Improvements in Plate Discipline: Rare but Effective

For even the most physically gifted hitters, offensive success is highly dependent on one’s approach at the plate. No two pitches are exactly the same, and the batter has less than half-second to glean the characteristics that differentiate them — things like velocity, movement, and location. The ability to process and act on this information is key, and can really separate a good hitter from a bad one. Anyone who’s watched more than a few baseball games could tell you that some hitters hinder themselves by swinging at bad pitches, but while plate discipline’s pretty easy to see, quantifying it — or even defining it — is a little more tricky.

One way to measure a player’s plate discipline is through a signal-detection theory approach, which was first introduced by Russell Carleton, and has since been replicated elsewhere. The idea is to calculate how frequently a hitter makes the “correct” decision on the pitches thrown to him — that is, swinging at pitches in the strike zone and laying off pitches outside of it. By no means is this a perfect representation of plate discipline, as there are plenty of instances where swinging at a ball or letting a strike go by isn’t necessarily a bad decision. Lots of other factors come into play, such as pitch type, ball-strike count, and a pitch’s location within the strike zone.

For instance, this metric would penalize a batter for not offering at a breaking ball on the outside corner. Although this pitch would be called a strike, swinging at it might not be in the batter’s best interest if the likely result is a weak grounder. Instead, the smart choice might be to let that pitch go by in the hopes of seeing something more hittable later in the at bat.Overall, though, I think this methodology churns out a pretty good proxy for plate discipline — at least for the purposes of this article. From here on out, I’ll refer to this metric as Correct%.

For reference, here’s a look at the players who had the highest and lowest Correct% in 2014 with a minimum of 250 PA’s. And yes, this implies that George Springer has the best plate discipline in baseball. More to come on him later on.

Top 10 in Correct%, 2014

Name	O-Swing%	Z-Swing%	Zone%	Correct%
George Springer	24.1%	72.9%	49.5%	74.4%
Yunel Escobar	22.9%	69.4%	48.6%	73.4%
Chris Iannetta	19.4%	66.3%	52.1%	73.1%
Matt Joyce	21.7%	66.6%	48.8%	72.6%
Andrew McCutchen	23.7%	68.3%	47.4%	72.5%
Yonder Alonso	25.8%	70.6%	48.3%	72.5%
Dexter Fowler	22.9%	66.4%	45.9%	72.2%
Freddie Freeman	30.7%	75.7%	42.1%	72.0%
Jhonny Peralta	27.6%	71.3%	48.5%	71.9%
Chris Young	22.7%	66.7%	52.4%	71.7%

Bottom 10 in Correct%, 2014

Name	O-Swing%	Z-Swing%	Zone%	Correct%
Endy Chavez	34.7%	54.4%	49.6%	59.9%
C.J. Cron	41.0%	61.8%	44.3%	60.2%
Dee Gordon	33.7%	54.6%	51.6%	60.3%
J.J. Hardy	28.2%	50.0%	52.9%	60.3%
Danny Santana	39.8%	60.4%	51.7%	60.3%
Ben Revere	28.5%	51.8%	54.3%	60.8%
Corey Dickerson	41.5%	63.9%	45.6%	61.0%
James Jones	36.8%	58.7%	49.3%	61.0%
Salvador Perez	44.1%	67.6%	46.5%	61.3%
A.J. Pierzynski	46.6%	73.5%	41.2%	61.7%

Ostensibly, plate discipline seems like something a hitter would be able to learn — at least more so than traditional tools like power, speed, and arm strength. Players can make themselves a little stronger or faster, but only by so much. Even with all of the training in the world, a Molina brother would still be closer to “slow” than “fast.” Generally speaking, you’re either born with these skills, or you aren’t. But it feels like plate discipline should be different. Unlike power and speed, which depend on a player’s physical body, a hitter’s approach has more to do with his mind. Choosing whether or not to swing at a particular pitch is a decision, rather than a reflection of physical prowess.

Nevertheless, most major league hitters don’t get decidedly better (or worse) at making this decision, even after years of big-league experience. Kiley McDaniel recently noted that hitters rarely improve their plate discipline after their first three years as a pro, and that hitters who played year-round in high school tend to top out much earlier. The data seem to back up McDaniel’s claims. As research done by Bill Petti demonstrated, none of the core plate discipline metrics vary by more than 3% over the course of a player’s career.

This probably has something to do with the fact that plate discipline is more about pitch recognition than about making a conscious decision. In other words, it has to do with quickly processing moving objects, which apparently isn’t something most players can will themselves to do. That’s not to say it can’t be taught, and recent research suggests that brain training techniques might be able to do just that. But while the results of this study are promising, they’re far from conclusive. At this point, it’s just one experiment with a relatively small sample size. It also dealt exclusively with college players, many of whom are still relatively new to baseball, and it’s not entirely clear if a more polished hitter would reap the same benefits.

