BILL JAMES ONLINE

Swing Hard; maybe you'll run into something

September 17, 2021
             I had a question from a reader a couple of weeks ago; I can’t remember the exact wording, but the essence of it was, are there some players now, with all of the strikeouts, who would be better off if they would put the ball in play more often?  Is the number of those players rising?   Could you just perhaps figure runs created with the ball in play and runs created on home runs, walks and strikeouts, maybe with a basic Runs Created formula?

            I’m not sure whether the reader intended for me to read this into it, but what I took from the question was, if the number of players who are better off just  putting the ball in play is rising, will this eventually—or even SOON—begin to cut into the epidemic of strikeouts?

            To deal with a technical issue up front, one cannot use a basic runs created formula for this purpose.   The reason that you cannot is that:

A)    Total bases are a foundational element of Basic Runs Created,

B)    The only total bases in the "three true outcome" or "immediate outcome" category are home runs,

C)     Some players hit zero home runs,

D)    Anything else, multiplied by zero, produces zero. 

 

As a result of those things, any player who hits zero home runs would  calculate as having zero runs created on Immediate Outcome plays, even if he had 120 walks.  A player with 100 walks and a very low number of home runs, like Richie Ashburn or Nellie Fox, would calculate as having an unrealistically low number of Runs Created on three true outcome plays. 

So we can’t do that.  What we can do, however, is place the players in what I call a phatic context.  We put the player into a context of eight more-or-less average players, add the focus player as the ninth, figure the Runs Created by the nine players, and subtract the Runs Created by the other eight players, the phatic context.  (Phatic means more or less "empty of meaning.")  The result is the Runs Created by the Player. 

So we break out each player’s Times on Base, Total Bases, Plate Appearances and Outs Made (a) when the ball was in play, and (b) on Immediate Outcome plays, and then we figure the Runs Created per 27 outs for each group and for each player. 

To cut to the heart of the matter:  Immediate Outcome plays are far, far more productive, on average, than balls put into play.  Historically, 85% of players with 300 or more Plate Appearances have been more productive on Immediate Outcome plays than when putting the Ball in Play.   Most of the exceptions have been the worst hitters in the league.  The Mark Belangers and Mario Mendozas and Mallex Smiths and Yolmer Sanchezes of the world are often more productive when they put the ball in play than when they don’t.  But the Ted Williamses and Babe Ruths and Mickey Mantles and Barry Bondses and Mike Trouts; those guys are VASTLY more productive on three true outcome plays than when putting the ball in play. 

As strikeouts have increased, the percentage of players who are more effective when putting the ball in play HAS risen very substantially, from less than 3% in the  early 1950s to almost 30% now.  Thus, it IS possible that this trend could be putting pressure on players to adjust, and, particularly if the trend continues, it could conceivably lead to changes in how the game is played. 

 

1)      The Study Group

I studied seasons, rather than careers.  I took my spreadsheet encyclopedia of all batter seasons, and eliminated (a) all seasons with less than 300 plate appearances, (b) all seasons with zero strikeouts, since those would all be early and bad-data seasons, and (c) all seasons before 1910, when official counts of strikeouts begin.   This left me with 20,755 player/seasons. Then I eliminated all data that was not relevant to the subject—runs scored, RBI, overall on base and slugging percentage, etc.

 

2)      The Method

I figured runs created for each player on BIP (Ball in Play) and IO (Immediate Outcome) plays, then compared the two.   I counted Hit By Pitch as an Immediate Outcome, and I counted stolen bases and caught stealing as Balls in Play.  Then I figured the percentage of players in each five-year period who were more effective on immediate outcome plays than when the ball was put in play.

 

3)      Conclusions

In the first five years of the study, 1910 to 1914, 79% of players were more productive on immediate outcome plays than when putting the ball into play.  This number, however, began rising immediately; perhaps it was already rising before 1910, although, without strikeout data, we don’t know that.  

Anyway, from 1915 to 1919 this number nudged up to 81%.  With the coming of the lively ball (1920) it jumped suddenly to 91%.  

