Zone% by Batting Order Position

Giancarlo Stanton is struggling and Giancarlo Stanton’s teammates are terrible. These two things are true. These two might be related. It would make sense that these things are related, because the drop-off from Giancarlo Stanton to the guys hitting behind him is absurdly large.

Stanton has played in 16 games this season, and has hit third in all 16 of those games. The cleanup spot behind him has been a rotation of Greg Dobbs (8 games), Placido Polanco (5 games), and Joe Mahoney (3 games). If you go by the rest-of-season ZIPS projections, Mahoney is the best hitter of the bunch, forecast for a .677 OPS, with Dobbs and Polanco both coming in at .650. Weighted for the number of games played, then, you could say that Stanton has been “protected” by three players with an aggregate OPS projection of .665, a 269 point drop off from his own .934 rest-of-season ZIPS forecast.

To put that in context, the drop-off between a #8 hitter in the NL (.661 OPS) and the pitcher’s spot in the line-up (.485 OPS) is only 176 points of OPS. There is a larger relative difference in expected performance between Stanton and his protectors than there is between a position player and a pitcher. And we know that #8 hitters get pitched differently than other hitters in the line-up because the pitcher is hitting behind them, which is why they receive more intentional walks than any other line-up position. And I think it’s generally accepted that even when they aren’t getting walked intentionally, they’re still pitched to in a different way because the pitcher is behind them, and they’re offered the least protection of anyone in baseball.

The numbers certainly support the extreme uptick in intentional walks, but what do the numbers show if we take out the IBBs? Here are the 2012 NL walk rates by batting order position, both with IBBs included and then removed.

Split PA BB% UIBB%
Batting 1st 12,002 7.6% 7.4%
Batting 2nd 11,720 7.3% 7.1%
Batting 3rd 11,456 9.7% 8.7%
Batting 4th 11,197 9.2% 8.3%
Batting 5th 10,956 8.7% 8.2%
Batting 6th 10,666 8.1% 7.5%
Batting 7th 10,360 7.3% 6.9%
Batting 8th 10,010 8.1% 6.4%
Batting 9th 9,696 5.3% 5.0%

Or, if you prefer a graph, here’s the UIBB% by batting order position.


Whether the lack of walks from the #1 and #2 spots in the order are evidence of protection or poor line-up construction is a topic for another day, so for now, let’s just focus on the UIBB rates from the 3rd-8th spots. Essentially, NL teams begin stacking their hitters from best to worst at that point, and NL pitchers respond by issuing non-intentional walks in a linearly decreasing fashion after they get through the fifth spot in the line-up. It’s not a perfectly straight line, but it’s close.

But, walk rate isn’t necessarily what we’re looking for if we’re looking for line-up protection, right? The idea is more that batters without protection will see fewer pitches to hit, because the walk is less harmful if the guys behind the guy who walks aren’t likely to drive him in. So, instead of just looking at walk rate, we should look at Zone% by batting order position, to see whether we see a big change among #8 hitters in the NL. Thanks to Jeff Zimmerman, we’ve lined up the data from PITCHF/x for the last five years by a player’s spot in the batting order. Here’s the data for both leagues.

NL Zone%   AL Zone%
1 47.5%   1 51.4%
2 49.3%   2 50.1%
3 48.1%   3 50.9%
4 49.9%   4 50.0%
5 49.0%   5 47.9%
6 50.6%   6 46.3%
7 49.7%   7 47.6%
8 51.8%   8 46.2%
9 52.3%   9 48.5%

This is a pretty fascinating result. #8 hitters in the NL see almost as many pitches in the strike zone as #9 hitters, the spot predominantly occupied by pitchers. No batting order spot occupied by a position player sees more pitches in the zone than the guys hitting directly in front of the pitchers. The guys with the least protection see the most strikes.

This data seems to suggest that the quality of the batter at the plate is the primary factor behind how many strikes he’ll be thrown, not the quality of the batter on deck. #8 hitters in the NL are bad hitters, which is why they’re hitting eighth, and it seems that pitchers aren’t afraid to attack them because of their own weakness, even though they have the pitcher’s spot due up next.

In the one spot where we know that IBBs offer real evidence of “protection”, we do not find much evidence of that protection carrying over to situations where the pitcher chooses to pitch to the batter.

But, what if we get back to Stanton and the Marlins? After all, he is seeing the third lowest rate of pitches in the strike zone of any hitter in baseball this year, and his walk rate has jumped from 9% last year to 14% this year. Just on its face, this seems like a pretty clear indication that the lack of protection behind him is the driving force for his big drop in pitches in the strike zone.

Here’s the only problem: Stanton had no protection last year either. During the 117 starts he made in 2012, here is who was hitting behind him:

Player Games OPS
Logan Morrisson 43 0.707
Gaby Sanchez 20 0.556
Carlos Lee 19 0.654
Greg Dobbs 18 0.698
Justin Ruggiano 5 0.909
Omar Infante 4 0.754
Austin Kearns 3 0.733
Donovon Solano 2 0.717
Rob Brantly 1 0.832
Chris Coghlan 1 0.394
Donnie Murphy 1 0.661

The weighted average of his protectors OPS last year was .680, not dramatically better than the guys hitting cleanup behind him this year. And yet, pitchers threw him strikes, and he hit home runs, and no one really talked about Stanton’s lack of protection. He only drew nine intentional walks all year. It sure didn’t seem like pitchers were going out of their way to pitch around Stanton to get his weaker hitting teammates.

