Pete Crow-Armstrong Is Too Good Of A Hitter To Keep Bunting So Much

[matto1233]

Image courtesy of © Katie Stratman-USA TODAY Sports

Last Wednesday, with two on, two out, and the Cubs trailing by two in the bottom of the sixth, Pete Crow-Armstrong stepped to the plate. This was a fantastic opportunity for the North Siders, who were going for the sweep against the Miami Marlins, to scratch a few runs across the board and get back in the ballgame. Except that Crow-Armstrong, who, through Saturday’s games, has been the 19th best qualified hitter in baseball by wRC+, bunted. He did succeed in doing so, despite it being a terrible bunt, thus loading the bases for Nico Hoerner.

For what it’s worth, and you likely already know this, Nico Hoerner struck out to end the inning, and the Cubs went on to lose the game, 3-1. The end result is somewhat irrelevant. Hoerner could have hit a grand slam there, and I’d probably still be writing this article. For me, the ends don’t always justify the means. In this case, they did. I am here to talk about one of the Cubs’ best hitters deciding to bunt with two on and two out while down two runs. 

I have been here before with the young center fielder. On August 1 of last year, I wrote an article pleading with him to stop bunting so much. Through that point of last season, Crow-Armstrong was hitting just .222 on 13 bunt attempts, which was the worst mark in baseball of anyone that had put down at least 10 bunts. 

He has been considerably better since. He had a .625 batting average on bunts in August and September of last season, according to Baseball Savant. He also had four sacrifice bunts, which of course, don’t count against his batting average. I’m not sure it’s fair to ever give Crow-Armstrong a sacrifice bunt, because every bunt he lays down is an attempt at a hit. If it ends up as a sacrifice, that is simply a positive consequence. That being said, I’ll be nice and not knock him for it. 

This year, he’s hitting .400 on five bunt attempts, with three sacrifices. Hooray! I would have been thrilled with that last season. Now, I'm not, mostly because Crow-Armstrong is a good hitter these days. A flawed one, sure, but the current results can’t be argued with.

Back to Tuesday night’s decision to bunt. In deciding to bunt, he took away any chance that he had to drive in a run. The best-case scenario that he created for himself was for the Cubs to end up with the bases loaded and two outs. Using the Win Probability Added (WPA) Inquirer at FanGraphs, we can calculate the average WPA that Crow-Armstrong was providing by bunting, or by swinging away. 

For example, if he bunts, we can assume that would result in a single 40 percent of the time, and an out 60 percent of the time, since he is hitting .400 on bunts this season. The Cubs would go from a 24.7 percent chance of winning with runners on first and second with two outs, to a 29.9 percent chance of winning with the bases loaded and two outs, a 5.2 percent difference. He has a 40 percent chance of creating that 5.2 percent difference, so if we multiply those two together, we get 2.1 percent. 

But he also has a 60 percent chance of making an out, and in this case, an out would drop the Cubs’ win probability by 6.6 percent. Multiply those together, and we get -4 percent. If we subtract that from our first number, we get -1.9 percent. This means that, with Pete Crow-Armstrong squaring to bunt in that situation, he is subtracting 1.9 percent from the Cubs’ win probability, on average. 

Now, let’s run that same calculation, for this particular scenario, based on his full season statistics. Let’s also assume a single scores one run, and a double scores two:

Event	% chance of happening	WPA added or lost	Average WPA added or lost
Single	13.51%	13.4%	1.8%
Double	5.95%	30.6%	1.8%
Triple	0.54%	31.4%	0.2%
HR	6.49%	45.9%	3%
BB/HBP	5.41%	5.2%	0.3%
Out	68.11%	6.6%	-4.5%
Total			2.6%

Adding all of that up, and we can see that an average plate appearance from Pete Crow-Armstrong in that situation would result in 2.6 percentage points of win probability added. A much better figure than the average bunt, which as we saw above, would result in a 1.9 percentage point loss in win probability. 

Is this logic perfect? No, and I can mention the variables that this does not address, some of which support the bunt, and some of which do not:

1. The next batter was Nico Hoerner, who is a decent player and has been a roughly league average hitter for several seasons now. He is also one of the Cubs’ worst hitters. It’s not like Kyle Tucker was due next. 

2. A single will not always score one run, and a double will not always score two. 

3. The pitcher was Calvin Faucher, an above-average relief arm. Perhaps Crow-Armstrong didn’t like the matchup. 

4. There is some very small chance that the Marlins throw that ball down the right field line and the Cubs score two runs on the bunt, especially given the weather conditions at the time. I am assuming there is a zero percent chance of that happening in the above calculations.

I don’t think any of that is bridging the gap between average WPA on a bunt vs. swinging away, though. Ironically enough, Pete Crow-Armstrong stepped to the plate with two on and two outs in a tie game on Friday afternoon against the White Sox. He swung away, and deposited the ball into the right field bleachers, giving the Cubs a 5-2 lead that they would never look back from, adding 26% to the Cubs’ win probability.

Hopefully he learned his lesson. Pete Crow-Armstrong is just a good hitter now. Period. He’d do well to remember that.

