your listed chart; if the differentials on the lists were similar, it might suggest that much of what is being shown is random-ish -- unless, the identities of the players on the additional lists were such that we could tell logically why the shift wouldn't and didn't work well against them.
12:12 PM Oct 22nd
 
 
©2019 Be Jolly, Inc. All Rights Reserved.|Web site design and development by Americaneagle.com|Terms & Conditions|Privacy Policy