In the 1920s there were few strikeouts, and the number of home runs was increasing rapidly.  Thus, few players struck out often enough to offset the value of their walks and homers, even if they didn’t hit very many homers.  91% were more productive on walks, strikeouts and home runs than by putting the ball in play. 

In the late 1920s this increased to 94%.  In the early 1930s it dropped back to 91%; in the late 1930s, back to 94%.   In the early 1940s, the War years, it was 95%; in the late 1940s, a higher 95%.   In the early 1950s, 1950 to 1954, 97.3% of players were more productive on Immediate Outcome plays than when putting the ball in play. 

Focusing on that era, Yogi Berra in 1950 created 133.58 runs per 27 outs on Immediate Outcome plays, whereas he created only 3.88 runs per 27 outs when putting the ball into play where fielders could make an out on it.   He hit 28 homers and drew 55 walks, so that’s. . . .what, about 55 runs created on Home Runs and walks, right?   He struck out only 12 times.  So you take the 55 or whatever it is, 59 I guess, multiply that by 27, divide by 12, it’s 134 runs created per 27 outs. 

Nobody else in that era is over 100, but guys like Musial, Kluszewski, Mize, DiMaggio, Jackie Robinson and Ted Williams are sometimes over 50. 

For Balls in Play, everybody is around 4.00.  That’s an overstatement; if a player has a severe uppercut, he gets lots of pop outs and will have a low rate of Runs Created per 27 outs on Balls in Play.  Mark McGwire one year had zero runs created on Ball is Play, although he is the only player who ever did.  If a player has a very high Batting Average on Balls in Play in a season, then he will have a high rate of Runs Created per 27 outs on balls in play, higher than 4.00.  But in that 1950 to 1954 era, the highest rate for any player was 6.97, by Smoky Burgess in 1954, followed by 6.76, Mickey Mantle in 1952.   That’s a LONG way from 50, or 133.58, like Berra.   97% of players in that era were more effective on Immediate Outcome Plays than on Challenge-the-Defense plays. 

      I understand that announcers and coaches and people like to say "Good things happen when you put the ball in play", and they like to say other things like that.  Hitters have always been taught to meet the ball and put it in play.  But if that was actually TRUE, then we wouldn’t be where we are today.   We have wandered into the strikeout-and-home run quagmire because it produces more runs; that’s the cold, hard truth.  

      It’s not as true as it used to be.  By 1954, only 37 players in the history of baseball had struck out 100 times in a season, not counting some 19th century player who supposedly did it, Sam Wise or somebody.  But from 1955 to 1959, our next five-year group, another 16 players were added to the Hall of Infamy, the players who had struck out 100 times in a season.  Jim Lemon struck out 138 times in a season; my God, what is the world coming to? 

From 1960 to 1964, 73 players struck out 100 times in a season.   From 1965 to 1969, 122 players struck out 100 times in a season, with Bobby Bonds reaching 187 in 1969.   These players weren’t even ASHAMED of striking out.   You could tell them and tell them to put the ball in play, and they just wouldn’t listen. 

As strikeouts became more common, the percentage of players who were more effective just putting the ball in play than on Immediate Outcome Plays began to creep up.   From 2.7% in the early 1950s, it moved to 4% in the late 1950s, 10% in the early 1960s, and 15% in the late 1960s.  In the early 1970s, after the mound was lowered in 1969, the percentage of players who were better off putting the ball in play dropped slightly, to 12% in the early 1970s, 14% in the late 1970s, 15% in the early 1980s, a slightly higher 15% in the late 1980s. 

In the 1990s, both halves, 18% of player were more effective putting the ball in play.  From 2000 to 2004 it was a then-record 24%, which dropped back to 19% after the steroid ban in 2005; not sure if the steroid ban caused that to happen or something else.   But from 2010 to 2014, strikeout totals were so high that 29% of players were more effective when putting the ball in play—ten times as many as in the early 1950s.   In the 2015-2019 era, this percentage dropped to 26%, which could be either an aberration, of the beginning of a historic shift.   We have no way of knowing which at this time, although my guess would be that the worm has not yet turned. 

