Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Wednesday, February 19, 2014

ESPN: Those Ortiz shifts? They do hurt

The 10th edition of the Hardball Times Annual is out, with an eclectic mix of baseball coverage unlike anywhere else. Admittedly, I’m way over my skis on some of the deeper statistical analysis that is included, although it is indicative of the extraordinary impact advanced metrics has made on the industry.

But one story that immediately grabbed my attention was Jeff Zimmerman’s piece on defensive shifts, entitled “Shifty Business, or the War Against Hitters.’’ Employing date he collected from InsideEdge, Zimmerman reports that Red Sox DH David Ortiz hit against shifts more than any player in baseball in 2013, and it had a measurable effect on his performance.

Ortiz batted 266 balls into the shift (home runs, of course, were not counted). When there was no shift, his BABIP (batting average on balls in play) was .330; against the shift, it was .312, a drop of 18 percentage points. When there was no shift, Ortiz’s percentage of doubles and triples was 14.7 percent; against the shift, it was 8.7 percent, a decline of six percent.

No American League hitter was impacted more by shifts than Orioles slugger Chris Davis. With no shift, Davis had a .425 BABIP and a 17 percent extra-base hit rate (not including homers). Against the shift, his BABIP was .302 (123 percentage points less) and his extra-base hit percentage dropped to 10.1 percent.

Zimmerman also cited research by Bill James and Baseball Info Solutions that showed Ortiz pulled ground balls 82 percent of the time and line drives 60 percent of the time—“pull” defined in these instances as any ball hit to the right of second base.

Thanks to Pat.

Repoz Posted: February 19, 2014 at 11:39 AM | 25 comment(s) Login to Bookmark
  Tags: red sox, sabermetrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Jose Is The Most Absurd Thing on the Site Posted: February 19, 2014 at 11:47 AM (#4658940)
When there was no shift, Ortiz’s percentage of doubles and triples was 14.7 percent; against the shift, it was 8.7 percent, a decline of six percent.


That surprises me. I would expect that against the shift Ortiz' hits would skew to the extra base hits more rather than less often.
   2. dave h Posted: February 19, 2014 at 11:53 AM (#4658947)
Don't have the original article, but presumably the BABIP is insufficient to show the shift is helping, since the times that there is or is not a shift are not distributed randomly. Specifically, fewer shifts and higher BABIP will both be correlated with runners on base.

Is there a way to set a lower or upper bound by looking at video in some sort of quantitative way?
   3. Yoenis Cespedes, Baseball Savant Posted: February 19, 2014 at 12:04 PM (#4658953)
I'm surprised the author didn't match Ortiz's shift/non-shift at-bats based on game state to control for non-random distribution of the shift.
   4. ellsbury my heart at wounded knee Posted: February 19, 2014 at 12:30 PM (#4658970)
That surprises me. I would expect that against the shift Ortiz' hits would skew to the extra base hits more rather than less often.


I think the idea of the shift is partly to take away some doubles down the RF line. It could certainly change the approach of the hitter if they try to go the other way with less power. Or maybe teams that shift a lot tend to have better pitchers anyway.
   5. Hang down your head, Tom Foley Posted: February 19, 2014 at 12:34 PM (#4658972)
Reading the headline, I imagined Ortiz wiping away a tear.
   6. ellsbury my heart at wounded knee Posted: February 19, 2014 at 12:34 PM (#4658973)
I'm surprised the author didn't match Ortiz's shift/non-shift at-bats based on game state to control for non-random distribution of the shift.


I'm surprised the author of this post didn't read the excerpt - which is a blog post about analysis in a book that isn't online. Maybe the actual analyst did do that! Although I doubt it, because it would be a lot more work to do for probably not much difference.
   7. PreservedFish Posted: February 19, 2014 at 01:06 PM (#4658983)
When there was no shift, his BABIP (batting average on balls in play) was .330; against the shift, it was .312, a drop of 18 percentage points.


I'm not a math pedant, but wouldn't this be a drop of 1.8 percentage points? Is a percentage point even a thing?
   8. Joyful Calculus Instructor Posted: February 19, 2014 at 01:17 PM (#4658996)
[7] Yes to both questions. Also, https://xkcd.com/985/
   9. Karl from NY Posted: February 19, 2014 at 03:02 PM (#4659095)
That surprises me. I would expect that against the shift Ortiz' hits would skew to the extra base hits more rather than less often.