So, yes, Giancarlo Stanton’s Zone% is down this year, and yes, Giancarlo Stanton has really lousy teammates. But, it doesn’t seem like he’s really that much worse protected than he was a year ago, and the Zone% data for batting order positions doesn’t really suggest that protection is the driving force behind how many strikes a batter is thrown. Stanton’s going to see a lot of pitches out of the strike zone because he’s a scary hitter, and pitchers don’t throw strikes to scary hitters. He might see fewer pitches in the zone this year, but I wouldn’t rush to that judgment just yet.

As Jeff noted yesterday, Albert Pujols has seen a similar drop in percentage of pitches in the strike zone, and the Angels just spent $125 million to protect him with the scary left-handed bat of Josh Hamilton. If we’re going to explain Stanton’s drop in Zone% as being a factor of his teammates lack of intimidation, that theory would also have to hold up in situations where intimidating teammates are added to the mix. The Pujols/Hamilton dynamic, and the way pitcher’s attack #8 hitters in the NL, suggests that it’s just not that simple.

Print This Post

Dave is a co-founder of and contributes to the Wall Street Journal.

32 Responses to “Zone% by Batting Order Position”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Eminor3rd says:

    Wait, so pitchers get intentionally walked 0.3% of the time?

    Vote -1 Vote +1

    • davef says:

      guessing it’s people who pinch hit for the pitcher in the 9th slot being IBB’d.

      +7 Vote -1 Vote +1

    • the sauce says:

      I was going to guess that pitchers would walk Stephen Strasburg hoping that not embarrassing him would prevent their getting hit by a 99 mph Strasburg special.

      Then I checked all pitchers with 10 PA last year and saw nary an IBB. I feel pretty confident in saying davef has it right.

      Vote -1 Vote +1

    • IHateJoeBuck says:

      Do the NL stats factor in interleague games? That would cause a few more IBBs in the 9 slot.

      Vote -1 Vote +1

  2. Frosted Tips says:

    Good analysis. However, I think it might be a little simplistic to assume opposing teams and pitchers are only looking at the OPS (or on base percentage) of the guy on deck. Logan Morrison may have had a similar OPS last year to Placido Polanco’s projected OPS, but I would think an opposing pitcher would be slightly more intimated seeing “former highly regarded prospect Logan Morrison” on deck compared to “over the hill veteran with no power Placido Polanco”. Of course, that may not change the way Stanton is pitched to, I just wouldn’t assume most pitchers are smart enough to ignore the “name value” of the hitter and just focus on the numbers.

    Vote -1 Vote +1

    • Frosted Tips says:

      Although, this theory does go against what appears to be happening in Anaheim, where opposing pitchers do not seem particularly intimidated by Josh Hamilton, despite his name recognition.

      Vote -1 Vote +1

    • JRoth says:

      Looking at the 4 guys with more than a few PAs behind Stanton, it’s mostly guys with respectable MLB track records and/or power (Sanchez had hit 19 HRs in each of the previous 2 years). Whatever their OPS in 2012, they weren’t nobodies that any self-respecting pitcher would prefer to face with a runner on board.

      That said, I certainly agree that the Pujols/Hamilton example cuts the other way (although, given what’s become of Pujols, I’m not sure he’s a comp for Stanton anymore – Stanton was better by 25 points of wRC+ last year).

      Vote -1 Vote +1

  3. Patrick says:

    The fascinating result is that, as you move down the batting order zone% clearly trends down in AL and up in NL. WTF could be driving that??

    +7 Vote -1 Vote +1

    • WWMcClyde says:

      Exactly. I want to know more about that. No idea what could be driving it.

      Vote -1 Vote +1

    • Aaron says:

      That is definitely the weirdest thing about this, and it’s not like we’re talking small sample size here.

      Vote -1 Vote +1

    • Tight rope walker says:

      We need to see the swing %. It’s likely that the lower hitters chase more pitches.

      Vote -1 Vote +1

    • Jon L. says:

      It is bizarre. By reputation, pitchers in the NL challenge hitters more with fastballs in the zone, while pitchers in the AL nibble more and try to get them to chase out of the zone. I don’t see much evidence for that here overall, but maybe it does reflect a small philosophical difference between leagues in how to approach weaker hitters.

      Vote -1 Vote +1

    • Owen says:

      Not only that, but the AL breaks into two very distinct chunks: 1-4 (50-51.4%) and 5-9 (46.2-48.5%). Makes me want to see wRC+ of those lineup spots.

      Vote -1 Vote +1

      • Ian R. says:

        It breaks down even more cleanly than that, as the #6 and #8 hitters see the fewest pitches in the zone, by a pretty significant margin.