Unsurprisingly, players’ swinging tendencies are very highly correlated from one year to the next, and tend to have some of the highest correlations among all hitting metrics. Here’s a look at players’ year-to-year Correct% going back to 2008, when reliable PITCHf/x data first became available. This considers all players with a minimum of 250 plate appearances in consecutive seasons, for a total of 1,399 data points.

Even with the threshold set as low as 250 PA’s, Correct% has a year-to-year correlation .77, which shows that simply looking at last year’s data gives us a very good sense of how disciplined a hitter will be going forward. There are exceptions, of course, but even they aren’t crazy outliers. The range for Correct% is about generally around 10 or 15 percentage points — the difference between someone like Andrew McCutchen (73 percent) and Dee Gordon (60 percent). Yet since 2008, just over 2% of players (33 of 1,399) moved their moved their Correct% by even four percentage points — in either direction — between consecutive seasons. And just nine players — or 0.6 percent — did so by five percentage points.

Simply put, its very rare that a hitter suddenly makes any wholesale changes to his approach at the plate, but it does happen. Below, you’ll find a list of the most prolific single-season Correct% improvers of the PITCHf/x era.

most prolific single-season Correct% improvers, PITCHf/x era

Name	Year	Δ O-swing	Δ Z-swing	Δ Correct
Cameron Maybin	2012	-5.9%	4.6%	5.3%
Trevor Plouffe	2012	-2.6%	8.1%	5.3%
Gerardo Parra	2011	-4.4%	6.2%	5.2%
Victor Martinez	2011	-0.3%	8.3%	4.8%
Fred Lewis	2009	-1.9%	6.8%	4.7%
Michael Cuddyer	2013	-5.2%	3.7%	4.6%
Ruben Tejada	2011	-0.5%	8.3%	4.6%
Ryan Spilborghs	2010	-5.1%	4.3%	4.5%
Dayan Viciedo	2013	-0.8%	8.5%	4.5%
Michael Brantley	2012	-1.4%	6.8%	4.5%

And the biggest decliners…

most prolific single-season Correct% decliners, PITCHf/x era

Name	Year	Δ O-swing	Δ Z-swing	Δ Correct
Didi Gregorius	2014	4.3%	-9.5%	-6.9%
Kosuke Fukudome	2011	3.8%	-7.1%	-6.1%
Andy LaRoche	2009	1.6%	-8.7%	-5.6%
Chris Duncan	2009	4.8%	-6.9%	-5.4%
Rod Barajas	2010	5.4%	-5.2%	-5.4%
Dioner Navarro	2014	5.1%	-4.9%	-5.1%
David Freese	2011	6.2%	-3.9%	-5.0%
Ramon Hernandez	2011	5.9%	-4.0%	-4.9%
Joey Votto	2010	5.0%	-4.7%	-4.9%
Matt Carpenter	2013	0.3%	-8.8%	-4.8%

I also want to bring to your attention some players who may not have had any drastic, single-season fluctuations, but made some substantial improvements (or diminishments) over the course of a few years. These are the hitters who had the largest sum of improvements from consecutive seasons going back to 2008.

Largest sum of improvements in consecutive seasons, 2008-2014

Name	Δ O-swing	Δ Z-swing	Δ Correct
Andrew McCutchen	1.6%	14.9%	7.8%
Michael Brantley	1.1%	11.9%	6.8%
Dayan Viciedo	-5.9%	7.2%	6.6%
Maicer Izturis	-4.7%	6.6%	6.1%
Trevor Plouffe	-6.9%	4.8%	6.0%
Carlos Gomez	0.3%	13.9%	5.6%
Travis Snider	-1.7%	8.5%	4.9%
Josh Donaldson	-8.6%	0.1%	4.7%
Mark Trumbo	-7.4%	1.5%	4.5%
Willie Harris	-1.1%	5.5%	4.4%

And the largest cumulative diminishers…

Largest sum of diminishments in consecutive seasons, 2008-2014

Name	Δ O-swing	Δ Z-swing	Δ Correct
Didi Gregorius	4.3%	-9.5%	-6.9%
Julio Lugo	8.7%	-3.3%	-6.3%
Dioner Navarro	8.8%	-2.2%	-5.6%
David Freese	6.4%	-4.7%	-5.4%
Chris Duncan	4.8%	-6.9%	-5.4%
Miguel Olivo	-2.5%	-14.4%	-5.3%
Cesar Izturis	8.2%	-1.4%	-4.7%
Matt Carpenter	-5.1%	-12.7%	-4.6%
Derrek Lee	4.7%	-4.1%	-4.6%
Manny Ramirez	0.1%	-8.9%	-4.5%

Interestingly, none of these lists seem to be particularly biased towards young or old players, but appear to be a mish-mash of veterans and hitters still in the nascent stages of their baseball lives. Four sets of 10 players is little more than anecdotal evidence of course, but the complete set of data tells the same story. A player’s age seems to have very little to do with the progression of his Correct%. Younger players might be slightly more prone to improvement (or averse to disprovement), but not by any significant margin.

