Should a Player Ever Just Take a Strikeout?

As you hear Len Kasper say at the end of the video above, “the old adage is, if you think you’re gonna hit into a double play, strikeout”. This is something I’ve heard being joked about before, but I wanted to see if there was actually something to it. Is there a situation in which a manager should tell a player to not swing? Just take the strikeout. The situation I looked at is bases loaded, with 0 or 1 out and the pitcher batting. I just felt like pitchers tended to hit into so many rally killing double plays in this situation. Therefore, I believed it may actually be better to just accept the strikeout and let the lead off hitter get his shot.

To begin, this question doesn’t come about unless pitcher hitting is really awful, and it is really awful. Just look at the graph above if you need any proof. The graph shows the average batting average at each position over the last 20 years. Every position is clumped together and then there’s pitchers. They just can’t hit at a major league level. Not only that, but they were worse than ever in 2017 and they are even worse in 2018 so far.

For answering the main question here, I looked at data with this exact criteria from seasons 2013 to 2017. This situation (bases loaded, 0 or 1 out, pitcher batting) has happened 331 times over that time. 43 have resulted in hits and 54 have resulted in grounded into double plays. Pitchers are hitting .143 in these situations.

I calculated the positive win probability by using the formula below. Win probability, also referred to as win expectancy, is the percent chance a particular team will win based on the score, inning, outs, runners on base, and the run environment. This is calculated based on historical data. I did this to get the win probability added from each at bat.

Win Probability Added = Win Probability after outcome (1B, 2B, 3B…) – Win Probability of Situation (bases loaded no outs)

Let’s call these the positive values. I did the same with negative values (double play, strikeout, ground out- fielder’s choice, pop out or fly out- not Sac Fly). All of these negative values except a double play, advance no runners, make one out, and decrease the win probability at the same rate. I considered both “home to first” and “second to first” double plays.

I then took then took the percentage each outcome occurred in this situation. For example, singles happened 12.33% of the time in this situation. I did this in order to weigh the algorithm for how often each outcome occurs in MLB pitcher at bats with the bases loaded. I multiplied this number by the win probability difference I found previously for each difference outcome. I then added all the values to create a total score of whether or not the at bats on average were positively affecting the team.

I repeated this same process, but with the “bases loaded, one out situation” for pitchers. I also evaluated the situations of “bases loaded one out” and “bases loaded two out” with all lead off hitters in these situations from 2013 to 2017. This accounts for what would happen after the pitcher were to strikeout if he didn’t swing at any pitches.

Okay, now let’s look at the results and compare them to the question I asked. The point is to see if the batter following the pitcher actually has a better chance of increasing the team’s chance to win than the pitcher does with one less out. Let’s look at the chart below of the results from my algorithm. The probabilities I used for the table below are based on first inning probabilities, but the same conclusions can be drawn from each inning’s data.

Outcome Expectancy

There are no units since these numbers come from my algorithm, I made that up so there is no proper unit. What is important to see is the comparisons and whether they are positive (likely to increase team’s chance of winning) or negative (decrease team’s chance of winning). We actually see something I didn’t expect. It’s better off with “no out bases loaded” for the pitcher up to take the strikeout. However, with 1 out and the pitcher up, you might as well let him hit. There is a much larger chance that you don’t score with the lead off hitter up and 2 outs than the pitcher up with 1 out. I attribute this to how important it is not to have two outs. With two outs in an inning, run expectancy drops tremendously. In addition, if you don’t swing with the pitcher with one out, there is no chance you score on sac fly. You also cannot score on a sac fly with your lead off hitter since there are then 2 outs.

This is a general rule of thumb to use, but of course game situations and the hitting ability of the batter will influence decision-making. Better hitters should be allowed to swing away more. Only two pitchers in the data set hit a grand slam: Madison Bumgarner and Travis Wood. Also, I know Reds pitcher Anthony DeSclafani did hit a grand slam off the Cubs last week, but it was in a two out situation in which the pitchers have nothing to lose and might as well swing away. Guys who show they can hit should swing away still, but majority of pitchers should abide by these rules.

The other idea for pitchers is to bunt in this situation. That can absolutely work if you have a pitcher who is a skilled bunter. Jon Lester has done this before and is a very good bunter. It’s better off to just bunt with him in this situation. In the video we see Lester execute a bases loaded sacrifice bunt to perfection. 

However, if you do bunt you could risk a double play or a fielder’s choice at home and your pitcher then has to run the bases. 

I was curious about this whole situation and was surprised that striking out with one out and the pitcher up wouldn’t be the better option, but it is with no outs. I understand some people will counter that with that players want to compete and won’t want to just strikeout, but I’m just saying with no outs and most pitchers up, it gives the team a better chance if they strikeout. Obviously teams need to know their own pitchers abilities when making decisions and this theory should not be set in stone. This situation doesn’t come up too often, but it’s still interesting to see how each team handles it.