Taking the close pitch with two strikes

One of the more enjoyable aspects of being exposed to so much of the 2013 world champion Boston Red Sox over the last few weeks was watching their hitters’ remarkable plate discipline in just about every situation. Repeatedly throughout this post-season I’ve caught myself thinking, my god, how in the world did he take that pitch? So it was no surprise to find out that Fangraphs ranks Boston’s overall swing rate as the lowest in the majors this season with just 43.7 percent.

But it seems that this keen demonstration of non-aggression stands out particularly when there are two strikes in the count. Maybe it’s just me, but I can still remember as far back as little league and I can still hear the calls from beyond the chain-link fence, “two strikes now, swing at anything close!” That’s a bit of an oversimplification at the major league level, I imagine, but I still wince watching an especially relaxed hitter like Mike Napoli sit back and watch a four seamer zip just off the corner of the plate in an 0-2 count.

All this got me wondering about the general major league habits of taking close pitches. So in order to get an idea of how often the league or a team or an individual player balks at such a borderline pitch, I borrowed from two of Fangraphs greatest analysts, Bill Petti and Jeff Zimmerman, and their definition of the strike zone’s “edge”.

Bill and Jeff have done some excellent work studying various aspects of what they’ve dubbed “edge percent”, especially with regards to how pitching on the corners relates to overall pitching success. In fact, I have no problem saying it’s probably my favorite area of work to come out of the saber world this year. Today I’ll be using the PITCHf/x definition of the edge of the plate that Bill detailed in this article, but you can do yourself an enormous favor by exploring any of the related posts on “EDGE%”.

Firstly, to get a general sense of what ‘normal’ plate discipline looks like, let’s look at the major league average swing percentages for pitches on “the edge” from 2010-2013 by count. In the table below “EDGE%” will refer to the percentage of pitches on the edge, “EDGE called strike%” will refer to how often a pitch taken on the edge is called a strike, and “EDGE swing%” will refer to how often the batter swings at the pitch on the edge.

Taking pitches on the edge by count (sortable)

Count EDGE% EDGE Called Strike% EDGE Swing% Sw/cs Ratio
3-0 16.5 81.4 8.3 0.10
0-0 17.6 72.4 30.7 0.42
2-0 17.9 76.4 46.9 0.61
1-0 18.0 73.5 47.7 0.65
3-1 18.9 68.3 62.5 0.92
1-1 18.0 64.4 66.0 1.02
0-1 17.0 59.2 60.9 1.03
2-1 18.6 65.7 68.6 1.04
2-2 17.1 61.0 81.6 1.34
1-2 14.8 56.8 78.6 1.38
3-2 18.9 60.8 84.8 1.39
0-2 12.5 50.2 74.5 1.48

No surprise that hitters are most comfortable letting the close pitch go by in the most hitter-friendly counts— 3-0, 2-0, and 1-0 especially. Additionally, in this table we see more evidence that the tendency to take first pitch is becoming more and more popular in recent seasons and this is clearly just as true for close pitches.

Conversely, batters seem to do what they can to protect the plate with two strikes, or at least for the most part. Swing rates on the edge are highest in full and 2-2 counts, but batters are noticeably more likely to take the close pitch in 0-2 counts with the high probability of a breaking ball in the dirt looming. (This is demonstrated by the extremely low 12.5 EDGE percent.) It is possible that batters quite often may just decide not to swing in 0-2 counts before the pitch is even made.

Of course, these are only the league averages. What if we look at players who are more likely to let the close pitch go by in pitcher-friendly counts? What kinds of patterns will we see there?

