Blue Jays and the Unloved Strikeout: Why a K can be OK

Strikeouts are often perceived as worse than other outs.  Are they?  And how should the answer to this question affect the Blue Jays’ thinking?

Featured Image Courtesy Of DaveMe Images. Prints Available For Purchase

For a hitter, strikeouts are evil.  You did not advance the runner, or give the fielders a chance to make an error.  Or the ball could disappear into a drainage hole, or lose its cover, or the fielder could be attacked by a gang of seagulls.  All these opportunities are lost when a player strikes out.

Of course, the catcher could drop the third strike and the batter could reach first.  But that is rare – far more rare than a fielding error on a ground ball.  True, that ground ball or liner to an infielder could also result in a double play, but double plays are also relatively rare.  Teams turn double plays, on average, in only 10-12% of potential double play situations.

So what *is* the answer?  Does a strikeout hurt a team more than a different type of out, on average?  And how would you answer that question?

Way back in 1963, an ahead-of-his-time baseball statistician named George Lindsey created a tool called a “run expectancy matrix”.  That matrix (also called a RE24 matrix) shows the average expected number of runs for each combination of outs and men on base.  So, for example, using a matrix based on data from 2010-15, when a team has a man on first with nobody out, they will on average score 0.859 runs.  If the next player strikes out, the expected runs (man on first, one out) decreases to .509.  If, instead of striking out, the second player grounds out but advances the runner to second, the expected runs are .664.  And if the second hitter grounds into a double play, the expected runs (2 out, nobody on base) decrease to .098.  So advancing the runner increases expected runs by .155 (.664-.509) but a double play costs almost three times as much (.509-.098=.411)

In their book “The Book – Playing The Percentages In Baseball” (otherwise known as the Statsgeek Bible) Tom Tango, Mitchel Lichtman and Andrew Dolphin (“the Gurus”) did the math.  For each possible combination of outs and men on base, they calculated the effect on expected runs of a strikeout and of a non-strikeout out based on thousands of actual plays.  Their finding are summarized in this table.

Most of their findings are highly intuitive.  When there is nobody on base, it does not matter – an out is an out.  And when there are two outs, it also does not matter – an out ends the inning, regardless.  They found that a strikeout was better than a non-K out with runners on first, or first and second, due to the higher risk of a double play.  But a ground ball or fly ball out is significantly better than a strikeout with a man on third.

It would be easy to take the average of the plusses and minuses in this table.  But that would be misleading.  It is far more common for a player to come to bat with the bases empty – or with a man on first – than it is to come up with the bases loaded.  And hitters in different spots in the batting order frequently face different on-base situations with different frequencies.  So the Gurus took it one step further, and calculated the average difference between a strikeout and a non-K out for each position in the batting order.  Their finding are summarized in this table.

The bottom line?  The effect on expected runs of a strikeout is basically the same as a non-strikeout out.

The Book was published some years ago (in 2006).  The results are generally still valid, but there have been developments since then that make strikeouts potentially even more attractive (or at least less unattractive?).  The first such development is pitcher usage.  In 2006, starting pitchers in the American League pitched 13,224 innings.  By 2019, that figure had decreased almost 10%, to 12,125 innings.  A strikeout takes, on average, 50% more pitches than an out from a ball in play.  Running up a starter’s pitch count in 2021 is, if anything, even more important now than it has been in the past.

The second change is in the number of home runs.  Ten years ago, the average home runs per team per game was 0.94.  In 2019, it was 1.23 – an increase of 30%.  As the number of home runs increases, the value of advancing the runner decreases and the negative impact of eliminating a runner (through a double play) increases.  This makes groundouts even less attractive than they were in the past.

So why should the Blue Jays care?

The Jays will be facing decisions, both with existing players and with potential free agents and trade targets, which involve players with above-average strikeout rates.  How much weight should the Jays put on this statistic?

As for example – it is entirely possible that the Jays will be looking to sign a good young free agent third baseman this offseason.  One potential candidate is Javier Báez.  Javy ticks a lot of boxes – he is young (he will play 2022 at 29 years old), a good-to-elite defender (+14 career DRS/1200 at third, +8 at shortstop), and his 13.4 fWAR from 2018-21 is 26th highest in baseball.  But his 2021 strikeout rate of 33.6% is more than 10% higher than the league average.   Should this K% be a deal-killer?  Or would it be more appropriate to look at his 2021 on-base percentage of .319 – which is above the AL league average of .316 – and say that it doesn’t really matter how he gets there?

The bottom line

As Babe Ruth once put it, “I’ve never heard a crowd boo a homer, but I’ve heard plenty of boos after a strikeout.”  And he was right.  But the Bambino held the record for career strikeouts – 1,330 – for 30 years, until it was surpassed by Mickey Mantle in 1964.  And Babe also said that a player should “Never let the fear of striking out keep you from coming up to bat“.    If one of the greatest players of all time was not afraid of Ks, perhaps fans – and the Jays – should not be as well?