View full article

[Jeff Alson]

I enjoyed reading this well written article, but I want to add some points. You mention 4 variables that were not included in the quantitative analysis, but failed to point out that Hoerner has been the Cubs' best hitter with RISP, over .400 the last time I checked, higher than PCA. This means that all 4 of the omitted variables favored the bunt. And there was a 5th omission-- because there were 2 outs and because PCA has been red hot, the 3B was playing much deeper than usual for PCA (which is why his terrible bunt was successful!), so his expected success rate was higher than .400. We cannot quantify all these variables, but my guess is that the choice to bunt or not was fairly neutral. And I like that PCA was even willing to consider a " small ball" move and not let his decision be dictated by fan chanting or ego. That bodes well for the future decisions.

[Brock Beauchamp]

3 minutes ago, Jeff Alson said:

I enjoyed reading this well written article, but I want to add some points. You mention 4 variables that were not included in the quantitative analysis, but failed to point out that Hoerner has been the Cubs' best hitter with RISP, over .400 the last time I checked, higher than PCA. This means that all 4 of the omitted variables favored the bunt. And there was a 5th omission-- because there were 2 outs and because PCA has been red hot, the 3B was playing much deeper than usual for PCA (which is why his terrible bunt was successful!), so his expected success rate was higher than .400. We cannot quantify all these variables, but my guess is that the choice to bunt or not was fairly neutral. And I like that PCA was even willing to consider a " small ball" move and not let his decision be dictated by fan chanting or ego. That bodes well for the future decisions.

Welcome to North Side Baseball!

[Rcal10]

15 minutes ago, Jeff Alson said:

I enjoyed reading this well written article, but I want to add some points. You mention 4 variables that were not included in the quantitative analysis, but failed to point out that Hoerner has been the Cubs' best hitter with RISP, over .400 the last time I checked, higher than PCA. This means that all 4 of the omitted variables favored the bunt. And there was a 5th omission-- because there were 2 outs and because PCA has been red hot, the 3B was playing much deeper than usual for PCA (which is why his terrible bunt was successful!), so his expected success rate was higher than .400. We cannot quantify all these variables, but my guess is that the choice to bunt or not was fairly neutral. And I like that PCA was even willing to consider a " small ball" move and not let his decision be dictated by fan chanting or ego. That bodes well for the future decisions.

While I agree with everything you said, I don’t think bunting there was the right move. He should have been swinging. That said, I don’t think PCA attempting to bunt for a hit 5 times this year is a lot. It depends when he bunts. If no one is on base, bunting for a hit is a good idea. Doesn’t matter how many outs there are. If he can bunt to a .400 average that’s better than his actual BA. I realize bunting only gets him to first base, while swinging can get him extra bases. But if he does get a hit in a bunt he can always steal second base. Plus by bunting some, that puts the corner infielders in a little. Hell , might even put the 2nd baseman and SS in a little. Which gives him a better chance to get a ground ball through. As I said, the particular AB being discussed in the article, I agree bunting was not the move. But bunting should be a part of his game and with only 5 attempts this year, I don’t think he is bunting too much. 

[matto1233]

2 hours ago, Jeff Alson said:

I enjoyed reading this well written article, but I want to add some points. You mention 4 variables that were not included in the quantitative analysis, but failed to point out that Hoerner has been the Cubs' best hitter with RISP, over .400 the last time I checked, higher than PCA. This means that all 4 of the omitted variables favored the bunt. And there was a 5th omission-- because there were 2 outs and because PCA has been red hot, the 3B was playing much deeper than usual for PCA (which is why his terrible bunt was successful!), so his expected success rate was higher than .400. We cannot quantify all these variables, but my guess is that the choice to bunt or not was fairly neutral. And I like that PCA was even willing to consider a " small ball" move and not let his decision be dictated by fan chanting or ego. That bodes well for the future decisions.

Thanks!

Personally, I’d tend to chalk up Nico’s RISP batting average as sample size that would average out over time. However, I had no idea he was hitting that well with RISP, and you make fair points!

[matto1233]

2 hours ago, Rcal10 said:

While I agree with everything you said, I don’t think bunting there was the right move. He should have been swinging. That said, I don’t think PCA attempting to bunt for a hit 5 times this year is a lot. It depends when he bunts. If no one is on base, bunting for a hit is a good idea. Doesn’t matter how many outs there are. If he can bunt to a .400 average that’s better than his actual BA. I realize bunting only gets him to first base, while swinging can get him extra bases. But if he does get a hit in a bunt he can always steal second base. Plus by bunting some, that puts the corner infielders in a little. Hell , might even put the 2nd baseman and SS in a little. Which gives him a better chance to get a ground ball through. As I said, the particular AB being discussed in the article, I agree bunting was not the move. But bunting should be a part of his game and with only 5 attempts this year, I don’t think he is bunting too much. 

I do think that bunting could be a legitimate part of his game. It needs to be applied in better situations than this, though. Which I know you’re agreeing with anyway. 

A couple of points on him bunting, in general:

1. He has laid down eight total bunts this season, but three of those went for sacrifices. So technically a few more than five, but only five count against his batting average. 

2. If it is going to be a legitimate part of his game, he needs to be better at it. Even if he maintains the .400 batting average, that would put him in a three way tie for 10th out of 15 hitters who have laid down at least five bunts. The league as a whole is hitting .514 on bunts.