Why Shifting Does Work

May 7, 2016

Renowned Baseball Prospectus researcher Russell Carleton wrote an interesting article earlier this week on Baseball Prospectus that utilized the Baseball Info Solutions shifts data made available by the great folks at FanGraphs and questioned whether defensive shifts are actually an effective technique. However, there are a couple of small flaws in the technique that lead to misleading results.

Carleton came up with a very good approach to apply hitter batting averages on balls in play (BABIP) in non-shifted situations to their number of shifted balls in play. This allowed him to compare hitters’ actual totals of hits to their expected totals of hits, finding that the actual hits exceeded the expected hits and concluding that the shift has been detrimental in net.

However, Carleton’s choice of technique creates a few problems. His technique considered single season BABIP in non-shift situations with a minimum of 100 such plate appearances in that individual season. Unfortunately, this eliminates a large number of the most shifted batters in baseball. In 2012, for example, the most shifted batters in baseball were Carlos Pena, Adam Dunn, and David Ortiz. None of them reached Carleton’s threshold of 100 balls in play in non-shift situations, meaning Carleton is leaving out the three players facing the shift the most. In 2015, when shifts were much more prevalent, this approach throws out 7 of the top 11 shifted batters.

More importantly, this approach relies on single season BABIPs based on as few as 100 plate appearances. Contemporary analysis has taught us that while hitter BABIP is more stable than pitcher BABIP, even full season samples should be regressed heavily towards the league BABIP or hitter’s historical BABIP.

For comparison, we can tweak Carleton’s approach ever so slightly, which will lead us to reach the exact opposite conclusion. Instead of treating each season individually, we can group the full 2012-15 sample together and apply the same 100 no-shift plate appearance minimum over 2012-15. This will allow for more stable non-shift sample sizes to help calculate the Expected Hits and allow more players into the sample.


Shifts Data 2012-15 (from FanGraphs)
Hits (Expected) Hits (Actual) Difference
Individual Seasons 9,177 9,222 -45
Combined Seasons 12,540 12,053 487


As you can see, this small change of approach leads us to a much stronger conclusion in the opposite direction, demonstrating that shifts have in fact been working on a large scale.

Neither of these is a perfect approach, of course. You can focus on groundballs and short line drives specifically, for example. We’ve tried this and many other approaches (some of which were previously published Stat of the Week articles here, here, here, here, and here and in the four volumes of The Fielding Bible), and have unanimously reached the conclusion that the shift works. In fact, the results of this tweaked version of Carleton’s study match up quite well with our previous research. In this case, a slight flaw in Carleton’s technique seems to be skewing the results and conclusions, but that should encourage, not detract, from the larger discussion. Carleton is a fantastic analyst, and we’re thrilled to see him dig into a topic so near and dear to our hearts.


COMMENTS (10 Comments, most recent shown first)

Ben, yes the data that showed up-the-middle ball-in-play averages climbing includes all balls in play, not just grounders, and is from the Hit Location table in BB-reference's batting splits.
8:20 PM May 12th

I'm not sure why the pulled and up the middle BABIPs have changed the way that you cite. Is that including all BIP, not just grounders? There could be many factors in play there, just like there are many factors more powerful than shifts impacting the league's decreasing BABIP.

6:59 AM May 12th
Brock Hanke
chuck - Thank you very much. I understand why you can't report on things that no datbase tracks conveniently. Given that, you've done a fine job of responding to me. Thank you!
7:25 AM May 11th
Here’s another small chart of in-play shift data vs. non-shift.
These are the rates of singles to balls-in-play and (doubles+triples) to BIP.
Same deal- shift average / non-shift average, times 100.

Year .. 1b .. 2b+3b
2010 . 104 . 100
2011 . 093 . 101
2012 . 096 . 116
2013 . 092 . 103
2014 . 097 . 098
2015 . 095 . 103

As before, it’s apparent the shift cuts down on hits- just not all kinds of hits. It reduced single/BIP average by about 5%. Pro-rated to the number of PA's with shifts, the averages are:
95.5 for singles (reduces singles in play by 4.5%)
102.8 for doubles and triples (doubles and triples increase by 2.8%)

It would be great to see the home run data, to see how much, if any, players have compromised their HR/PA or HR/batted ball rates in adjusting to shifts.
10:51 AM May 9th
Brock, I agree. ALL data recorded when a batter is in a shift is worth putting into that basket: balls in play, home runs, walks, strikeouts. Then a full comparison can be made between those PA's and his non-shifted ones. Does his power diminish? Does he walk or strikeout more, or less? How does his runs created / 27 outs change?

