clock menu more-arrow no yes mobile

Filed under:

Mike Matheny and pitcher wins

Why does Mike Matheny repeatedly make what appears to be a tactical mistake, all for the sake of an individual player statistic, and what can be done to make this less of an issue going forward?

Jeff Curry-USA TODAY Sports

For baseball fans, second-guessing the manager has become one of the game's timeless traditions. When a questionable move doesn't work out, fans often take their frustrations out on the manager and accuse him of incompetence. At times, these criticisms are examples of hindsight bias, where a move is only questioned after it has already failed. Many people do not realize that the manager cannot see into the future and know how a certain move is going to turn out. The best a manager can do is make a move that gives his team the best chance of winning, and oftentimes, the difference between one move versus another is negligible. Sometimes, a bad move can have a good result, and a good move can have a bad result.

With all of that being said, our site has, at times, been known for being critical of Mike Matheny's in-game strategy, and I feel comfortable in saying that much of this criticism is justified, as our writers do their best to avoid logical fallacies like hindsight bias by using objective analysis. (Occasionally, we will even defend Mike Matheny when fans question a decision that doesn't work out.) One of Matheny's most frustrating habits is leaving a starting pitcher in a game longer than he should, often because he wants that pitcher to have a chance at a pitcher "win". On the surface, this type of move should seem ridiculous. A manager's number one priority should be for his team to win the game at hand, even if that means that the starting pitcher doesn't get credited with a win. With that being said, there are other factors that can influence the decision of when to pull a starting pitcher. For example, a manager will be more likely to pull a starting pitcher when he has a high-quality, deep, and rested bullpen at his disposal. On the flip side, amanager may stick with a starting pitcher longer than he normally would if the bullpen has been overworked in recent games.

Still, there have been multiple instances of Matheny leaving a starting pitcher in a game longer than he should, where the only plausible explanation is that he wants the starting pitcher to have a chance to win the game. I'm talking about situations where the bullpen is rested, the starter is well past 100 pitches (and is visibly losing effectiveness), and the team is either tied or down by a small margin. This was exactly the scenario Tuesday night, when Mike Matheny gave Lance Lynn multiple opportunities (three to be exact) to get the third out of the 7th inning, even though he looked gassed. First, Lynn nearly walked pinch-hitter Madison Bumgarner before allowing him to reach base on a clean single. Then, Lynn stayed in the game and gave up back-to-back walks to Gregor Blanco and Matt Duffy. With the bases loaded and his pitch count at 117, Lynn was finally pulled from the game. (One of the runners would end up scoring.)

In the context of one game, this type of move makes very little sense, since a fresh relief pitcher would likely have a better chance of retiring one batter than a worn out starting pitcher would, especially if that starting pitcher is facing hitters for the fourth time. If I were to read Mike Matheny's mind, though, and find his rationale for making such a decision, I imagine that it would include the following ideas.

  • showing faith in a pitcher boosts his confidence, which will lead to positive results in the future
  • giving a starting pitcher a chance to boost his stats with a win will make him happier, leading to good clubhouse chemistry and the belief that the manager cares about his players individually
I can't really speak much on either of these ideas, since terms like "confidence" and "clubhouse chemistry" are abstract concepts that are difficult to pinpoint. With that being said, it seems pretty clear that Mike Matheny is a manager that players respect and love to play for (a "players manager", so to speak), and I don't think it is too much of a stretch to claim that this good player-manager relationship could lead to some sort of positive impact on the field.

If we assume that player confidence and clubhouse chemistry are real things that a manager should take into consideration when making an in-game tactical decision, then the question that should be asked is this. Is it possible make in-game decisions that make sense statistically without having a negative impact on player confidence and clubhouse chemistry?

While I am no expert in areas of player confidence and clubhouse chemistry, I do think that it is possible to make decisions that make sense statistically without having a negative impact on the team because of hurting a starting pitcher's feelings by not giving him a chance at a win. To me, the biggest issue right now is that players put too much value in the pitcher win, which is a pretty terrible statistic overall.