 

      Throughout all of baseball history within the study, 1910 to 2019, players produced an average of 11.42 runs per 27 outs on Immediate Outcome Plays, and 4.01 runs per 27 outs when the ball was in play.   In the years 2015 to 2019, players averaged 7.61 runs created per 27 outs on Immediate Outcome Plays, and 4.27 runs per 27 outs when the ball is in play. 

The runs created on Immediate Outcome Plays have dropped because there are more strikeouts in the mix.  

The runs created on Balls in Play have increased because

(1)  There are more hard-hit balls than there used to be, and

(2)  The effect of there being more hard-hit balls is significantly greater than the effect of defensive shifts. 

 

Of equal relevance here is the standard deviations.   Over time, players averaged 11.4 runs created on Immediate Outcome plays, with a standard deviation of 9.9.  On Balls in Play, they averaged 4.0 runs created, with a standard deviation of 1.3.   In other words, everybody is pretty much the same on balls in play.   There are BIG difference among players in strikeouts, walks and homers.  Ted Williams is a LOT different than Freddy Galvis.  When the Ball is in Play, everybody is sort of the same guy; not EXACTLY, but relatively. 

      In the 2015-2019 era, players averaged 7.6 runs created on Immediate Outcome Plays, with a standard deviation of 4.3.   The standard deviation has shrunk because there aren’t any Ted Williamses and  Joe DiMaggios and Yogi Berras around any more; there are still Mickey Mantles and other superstars, but there is nobody around who is going to hit 46 homers with 37 strikeouts, like Joe DiMaggio did in 1937.  Since EVERYBODY strikes out now, the standard deviation of those plays has dropped.   (Balls in Play in the most recent era, the average is 4.3 with a standard deviation of 1.4.)

 

NAMING NAMES

 

            OK, so the idea that more players would be better off if they just cut down their swing and put the ball in play more often is not yet TRUE, but we are, in fact, moving in that direction.   It is becoming more true than it used to be.  Who are these players who would be better off making more contact?

            The most prominent names are Javier Baez and, before this season, Yoan Moncada.   The only two players in history who have hit 29 or more home runs, but who were still more effective when they put the ball in play than when they got an immediate outcome are:

1)      Javier Baez in 2018, and

2)     Javier Baez in 2019.

Baez in 2018 hit 34 home runs, but created 6.53 Runs per 27 outs when getting an immediate outcome, 6.58 when putting the ball in play.  In 2019 he was at 5.55 when getting an immediate outcome, 5.96 when the ball was in play. 

 

Yoan Moncada was like that in the past, but this year he has cut down his strikeouts and increased his walks, so that is no longer true.  Miguel Sano is like that.  Kris Bryant, Mike Zunino and Trevor (Whats Your) Story. 

Some really good hitters were like that early in their careers, but figured it out:  JD Martinez, Marcell Ozuna, Teoscar Hernandez and Fernando Tatis Jr.  Some guys, like Chris Davis, go the other way; they start out producing good numbers of runs by hitting homers, then wind up striking out so much that eventually you have to feed them to the sharks.  Lots of BAD hitters are in that group, but my point was—and it could have been a mistake to lead with this—but my point is that even some GOOD hitters are in that class now.  But all of the guys like Jake Marisnick, JD Stewart, Adalberto Mondesi, Michael Chavis, Wil Myers. . ..all of those kind of guys would be better off if they could put the ball in play more often.  Whit Merrifield is in that class, but Merrifield is different in that he KNOWS that he is better off putting the ball in play more often, so he tries to do that.   There are a few guys like that, not many.   The guys who come up highly touted, hit .215 for a couple of years and disappear—ALL of those guys would be more productive than they are if they would put the ball in play more consistently. 

 

 

There are a lot of other directions that this research could go in.   What if we considered Home Runs to be Balls in Play; what would the splits be like then? 

But if you do that, you’re just isolating strikeouts and walks from all other plate appearances, which is well-plowed ground.  Or, "What if we split the strikeouts in half, and considered half of them to be unavoidable, then compared players based on the other half?"  