I just took a look at league splits on the Fangraphs pages, and they only show what happened on balls in play during shifts. From that set of data, one can see the differences, at least, in ball-in-play (BIP) average, BIP slugging and BIP isolated power, for when traditional shifts are used vs. not. They also have a category for non-traditional shifts, which I think is mostly when the infield is playing in to stop a runner at 3rd. There are a lot of SF within the non-traditional shift, and the averages in play are higher.

But the traditional vs no shift comparisons look like this for BIP average, slugging, and isolated power. (SF included, but not SH). I'll show it where the shift averages are divided by the non-shift averages, and multiplied by 100. So anything below 100 is more effective for the defense.

Ball-in-play only
2010 . 103 .. 101 . 098
2011 . 095 .. 095 . 096
2012 . 099 .. 102 . 117
2013 . 095 .. 096 . 102
2014 . 097 .. 097 . 096
2015 . 095 .. 098 . 101

Certainly it appears to work at cutting down average on balls in play. It's more of a mixed bag when looking at the effect it has on in-play power. Last year had the most shifts ever, and isolated power was a tiny bit higher with the shift on.
But you're right- until we can see the data for walks, K's and home runs this is only part of the picture of its effects.​
10:06 AM May 9th
Brock Hanke
I don't understand, at all, trying to evaluate shifts by BABIP. The alleged point of a shift isn't to decrease BABIP. It's to cut down on the hitter's total effectiveness. One big example - shifting against Ted Williams - was designed to keep him from hitting home runs by getting him to hit to left field, at least as much as it was designed to cut down on batten average. Ty Cobb showed Ted how to develop a swing that would hit the ball where it was pitched, but that only produced singles and doubles. Williams said something to the effect that he'd just blast hits past the shift. This, I would think, is a VERY important concept in shifting. I don't care if your shift INCREASES BABIP by twenty points, if it strips a power hitter of his power. In general, if I had the database to actually analyze shift effects, I wouldn't use BABIP at all. I would use the whole suite of hitting accomplishments. Does shifting increase or decrease homers? How about walks, since we know that pitchers pitch differently (focusing on inside) when the shift is on? For a small thing, does shifting increase runners already in place going from first to third or second to home? Just using BABIP is just scratching the surface. Until the people who analyze shifts get over just using BABIP they really aren't telling anyone anything very useful.
2:30 AM May 9th
I wouldn't just shift, I'd place my 4 infielders in the 4 most hit-to spots on each batter's spray chart, no matter where they happened to be (understanding a defender can reach his spot and the 2 spots on either side of his portion of the spray chart).
8:18 AM May 8th
Ben, I was looking at ball in play averages over the past 8 years or so. It looked like left-handed hitters, when they pull the ball, are hitting for a somewhat worse average than they used to, which suggests the shift's effectiveness. But when they hit balls up the middle, they're hitting for a higher ball-in-play (BIP) average.

I put these numbers up under another article recently, but here they are again:

When lefties pull the ball, their BIP average has gone down from .351 in 2010 to just .290 last season. However, their BIP average on balls up the middle has gone from .273 to .302 over the same period.

When righties pull the ball, their BIP average has dropped from .385 to .351; but when they've hit it up the middle it's risen from .262 to .283 (2010 to 2015).

For both lefty and righty hitters, the majority of balls are hit up the middle, so those increases have offset the big decrease in lefty-pull BIP average, keeping overall ball-in-play average from going down.

Assuming shifting is taking away more hits than if no shifts were employed, why the increases on BIP averages up the middle?
5:51 PM May 7th
Weren't DeKalbs some kind of energy converters in Heinlein's story, "Waldo?"
3:41 PM May 7th
And in his spare time he debunked property tax data claims made by advocates of the creation of the city of LaVista Hills in DeKalb County, Georgia, helping lead to its defeat by fewer than 150 votes. Russell Carleton, makin' a difference!
2:08 PM May 7th
©2019 Be Jolly, Inc. All Rights Reserved.|Web site design and development by Americaneagle.com|Terms & Conditions|Privacy Policy