I looked at 419 batters who saw at least 100 pitches on the edge with two strikes from 2010 to 2013. The correlation between their overall swing rate and their swing rate with two strikes on the edge was very low ( r = .10). In fact if we look at the players who swung most often in these situations, we get a nice variety of swing philosophies:

Swinging at the close pitch with two strikes

Name pitches K% ZONE% Swing% EDGE swing% EDGE swing% 2 strikes
Norichika Aoki 193 7.5 52.7 42.1 55.4 96.4
Miguel Tejada 152 10.7 51.1 50.6 62.9 95.4
Kevin Frandsen 106 8.6 50.5 48.6 63.0 93.4
A.J. Pierzynski 257 11.0 47.0 56.9 68.6 92.2
Vladimir Guerrero 120 9.4 40.8 58.6 76.0 91.7

At the top of the list is a very patient Norichiki Aoki, whose overall swing rate was lower than that of the 2013 Red Sox at just 42 percent. But, amazingly, Aoki has swung at 96 percent of pitches on the edge in two strike counts. Not too far below that number is the infamously wild hacker, Vlad Guerrero, and his stratospheric 58.6 percent swing rate.

So these two batters clearly have two very different approaches at the plate. In fact, if they have anything in common it’s that they both don’t strike out very often. Over that same time period, both Aoki and Guerrero have had strikeout rates below 10 percent (which is pretty impressive in our day in age), if not for different reasons. Aoki is as patient as they come until he’s forced to swing to protect the plate, while Guerrerro is just about always “protecting the plate” (so to speak) with a flailing, non-discriminating, yet amazingly effective wielding of the lumber.

Similarly, we have a mesh of styles at the opposite end of the spectrum with at least one characteristic in common:

Taking the close pitch with two strikes

Name pitches K% ZONE% Swing% EDGE swing% EDGE swing% 2 strikes
Jack Cust 121 30.8 49.6 39.7 50.2 60.3
Tyler Flowers 117 33.7 49.7 49.3 58.6 64.1
Casper Wells 145 26.5 50.1 48.5 61.5 64.1
Jim Thome 160 27.6 47.1 38.5 53.2 64.4
Geovany Soto 234 24.4 50.4 38.6 48.5 64.5

This is quite the collection of whiffers, yet with various degrees of plate discipline. We know that players like Cust, Thome, and even Geo Soto are notoriously discriminating with their swings, and that apparently doesn’t really change with borderline pitches in two strike scenarios.

But what is confounding is that there are quite a number of players like Casper Wells and Tyler Flowers who have unusually high swing rates in most situations, except in situations where it might benefit them to protect the plate. Chris Davis, Jeff Mathis, B.J. Upton, Brent Lillibridge, and J.P. Arencibia are among the others. Not surprisingly, then, all of these players have strike out rates north of 25 percent over that four year period. As it turns out, strikeout rate correlates much higher with EDGE swing rates with two strikes (at r = -.57) than does overall swing rate.

Back to the 2013 Red Sox

So what about my initial impressions about the world champs? Do the Red Sox let the ball go by more so than usual when we’d least expect it? The answer is yes, but they are not the most prolific in this regard. In fact, the champs ranked third most patient with regards to close pitches in two strike scenarios. Here are the top ten:

Taking the close pitch with two strikes, 2013 teams

# Team pitches EDGE Swing% w/ 2 strikes
1 Astros 1012 75.2
2 Mets 1119 76.4
3 Twins 1158 76.7
4 Red Sox 1114 77.2
5 Braves 1009 77.6
6 Rays 998 77.7
7 Cardinals 1036 78.9
8 Nationals 985 79
9 Yankees 981 79.5
10 Orioles 1031 79.6

So the team that won more games than any other in 2013 and that eventually took home the world series trophy was not quite as patient under these conditions as the team with more losses than any other team in 2013. Clearly then, this approach with two strikes has no bearing on the success of an offense, but it does make that group of hitters unique and by my judgement fun to watch, if not in an excruciating way.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Card Corner: 1973 Topps: Don Baylor
Next: 10 things I didn’t know about birthdays in 2013 »

Comments

  1. NatsLady said...

    Questions.

    (1)Did you eliminate pitchers from these stats?

