Buck Showalter screwed up, but so did Dan Duquette

There are already countless articles justifiably criticizing Buck Showalter for leaving Zach Britton, one of the best relievers in baseball, to rot in the bullpen while his team went through multiple high leverage situations on their way to elimination. Since so many have gone into detail on Showalter’s faulty strategy, I’ll cover it only briefly.

There were lots of opportunities where Britton could have helped, but the time where Showalter absolutely needed to bring him in was the bottom of the ninth inning. It was a tie game with a murderer’s row coming up in Josh Donaldson, Edwin Encarnación, and José Bautista, either of whom could easily end the game with one swing. To be fair, Brad Brach is no slouch himself, and he had the platoon advantage. This season he had a 2.62 RA9, 2.92 FIP, and struck out 29.6 percent of batters he’s faced. If we look at the leverage index for the game and how those numbers are qualitatively described, we can see that the beginning of the inning qualified as a high leverage situation. Once Bautista came to the plate, it be came a very high leverage situation. (Yes, I know fancy stats aren’t needed to prove that it was a high leverage situation, but this is a saber-slanted site.)

When Bautista did step up in the ninth inning, what the Orioles needed more than anything was a double play, yet the pitcher with the 80 percent groundball rate still wasn’t put in the game. Thankfully, Darren O’Day got the double play they needed anyway. However, O’Day’s groundball rate has been only 34.8 percent since the start of 2015. The Orioles and Buck got lucky.

The game ended because Showalter believed it would be a good idea to put in Ubaldo Jiménez, he of the 5.88 RA9 and 4.43 FIP this season. Jiménez’s delivery is so erratic and unorthodox that it’s impossible to repeat, which is why he’s so bad and inconsistent. He got no outs and only threw four pitches: Two balls, a single, and a home run. Game over. You can debate whether Brach or Britton makes more sense for the 9th, and whether O’Day was a worthy choice to try to get the double play. Showalter had a ton of options, but there’s a huge chasm between the ones that involve Zach Britton, the best reliever he had and possibly in all of baseball, throwing any number of innings at all, and the ones that don’t. If he took a line in the former group, Showalter couldn’t have screwed up too bad. He took a line in the latter, though.

I believe that Showalter is a good manager, but when pressed on his decision making he made the usual excuses you hear from managers. Britton was held back because of teh savez. Jiménez was put in because he had a 2.82 ERA in the second half, as if he was a completely different pitcher than the first half, or the rest of his career. Buck weighed Jiménez’s 60.2 most recent innings pitched over the 938.2 previous ones where he had a 4.69 ERA. There’s almost no way that Britton is a 0.94 RA9 pitcher by true talent, but even the most pessimistic view of his true talent would have him as a far, far better option than Jiménez in a high leverage situation.

It’s important to make clear that even if Showalter’s choices “worked,” as they did with Brach and O’Day in the ninth, he still would deserve to be criticized just as much. Evaluate the process, not the result. This is especially important in a sport as random and variable as baseball. You really don't need analytics to tell you to use your best pitcher when the game, and the season, are on the line.

Had Britton pitched the ninth and blew the game, well, at least you went down with your best pitcher on the mound, instead of waiting for a save opportunity that never came. I doubt anybody would’ve criticized Showalter then. (No one is criticizing Terry Collins for going out with Jeurys Familia on the mound.) Orioles fans would’ve probably slept better last night, too.

It’s amazing how managers don’t learn from their own tactical mistakes or those of others. Showalter made a similar mistake 21 years ago. In Game 5 of the 1995 ALDS, he left Jack McDowell in with a one-run lead in the bottom of the 11th instead of turning to their closer, John Wetteland, because he gave up a grand slam in the previous game. That's going by one batter, even worse than going by 60.2 IP. The Yankees promptly lost and were eliminated. More recently, we had Fredi González letting Craig Kimbrel waste away in the bullpen while the Braves got knocked out of the 2013 NLDS.

I don’t believe that Showalter deserves to be fired because of his blunder. I do believe, however, that he deserves to be called into GM Dan Duquette’s office and be excoriated over his decision making. If I were a GM, I’d be very disturbed that my manager is valuing saves and small sample sizes in 2016. If I were Peter Angelos, I would also be calling Duquette into my office and asking him why his direct report of over six years was managing to a save and making decisions based on small sample sizes.

The paragraph above is how most work places function. If an employee screws up or is performing poorly, it makes his or her boss look bad. The boss will correct the direct report, and if the employee fails to improve, he or she will be terminated. No bosses let their direct reports do whatever they want, because the bosses will then get in trouble with their own bosses. Yet this seems to be how GMs treat managers. They might advise them, but they never seem to correct them for poor lineup construction or poor in-game tactics. It’s likely because the managers resist. That’s not the way the manager/direct report relationship is supposed to work, or the way it works in almost any other field. You do what your boss tells you to do, or you find a new job.

Baseball has seen more than its fair share of poor in-game managers, but their GMs, the guys responsible for hiring them and managing their performances, tend to receive little to no criticism for their in-game failures. Again, in most work places, a boss is responsible for making sure that his or her direct reports receive the training they need to do their jobs properly. So why don’t managers receive training in game theory? Or at least an expert to consult with in the dugout? There’s little a GM can do to improve a manager who lacks coaching and personnel management skills, but teaching basic tactics should be very doable.

One game, even one as important as a Wild Card game, doesn’t mean that Showalter wasn’t a great hire. He was and still is. Just like Jiménez’s 60.2 recent innings don’t really move the needle on our evaluation of him, Showalter’s most recent game shouldn’t really change what we think of his career. However, when a poor manager is fired, the GM involved seems to rarely get criticized for hiring a poor employee. Matt Williams was as bad a manager as I can remember, but I don’t recall Mike Rizzo getting much or any criticism for the hire. In another line of work, he likely would’ve gotten chewed out by his boss for such a poor hire.

I find it quite bizarre how disconnected GMs and managers are about how they think about baseball, especially considering how often they interact. Front offices have become incredibly smart in recent years, but managers have yet to catch up. And Showalter is one of the smarter managers in the game! He is hardly alone when it comes to managing to a save.

GMs clearly understand that the save as a metric is worthless. Ironically, Dan Duquette is a great example of this. He traded Eduardo Rodríguez for Andrew Miller in 2014, even though Miller had never gotten a save. The Yankees then gave him a 4-year, $36 million deal. Duquette gave O’Day a 4-year, $31 million deal last winter, even though he has only 14 career saves.

If managers ever improve their in-game tactics, it’ll happen because their GMs forced them to do so. During the interview process, they should make it perfectly clear that the candidate will be trained in proper in-game tactics and will be strictly held to those standards. If the candidate doesn’t like the sound of that, then he is free to decline the job. Perhaps some of the bigger names will do just that, but the fact of the matter is that there are only 30 manager jobs in baseball. Some qualified candidates will eventually step up and step out of their comfort zones to learn how to do an important part of their jobs properly, and their teams will benefit as a result.

Orioles fans have every right to be mad at Showalter. But don’t let Duquette off the hook. He’s the one who either failed to provide the training to his direct report of over six year in how to manage games, or failed in getting him to apply the training. In almost any other line of work, Duquette would’ve gotten in big trouble with his superiors. It’s time to start criticizing GMs of bad tactical managers just as much as we criticize the managers themselves. Why they choose to enable such managers when their mistakes reflect poorly on them, and why the media never criticizes them for doing so, I’ll never understand.