That's not how to counterplay a shift. You go opposite field for a single. Shifts are only employed against hitters who don't have XBH power to the opposite field.

Also, I would guess the 2Bman playing basically in shallow right field lets the RF and CF play a bit deeper, to take away a few more XBH.
   10. tfbg9 Posted: February 19, 2014 at 03:06 PM (#4659103)
That's not how to counterplay a shift. You go opposite field for a single. Shifts are only employed against hitters who don't have XBH power to the opposite field.


They are deployed against guys that tend to hit very few ground balls to short and third.
   11. Curse of the Andino Posted: February 19, 2014 at 03:37 PM (#4659139)
That's not how to counterplay a shift. You go opposite field for a single. Shifts are only employed against hitters who don't have XBH power to the opposite field.


Weirdly, Chris Davis does have power to the opposite field. Serious power. His monster year was really a monster first half and he struck out a heck of a lot more after the all-star break. (SSS and all that.) Love the guy, but I don't expect quite as much from him this year.
   12. Jose Is The Most Absurd Thing on the Site Posted: February 19, 2014 at 03:38 PM (#4659140)
Weirdly, Chris Davis does have power to the opposite field. Serious power. His monster year was really a monster first half and he struck out a heck of a lot more after the all-star break. (SSS and all that.) Love the guy, but I don't expect quite as much from him this year.


Ortiz has great power the other way too (there may be a park effect there). It seems to me that a fairly high number of guys who get shifted are actually shaded to the opposite field in the outfield.
   13. Jeltzandini Posted: February 19, 2014 at 03:41 PM (#4659143)
Given that shifts have exploded just as teams have acquired Glengarry quality data, it would be pretty surprising if it turned out that shifts weren't effective.

There will be a focus on countermeasures. We'll arrive at a Nash equilibrium.

   14. bfan Posted: February 19, 2014 at 04:01 PM (#4659159)
There will be a focus on countermeasures.


I saw Brian McCann, a noted shift victim, bunt down the 3rd base line for a hit last year, so there you go.
   15. Walt Davis Posted: February 19, 2014 at 04:02 PM (#4659162)
the 95% ci on 266 balls is about 6%. the babip diff looks random. and why would an if shift dramatically reduce xbh? i will guess the primary effect is the pitcher keeping the ball down and in -- pitching to the shift.
   16. TJ Posted: February 19, 2014 at 04:26 PM (#4659189)
Walt, if you are right, I wish you would have mentioned that to Joaquin Benoit in last year's playoffs...
   17. Nathaniel Dawson Posted: February 19, 2014 at 07:53 PM (#4659344)
I'm not a math pedant, but wouldn't this be a drop of 1.8 percentage points? Is a percentage point even a thing?


This is one of my biggest pet peeves in sabermetric articles, when the author doesn't know the difference between percent and percentage points. Like this:

When there was no shift, Ortiz’s percentage of doubles and triples was 14.7 percent; against the shift, it was 8.7 percent, a decline of six percent.


No, that's a decline of 59%.
   18. Moeball Posted: February 19, 2014 at 08:26 PM (#4659363)
Ortiz batted 266 balls into the shift (home runs, of course, were not counted).


Presumably, when you put a shift on against a LH batter such as Ortiz it is with the idea in mind to make him hit the ball to the right side into the shift and to cut down his BABIP. But to do this you must be willing to pitch the ball over the inside part of the plate. If all you do is pitch to the outer portion, you risk some shots going to left field where no one is fielding. But if you pitch him on the inner part of the plate you also risk him being able to pull one for a HR to RF and thus rendering the shift as useless.

So my question is this? Does Ortiz HR more often with shifts on or with them off? Where does he hit those HRs? I’m curious as to how he actually gets pitched with the fielders in various formations. Do pitchers go outside on him more often when a shift isn’t on? At first glance I would think so but who knows?
   19. ptodd Posted: February 19, 2014 at 10:56 PM (#4659426)
4. ellsbury my heart at wounded knee Posted: February 19, 2014 at 11:30 AM (#4658970)

That surprises me. I would expect that against the shift Ortiz' hits would skew to the extra base hits more rather than less often.