    (2)A lot of times looking at the close pitch on 2 strikes depends on the situation (runners on base;what kind of zone does the ump have;closeness of the game; number of outs in the inning, pitchcount of the starter,  etc). How much can you factor these “at-bat plans” into the stats, if at all?

  2. MGL said...

    The data is very confusing to me. For example, at a 3-0 count, hitters are swinging at 8.3% of pitches on the edge. that seems awfully high. The average for all pitches at a 3-0 count is 7%! So are we expected to believe that batters swing at more pitches on the edge than overall? That makes no sense to me. That would require a lot of swinging at balls completely outside the zone!

    And of the 92% of the edge pitches that they don’t swing at at a 3-0 count, you would assume that those pitches as a whole would tend to be the worst edge pitches, the ones more likely to be outside the zone than inside the zone (edge is defined as 3 inches on either direction – you should have at least mentioned that rather than require the reader to go to the other link).

    Yet 81% of them are called strikes? OK, the strike zone is a little larger at a 3-0 count, but 81%?

    Even at a 0-0 count when batter take, again, those edge pitches would tend to be the ones more likely to be out of the zone. Yet 72% of them are called strikes. Again, that confused me.

    One thing I would love to know is when a batter swings at each of these counts, what percentage of those pitches would have been called strikes?

  3. Morgan Conrad said...

    O.K., minor elephant in the room.  My anecdotal evidence is that Red Sox and Yankees hitters get away taking close 3rd strikes cause umps give them the close calls.  Any ideas on how to test if this is true or just observer bias?

  4. James Gentile said...

    MGL, I don’t have a code written for inside-of-the-edge and outside-of-the-edge swing rates at the moment. But I am showing that Out-of-Zone swing% in 3-0 counts is about 1%, and In-Zone swing% is about 10% (using a quick query, though).  So I imagine your assumptions are right and that most of those edge swings are occurring on the inner-half of “the edge.”

    So when we look at the overall 3-0 swing rate of 7% you referenced, we’re also including a lot of balls outside the zone (45%) that are basically *never* swung at. Hopefully I can confirm all this with an inside-of-the-edge swing% later on.

    With regard to called strikes on the edge and 3-0, I double checked my results with the second table in this article:

    http://www.fangraphs.com/blogs/more-on-called-strikes-on-the-edge/

    Bill has 3-0 edge called strike rates at 85% for same handed-matchups and 78% for platoons and I have 81% combined. While the overall called strike rate in all counts is about 70% per Bill’s numbers in that same link, so the “strikezone” grows that additional “11%” in 3-0 counts. Does that seem to high?

  5. MGL said...

    “So when we look at the overall 3-0 swing rate of 7% you referenced, we’re also including a lot of balls outside the zone (45%) that are basically *never* swung at.”

    Right, I forgot about that.

    “so the “strikezone” grows that additional “11%” in 3-0 counts. Does that seem to high?”

    Probably not, but I’m not sure.

    I guess also I was unclear that 70% of all edge pitches not swing at are called strikes 70% of the time. I was assuming that around 50% of all pitches in the edge zone were strikes and that fewer than that were called strikes when not swung at. That was a bad assumption on my part.

  6. James Gentile said...

    @NatsLady I did not remove pitchers from the first table, but I should have. Here are the results from the same query with pitchers as batters removed:

    https://docs.google.com/spreadsheet/ccc?key=0AiJWR6a0B5vndERrQlA4TTRYc2Vya2J2ZWN1T3NERHc#gid=0

    The greatest difference between the two table does occur with an 0-2 count, but the difference is only .7 of a percentage point, so nbd.

    As far as adding controls to this type of inquiry, that is something I’ll likely be getting into in the future—that is, looking at what conditions have the biggest affect on swing rates. It is something I started here, but certainly hope to improve upon:

    http://www.hardballtimes.com/main/article/when-are-batters-most-aggresive/

  7. James Gentile said...

    Oh, and @Morgan, I did look at which teams get the most favorable strikezones just a few months ago, I’ll have to dig it up for you.

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>