        My theory is that that’s where managers tend to put their pure power threats: guys who can hit a fair number of homers, but lack the plate discipline needed to hit near the top of the order. It makes sense that pitchers would throw fewer strike to those guys, since A) they’re probably chasing bad pitches and B) a pitch in the zone may very well be punished. In contrast, the #7 and #9 spots are often viewed as secondary leadoff hitters, so the guys who bat there have very limited power.

        Just a thought, and I don’t have any data to back it up, but it makes intuitive sense.

        Vote -1 Vote +1

    • chuckb says:

      This is very strange and interesting. My guess is that tight rope walker is right, but only b/c I have no explanation for it.

      Vote -1 Vote +1

      • Mike says:

        I’d love to see the correlation between the 5-9 hitters in an AL lineup and what percentage of breaking balls they see. Are pitchers more interested in tricking hitters? Are the less likely to throw strikes to some to get to a more favorable matchup?

        Vote -1 Vote +1

  4. jesse says:

    I would wonder how the zone% differs with men on base. I would think that 8 hitters in the NL would have lower number with men on.

    Vote -1 Vote +1

    • Trotter76 says:

      Situation makes a huge difference in how #8 hitters are approached. With a runner in scoring position and less than 2 outs, you might want to walk him and go after the pitcher. However, with 2 outs, you want to go right after the #8 guy and have the pitcher lead off the next inning. I don’t know if all that comes out in the wash over the course of the entire league’s entire season, but it’s definitely something to consider.

      Vote -1 Vote +1

  5. David says:

    Wouldn’t the hitters in front of Stanton have a significant effect on his walks as well? Without Jose Reyes and Hanley Ramirez getting on base in front of him it’s much easier to walk him.

    +7 Vote -1 Vote +1

  6. isavage30 says:

    Regarding the NL numbers, I would think having the pitcher batting next cuts both ways, as far as how you’re going to pitch the guy in front of him. It it’s a situation where the #8 hitter’s coming up with men already on base, then it would make sense to be careful and work around the hitter and throw more pitches off the plate. But, with no one on base, pitching around and potentially walking the #8 hitter in an NL lineup would be a very bad result, and it would make sense to attack the hitter. He’s likely not a guy who you’d be afraid is going to hit one out, and even if you give up a hit, you have the pitcher next.

    Vote -1 Vote +1

  7. DowntownChico says:

    I love the questions this article raises. In particular, the results on the Zone%.

    It would take more in depth research, and would decrease the sample size, but there may be more clear answer if one were to look at the situational Zone%.

    For example, when the batter has 3 balls on them. This is the moment where the pitcher/cathcer/coaches have to make the choice to go at the hitter, or work around him.

    Vote -1 Vote +1

  8. Ctownboy says:

    Except with Dusty Baker, the Catcher mostly bats eighth.

    With the Reds having Corey Patterson, Wily Taveras, Drew Stubbs, Alex Gonzalez, Orlando Cabrera, Edgar Renteria and Zach Cozart on the team and hitting at or near the top of the order since Baker has been Manager, I seriously doubt the Catcher has been the WORST hitter on the team all of these years……

    Vote -1 Vote +1

  9. taprat says:

    I wonder if, in Stanton’s case this year, it’s as much about the rest of the team as the guy batting right after him. It’s a cliche, but there may be an element of truth to it when a pitcher says, “I’m not going to let *that* guy beat me. Make the other players beat me.” Similar to Bonds in San Francisco when he was the rest of his team was mediocre, when you know you’re only going to see the guy 3-4 times per game, and the entire rest of the team bites, you have to consider walking him every single time.

    Vote -1 Vote +1

  10. AJS says:

    I think this does point to evidence of protection, at least in the AL. Lineups tend to get weaker around the 6 hitter, and we see that the 5 hitter faces far fewer strikes than batters 1-4. I’m not sure why things aren’t the same in the NL, other than to say that NL lineup construction is different without the DH (traditionally a big bopper) in the middle of the lineup.

    Vote -1 Vote +1

    • chuckb says:

      But 6-9 see fewer pitches in the zone than 1-5 and I doubt that it’s b/c of how great those bottom of the order hitters are.

      Vote -1 Vote +1

  11. Patrick says:

    How did you get those numbers by position in the order?

    Vote -1 Vote +1

  12. Patrick says:

    How did you get those splits by position in the lineup?

    Vote -1 Vote +1

  13. Daniel says:

    As far as the Pujols/Hamilton thing goes, I’m pretty sure that the lack of decrease in ZONE% may have a lot to do with the Angels playing the Rangers 6 times, who know Hamilton’s glaring weaknesses and pitched around Pujols like Brendan Ryan was protecting him.

    Vote -1 Vote +1

  14. Patrick says:

    I did some more digging into why it might be that AL pitchers throw fewer strikes as they move through a lineup and NL pitchers throw more. Didn’t arrive at a good answer yet, but I can say that lineup construction is similar in both leagues.

    More detail on my blog

    Link text

    Vote -1 Vote +1

  15. Jonathan says:

    Is there an easy way on Fangraphs to sort team or league-wide performances by position in the batting order?

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>