I think the idea of the shift is partly to take away some doubles down the RF line. It could certainly change the approach of the hitter if they try to go the other way with less power. Or maybe teams that shift a lot tend to have better pitchers anyway.


Well, partly but mostly to take away singles hit up the middle or balls hit so hard to right field that the 2nd baseman can not handle it at normal depth (or is too short to catch the LD).

Its also an attempt to offer the batter a single the opposite way in return for not trying to hit a HR

Its probably a big reason that LHB'ers overall advantage over RHB'ers has dropped in recent years in terms of OPS, yet its overall effect on league BABIP seems negligible
   20. bobm Posted: February 20, 2014 at 12:32 AM (#4659458)
I'm not a math pedant, but wouldn't this be a drop of 1.8 percentage points? Is a percentage point even a thing?

This is one of my biggest pet peeves in sabermetric articles, when the author doesn't know the difference between percent and percentage points. Like this:

When there was no shift, Ortiz’s percentage of doubles and triples was 14.7 percent; against the shift, it was 8.7 percent, a decline of six percent.

No, that's a decline of 59%.


I think you mean 41% in absolute terms, but I would still use basis points here: a drop of 600 basis points instead of a "decline of six percent." It's clearer IMO.
   21. SoSHially Unacceptable Posted: February 20, 2014 at 12:48 AM (#4659461)
I think you mean 41% in absolute terms, but I would still use basis points here: a drop of 600 basis points instead of a "decline of six percent." It's clearer IMO.


See, I think it's just best to let the reader see that he he had a doubles and triples rate of 14.7 percent last year and he only hits doubles and triples in 8.7 percent of his AB against the shift. No other explanation is really necessary.

As for the clearness of basis points, I imagine only 50 percent of typical readers know what basis points are, and that number is 49 percent lower if you're just talking readers of the sports pages. (-:

   22. bigglou115 Posted: February 20, 2014 at 01:25 AM (#4659470)
the 95% ci on 266 balls is about 6%. the babip diff looks random. and why would an if shift dramatically reduce xbh? i will guess the primary effect is the pitcher keeping the ball down and in -- pitching to the shift.


But pitchers don't keep the ball down and in, and I'm not convinced they should. Guys who get shifted do so because they pull almost everything, your much more likely to get week contact against them when they try to pull a ball away. Its an oldie but a goodie, you try to turn on that away pitch you hit an easy pull grounder. You pitch inside your working with their pull tendencies to induce harder contact. If they were hitting away pitches away you wouldn't be shifting in the first place.

If I had to guess the drop is part randomness and part the secondary effects of the shift. The 1B gets to play closer to the line, taking away that narrow double band, and the 2B is playing deeper so the RF gets to take a few steps back as well.
   23. Walt Davis Posted: February 20, 2014 at 03:19 AM (#4659485)
But if you are already getting them to try to pull groundballs on outside pitches, you are already inducing lots of weak ground balls and also, inevitably, producing some reasonable number that go the other way.

I know everybody's shift-crazy these days but historically I can only recall the shift being employed against LHB pull hitters with patience. They're dead pull hitters because they're good at waiting for the inside pitch and the outside ones they might swing at are ones they can drive. But I could be wrong.

I'm sure the shift helps some -- obviously putting fielders where the guy is more likely to hit the ball is a good idea.

Ortiz made contact in 430 PA last year. His G/F ratio was .67 -- 40% GB. 40% of 430 is 172. Based on the ratio from the article, about 2/3 of those would have been hit while a shift was on (note he says "batted into the shift" but that's not possible. The other numbers are shift on/off so I assume 266 out of 400 BIP is also shift on). That's 114 ground balls when the shift is on, assuming independence. Shifting has to substantially reduce his BABIP on GB to take away even 3 hits.

Meanwhile, when the shift is on, that drop in XBH% amounts to 16 fewer XBH. Those have to be coming (mainly) from this FB production when the shift is on.

Why would the shift affect his production on FB?

I can't think of a reason it should unless the pitches are located differently. If the ball is kept down, the proportion of GB would presumably be higher, reducing XBH.

But then why not keep the ball down when the shift is off? Because this is too obvious?