There are many things that can lead to a starting pitcher getting a win, and a lot of them are outside of that pitcher's control. All he has to do is pitch five innings and leave the game when his team is ahead for good. A pitcher will usually end up with more wins if his teams scores runs when he starts, his defense plays well behind him, and his team's bullpen doesn't blow the lead after he leaves the game. A pitcher can help his cause by doing a good job of preventing runs, but that that does not necessarily guarantee that he will be the winning pitcher. Take Corey Kluber. He has a win-loss record of 8-13, despite posting a 2.80 FIP, 2.93 xFIP, and 4.9 fWAR in a league leading 186 2/3 innings. On the flip side, Drew Hutchison is 12-2 for the Blue Jays and was just sent to Triple-A.

Clearly, it does not make sense to put a lot of stock in pitcher wins when evaluating an individual player. Given the amount of information and statistical knowledge in the game of baseball today, we really don't have much of a need for such a statistic. While some may argue that a good pitcher win-loss record is an indicator of several important factors (high number of innings, good run prevention, pitching in clutch situations, contributing offensively, etc.), each of these factors can be broken down and measured individually, free from all the statistical noise that is present in the pitcher win statistic. I would be perfectly happy if baseball analysts took this kind of approach to pitcher wins.

With that being said, do players still have a good reason to value a flawed statistic like the pitcher win? It doesn't appear to be as big of a factor in Cy Young voting as it once was. Back in 2012, Joe Posnanski wrote an article titled "Pitcher wins don't determine Cy Young anymore," and since then, we have seen more examples of Cy Young winners not having the best win-loss record.

I would also imagine that teams don't value pitcher wins as much anymore when evaluating players and giving out contracts. In this day and age, teams do everything they can to gain an edge on their competitors by finding statistical measures with predictive value, and I'm sure that most front office people today understand the relative worthlessness of pitcher wins relative to other statistics. If this is actually the case, then I think it is safe to say that having extra pitcher wins probably won't lead to a higher salary going forward (assuming all other statistics are the same).

One area where win-loss record has not necessarily lost a lot of influence is All-Star voting. This season, we saw Clayton Kershaw and Corey Kluber left off the roster (at least initially, in Kershaw's case) despite being in the top five in all of baseball in FIP, xFIP, fWAR, and K%. Kershaw had a 5-6 record while Kluber was 4-10. One would think that the people who vote pitchers onto the All-Star team (players and managers) would have an especially good understanding of who the best pitchers are, since they manage and play against them on a daily basis. With their special knowledge of the game (which can supposedly only be understood by people inside the game), they should have a better understanding than we could ever have of who the best pitchers are, right?

If we are seeing a shift in the way the pitcher win is valued (and I would argue that this has happened throughout front offices and, to a lesser extent, the media), it has not happened in the minds of many players and managers throughout baseball. Perhaps teams need to be more proactive in educating their players (and managers, in some cases) about the statistics that they should really pay attention to. The Dodgers appear to be taking this approach under the direction of new farm director Gabe Kapler. Back in June, Pedro Moura of the Orange County Register wrote a piece about how Kapler was working with young players to understand statistics like exit velocity, wRC+, and strikeout-to-walk ratio, which are good predictors of future performance.

Here is a good excerpt from that article:

"Players are going to get bent out of shape about numbers," Kapler said this week. "It's inevitable. It's always going to be that way. We just want them to get bent out of shape about the right ones....It's fine if you want to mope around and get depressed because you're not performing. Just understand what it is that we care about and the real predictors of future performance versus your traditional counting metrics that are completely dependent on things that are outside your control."

I find it refreshing to see this kind of approach being taken inside the game itself, where it seems as though people have taken the to adapt to sabermetrics. Perhaps, in the not too distant future, we will see players have a better understanding of why the pitcher win statistic is seriously flawed and should not be valued highly. Pitchers won't feel slighted when they are taken out of a game where they could feasibly earn a win. Managers will be free to make pitching changes that make the most sense statistically. And I will not have to write 1700 words about a topic that should no longer be very relevant in our advanced understanding of the game of baseball.