Yeah, you could do that, but that is speculative or theoretical, so if you do that you don’t really emerge with any actual knowledge that you didn’t have before.  The point of research is come out of it knowing something you didn’t know before.  There is lots of other research that could be done in this area.  Don’t tell me I should do it; I’m a busy man.  

 
 

COMMENTS (7 Comments, most recent shown first)

mikefredo1
Guys like Baez would probably find more success by simply not swinging as hard (which, for Baez, is still extremely hard) .... assuming a batter sacrifices bat control for bat speed, and vice-versa: if a batter cuts down on his bat speed by, say, 10%, he lessens the likelihood of hitting a home run by a small value, but increases the likelihood of hitting a double or single (as opposed to a warning track pop-up) if he fails to square up the ball perfectly. In other words, the hitter increases his margin for error. After all, you score the same amount of runs for hitting a 400 ft. home run as you would if you hit a 500 ft. home run. This idea is probably statistically immune - just food for thought.
3:32 PM Sep 21st
 
abiggoof
Three true outcomes is faulty because of homers. Contact or no contact makes sense; swing or no swing, too. But TTO falsely groups opposite extremes.

Still, the value over time part was interesting.
2:36 PM Sep 20th
 
Joe_Start
"We have wandered into the strikeout-and-home run quagmire because it produces more runs; that’s the cold, hard truth." - Bill James

I think people are willing to accept that because that's they've been told more runs are good for the game. I kind of get that - the more runs you score, the better chance you have of winning.

But I would be happy to see fewer runs and more action on the bases. If every team did this, it wouldn't really change anything. The best teams will still win the 5-3 games as much as the win the 9-6 games.
5:29 AM Sep 18th
 
OBS2.0
Here it is -- thanks to jgf

boards.billjamesonline.com/showthread.php?16784-How-Do-You-Solve-A-Problem-Like-Javier-a&highlight=baez

Baez is the only MLBer since 1901 to have three games in his career with 5 Ks in 5 plate appearances. In fact, only four guys have done it twice.
3:03 AM Sep 18th
 
OBS2.0
According to the Mets announcers,, the team leaned heavily on Baez about a week ago to be more selective, and in his last 8 games he has walked 7 times while striking out only 4 times, batting over .400 in that span. To put that in perspective, in the preceding 8 games he had walked zero times and struck out ten times. Year to date to that point he had walked only 18 times and struck out 164.

Baez got some bad press....well he has got a lot of bad press, but he got some bad press on this topic when he struck out five times in five plate appearances in his sixth game with the Mets. It was the second time he had done that in his career, 5 Ks in 5 plate appearances. That was.....I forget the superlative. He may have been the only guy to have done that since divisional play, something like that.

At any rate, the correlation the Mets announcers are making is that by not swinging at balls two feet outside and in the dirt, he is getting pitches to drive. It was said that the coaches showed him the book that they had on pitching him, and that made the light bulb come on.
2:38 AM Sep 18th
 
FrankD
Interesting study. I assume that one of the adjustments for putting ball in play is not to upper-cut. What about speedsters like Willie Wilson. Its often reported that when he slapped at the ball and ran he was an effective hitter but later in his career he tried for HRs he'd hit deep flies. Was Willie more valuable when he was a slap-hitter? Do you have balls-in-play data for Willie Wilson or Ricky Henderson?
6:28 PM Sep 17th
 
evanecurb
This feels like the beginning of a whole new subset of research. Or maybe I'm just being hopeful because I like to see the ball in play. Here's an idea: could we study this matter in different ballparks? Is it possible that Johnny Ballgame is more effective putting the ball in play in a big ballpark like Petco or Citi Field, but in a home run park such as Fenway or Citizens Bank Park, he gets more value from immediate outcomes? I would think the park factors would have to be pretty disparate in order for this to make a difference with most hitters, but we never know until we look at the data. Another question: are groundball hitters, as a group, better off putting the ball in play or on the three true outcome plays? What about flyball hitters? High strikeout/low walk/medium power guys? High strikeout/medium walk/medium power guys?
5:16 PM Sep 17th
 
 
© 2011 Be Jolly, Inc. All Rights Reserved.