Possibilities -- the shift is generally used when the bases are empty. Having men on base generally improves batter performance and/or it's a sign a worse pitcher (or a pitcher having a bad day) is pitching. If men are on but 1B is open, Ortiz is more likely to be walked and can be even more selective.

Ortiz, empty: 287/335/559 (1 walk per 15 PA), 290 BABIP
Ortiz, men on: 330/445/568 (1 walk per <6 PA), 351 BABIP
Ortiz, men on, 1st open: 34 walks in 112 PA

Is it men on (or pitcher quality) or is it the shift? Who knows -- that's in line with his career BABIP split (282/327) but then he's been shifted on pretty much his whole career. I'm guessing that is a pretty extreme split.

Tex doesn't have a split like that. Tex's BABIP with nobody on nosedived in 2010 but then so did his BABIP with men on starting in 2009. McCovey didn't have a split like that. Who are the batters who weren't shifted on before but are now shifted on?
   24. Foghorn Leghorn Posted: February 20, 2014 at 12:11 PM (#4659608)
This analysis, IMO, shouldn't remove HRs, as 18 notes. Pitches aren't selected independent of fielder positioning and baserunners. So, as noted, there;'s likely a different pitch selection in play with the shift. HRs count in the overall effectiveness of the "team defense" with the shift on.

FWIW, ALL LHBs hit very few GB to short and third.

Lastly,
its common that batter make outs on GBs to the same handed side and outs in the OF to the opposite side.
   25. Answer Guy Posted: February 20, 2014 at 01:15 PM (#4659660)
Is it men on (or pitcher quality) or is it the shift? Who knows -- that's in line with his career BABIP split (282/327) but then he's been shifted on pretty much his whole career. I'm guessing that is a pretty extreme split.


I saw managers even shift Ortiz with runners on a few times last season, more or less conceding a SB to whichever of Ellsbury, Pedroia, or Victorino was on first at the time.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
cardsfanboy
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOT: Monthly NBA Thread - October 2014
(583 - 1:07pm, Oct 30)
Last: Los Angeles El Hombre of Anaheim

NewsblogJoe Maddon is to become Cubs manager, sources say
(92 - 1:05pm, Oct 30)
Last: spycake

Hall of MeritMost Meritorious Player: 1959 Ballot
(11 - 1:05pm, Oct 30)
Last: neilsen

NewsblogMadison Bumgarner, World Series legend - McCovey Chronicles
(57 - 1:05pm, Oct 30)
Last: Non-Youkilidian Geometry

NewsblogHeyman: Pablo Sandoval is on Boston's 3B wish list, but so is Chase Headley
(30 - 1:05pm, Oct 30)
Last: villageidiom

NewsblogOT: NBC.news: Valve isn’t making one gaming console, but multiple ‘Steam machines’
(977 - 12:58pm, Oct 30)
Last: Greg K

NewsblogSan Francisco Giants at Kansas City Royals - October 29, 2014 | MLB.com Box
(67 - 12:57pm, Oct 30)
Last: snapper (history's 42nd greatest monster)

NewsblogOT: Politics, October 2014: Sunshine, Baseball, and Etch A Sketch: How Politicians Use Analogies
(4642 - 12:56pm, Oct 30)
Last: snapper (history's 42nd greatest monster)

NewsblogRoad maps to pitching success in Game 7 | FOX Sports
(10 - 12:39pm, Oct 30)
Last: The Chronicles of Reddick

Newsblog2014 WORLD SERIES GAME 7 OMNICHATTER
(1437 - 12:25pm, Oct 30)
Last: Harveys Wallbangers

NewsblogVanguard after the Revolution | NBC SportsWorld
(52 - 12:13pm, Oct 30)
Last: McCoy

NewsblogPrimer Dugout (and link of the day) 10-30-2014
(9 - 11:55am, Oct 30)
Last: Pat Rapper's Delight

NewsblogStatcast: Butler chugs home
(1 - 11:38am, Oct 30)
Last: Davo's Favorite Tacos Are Moose Tacos

NewsblogJapan Times: Nakamura belts three-run homer in 10th to put Hawks one win away from Japan Series title
(10 - 11:32am, Oct 30)
Last: RMc is a fine piece of cheese

NewsblogAngell: The World Series is Almost Over
(1 - 9:22am, Oct 30)
Last: sotapop

Page rendered in 0.4094 seconds
52 querie(s) executed