Ok, so players occasionally become better at choosing which pitches to swing at. But to what end? How does a change in approach affect a player’s offensive output?  By comparing these players’ actual performances to their pre-season Marcel projections, I found that players who made improvements to their plate discipline generally outperformed their forecasted strikeout and walk rates, and most importantly, their expected wOBA’s.

Results for Δ Correct% Groups

Δ Correct%	Δ K%	Δ BB%	Δ wOBA
> 3%	-0.4%	+0.7%	+.004
3% > Δ > -3%	+0.3%	-0.1%	-.009
< -3%	+0.7%	-0.6%	-.013

Next, I’ll delve into a few case-studies, examining players who have drastically altered their plate discipline and/or posses unsusual swing tendencies.

George Springer

You’d never guess it by looking at his strikeout numbers, but per Correct% at least, George Springer was actually the most disciplined hitter in baseball in 2014. The 24-year-old rookie attacked pitches in the zone to the tune of a 73 percent Z-Swing%, but chased pitches outside of it just 24 percent of the time. The issue for Springer, however, was making contact, as his 61 percent Contact% was the worst among all hitters who met the 250 PA threshold last year. This distinction between plate discipline and bat-to-ball skills also came up in Kiley McDaniel’s evaluation of Rangers prospect Nick Williams, who struggles with the exact opposite problem: excellent bat-to-ball skills, but abysmal plate discipline. Springer’s a fascinating player, and it’ll be interesting to see if he’s able to keep succeeding with such an unusual blend of plate discipline and (lack of) contact skills.

Andrew McCutchen

Andrew McCutchen never improved his Correct% by more than four percentage points in a single year, but he’s still an easy choice for the lifetime achievement award. McCutchen’s swung at about 70 percent of balls in the zone the last couple of years, up from just 53 percent back in 2009. His O-Swing%, on the other hand, has barely budged, causing his Correct% has risen from 65 percent to 73 percent. When McCutchen first broke into the league, his Correct% placed him in baseball’s 21st percentile, but has been one of the very best at deciding when to swing in both 2013 and 2014. McCutchen’s plate discipline was never poor, per se, but he’s elevated it to elite levels by becoming more aggressive on pitches in the zone, but not those outside of it. McCutchen’s one of just a handful of players who’s gotten progressively better at reacting to pitches coming his way, and it’s likely played a role in his ascension to superstardom.

Michael Brantley

Like McCutchen, Brantley’s also turned himself into a fairly disciplined hitter since his debut; and also like McCutchen, he’s done it by capitalizing on more balls in the zone. In his first year of semi-regular duty in 2010, Brantley swung at just 48 percent of pitches in the zone and posted a dismal 61 percent Correct%, but he’s been steadily improving ever since — in both Correct% and overall offensive performance. With an impressive .327/.385/.506 batting line, Brantley set career highs in just about every offensive category in 2014, and his 60 percent Z-Swing% and 68 percent Correct% were no exception. Although his plate discipline still pales in comparison to McCutchen’s, Brantley’s 68 percent mark makes him better than average, and much better than the 23-year-old Brantley, who hit .246/.296/.327 in 2010.

Cameron Maybin

McCutchen and Brantley have been the poster boys of improving plate discipline these past few years, but the largest single-season gain belonged to Cameron Maybin. From 2011 to 2012, the toolsy outfielder’s Correct% jumped from 65 percent to 71 percent, moving him slightly below-average to nearly elite. Unfortunately for Maybin, his new-found discipline didn’t translate into improved performance the way it has for McCutchen and Brantley. Maybin improved his walk (+0.1 percent) and strikeout (-2.4 percent) rates that year, but his overall line plummeted thanks to drop-offs in his ISO and BABIP. Maybin has struggled to stay on the field since his 2012 campaign, so he hasn’t had much chance to show if his new-found discipline would parlay into better performance. Things haven’t looked promising though. In 95 games this past year, Maybin posted a respectable 69 percent Correct%, but could only muster a punchless .235/.290/.331 batting line.

Didi Gregorius

Didi Gregorious is an example of a player whose discipline has moved in the other direction. In fact, his 6.9 percent drop-off between this year and last was the most pronounced of the PITCHf/x era. As a rookie in 2013, he made the correct decision an impressive 71 percent of the time, but saw that regress to 64 percent in his sophomore campaign. In terms of percentiles, he dropped from the 95th all the way down to the 15th. Gregorious showed a higher propensity to chase bad pitches in 2014, but the real culprit was his passivity on pitches in the zone, as his Z-Swing% plummeted from 72 percent to 62 percent. Unsurprisingly, his offensive performance took a step back as well, with his K% (+1.3 percent), BB% (-1.8 percent), and wOBA (-.024) all trending in the wrong direction. It’ll be interesting to see which Didi shows up in 2015 following such an unprecedented dropoff.

Endy Chavez

Now for a few words on Endy Chavez, who was the absolute worst at figuring out when he ought to swing last year, making the right decision just 60 percent of the time. Chavez’s O-Swing% and Z-Swing% were both worse than league-wide rates, but his passivity on the balls in the zone was his biggest bugaboo. The veteran outfielder’s 54 percent Z-Swing% was nine percentage points below the league average, and is probably a big reason why he’s hit for such little power in his career. Chavez’s one saving grace has been his ability to make contact — on pitches both inside and outside of the zone. This strategy may not work much longer, however, as O-Contact% generally falls off in a hurry for players on the wrong side of 30.

Henderson Alvarez

Endy Chavez is pretty terrible at deciphering balls from strikes, but even he’s shown at least some discretion on pitches thrown to him. A Correct% of 60 percent is pretty bad, but is still significantly better than the 50 percent mark you’d see if he were simply guessing at random. But what would it would look like if a hitter had absolutely no sense of pitch recognition? What if he decided whether or not to swing before the ball left the pitcher’s hand? Ladies and gentlemen, I give you Henderson Alvarez: the man who swings at equal amounts of pitches both inside and outside of the strike zone. Well, almost. Alvarez’s Z-Swing% (59.8 percent) is a tad lower that his O-Swing% (60.5 percent), but is about as close as it gets. Alvarez has had 284 pitches thrown to him across 103 MLB PA’s, and has made the correct decision just 50.909 percent of the time — pretty darn close to what we’d expect to see if he were basing his decision off of a coin flip. Of course, Alvarez is a pitcher, and pitchers aren’t exactly known for their hitting prowess; but Alvarez is an outlier even among his pitching peers. Considering all players with at least 100 PA’s since 2008, Alvarez’s Correct% is easily the lowest. The next worst belongs to Anibal Sanchez, who checks in at 53.812 percent. Surprisingly, Alvarez’s unsystematic approach has worked relatively well for him thus far. It’s a small sample, obviously, but he holds a .227/.231/.295 career batting line, which is actually pretty good for a pitcher. However, something tells me his .339 BABIP might be unsustainable.

Joey Votto, Matt Carpenter & David Freese

Most of the guys who cracked the “disprovers” lists were unremarkable hitters who just happened to grow a little worse at deciding when to swing. There are also a couple of older players sprinkled in – like Manny Ramirez and Derek Lee — who’s dropoff corresponded with their decline years. But there are three other names that don’t look quite like the others. The player-seasons in question are Joey Votto’s 2010, David Freese’s 2011, and Matt Carpenter’s 2013. Not only were these guys consistently above-average hitters, but their declines in Correct% corresponded with good seasons, where each outperformed his projected wOBA by at least .018. The scatter plots above showed that the relationships between Correct% and offensive performance are far from perfect, so we should fully expect a fair share of outliers. But all three of these outliers have one thing in common: higher O-Contact%. Each improved his O-Contact% by at least eight percentage points that season, which may have played a role in their improved performances. It’s hard to say how all of these factors (Correct%, O-Contact%, wOBA) are related — if they’re related at all — but it could be worth some further exploration.

…

Improving one’s approach at the plate has obvious benefits. Players who do so tend to walk more and strikeout less, which parlays into improved overall offensive performance. Even so, plate discipline just isn’t an area where most players are able to make any noticeable gains. Cases like Andrew McCutchen and Michael Brantley show that considerable changes aren’t completely unprecedented, but guys like this are exceptions rather than the norm. By and large, good plate discipline is something a hitter either has or he doesn’t; and while most — if not all — hitters constantly strive to take a more disciplined approach, only a select few are able to really move the needle.

References & Resources

• Russell Carleton, By The Numbers, “Is Walk the Opposite of Strikeout?”
• Derek Carty, The Hardball Times, “Introducing Plate Discipline Stats”
• Neil Paine, FiveThirtyEight, “The Most Disciplined MLB Hitters”
• Kiley McDaniel, FanGraphs, “Scouting Explained: The Mysterious Hit Tool, Pt. 2”
• Bill Petti, FanGraphs, “Hitter Aging Curves: Plate Discipline”
• Chris Teeter, Beyond the Box Score, “Training the brain to enhance plate discipline”
• Bill Petti, Beyond the Box Score, “What Hitting Metrics Correlate Year-to-Year?”
• Tangotiger, 2004 Marcels