A Question About Value in a Losing Effort

Mike Trout isn’t going to win the AL MVP this year either. The Angels are 67-76 and haven’t been in the playoff race for months, and while it’s an individual award, it almost always goes to a player on a playoff team. The argument essentially goes something like this:

It’s not the best player award, it’s the most VALUABLE player award. And the (insert bad team here) would have finished out of the playoffs without him too. Those numbers he put up didn’t actually lead to a winning season, so he can’t be as valuable as (insert other good player on a playoff team), who helped carry his team to the postseason.

Essentially, this argument suggests that there is little or no value to be gained from wins that do not result in postseason berth, or at least playoff contention. The Angels would be the Marlins without Mike Trout, but when discussing MVPs, both teams are just seen as equally lousy, and there is little credit given for the separation created between bad teams. Losing is losing, and it doesn’t really matter what you do if your team doesn’t win.

Here’s my question, though: when using this criteria to determine the MVP, why is this point only valid at the seasonal level? If performance in a losing season has no value, then surely that same theory holds at the game level, right? If a player has a huge game but his team loses, well, they would have lost even if he hadn’t played. At the end of the day, his performance did not change the standings one bit. He might have played well individually, but the value of those contributions was negligible because his team lost.

Carried to its logical conclusion, this theory suggests that MVP candidates should only be considered based on their performance in the games that their teams won. After all, performance in losses didn’t affect their teams overall result, and those hits and runs all turned out to meaningless in the end, so, why would we consider the stats piled up in a losing effort?

Thankfully, Baseball-Reference has this split available in their play index, so we can look at the leaderboard of performances in team wins. Since it’s B-R, we’ll sort by OPS+, so just remember to adjust high OBP guys up a bit and low OBP guys down a bit, since OPS overrates SLG in relation to OBP. Here’s the list of the best AL hitters in winning efforts:

Miguel Cabrera 73 326 102 15 1 26 54 0.381 0.485 0.735 1.220 193
Chris Davis 76 325 96 26 0 32 37 0.343 0.418 0.779 1.197 183
Robinson Cano 76 328 99 22 0 20 69 0.352 0.436 0.644 1.080 181
Adrian Beltre 80 353 123 20 0 25 31 0.388 0.448 0.688 1.135 172
David Ortiz 72 327 97 25 2 16 54 0.362 0.462 0.649 1.111 168
Mike Trout 66 309 89 15 4 14 48 0.352 0.456 0.609 1.065 154

While Cabrera is trouncing Chris Davis in overall offensive numbers, their numbers in games that resulted in a win are actually very close. Davis has played in three more O’s wins than Cabrera has Tigers wins, and while he has six fewer hits, his distribution of hits have skewed far more towards the impactful side, as he has 11 more doubles and six more home runs. He’s slugging .779 in games the Orioles have won, compared to .735 for Cabrera in Tigers wins. Once you factor in Cabrera’s advantages in singles and walks, he still comes out as the better hitter, but it’s really close, at least in terms of hitting.

If you wanted to do a complete comparison, you’d also want to include defense and baserunning, and Davis probably comes out slightly ahead in both of those. At the very least, the relative performances of Davis and Cabrera in their own team wins would make it a coin flip between the two, since performance in a team’s losses don’t matter, and that’s where the real gap between them comes in. In losses, Cabrera has a .985 OPS, while Davis is just at .816, but since the argument is that there is no value in production that does not result in a win, those numbers should theoretically all be ignored.

As you can probably imagine, I don’t really buy into this line of thinking. I think wins and losses are measures of a team’s overall performance, and the MVP award is a measure of individual value. I do not think that we should give a player MVP credit for the performance of his teammates, which is something he has no control over. And I do not think that performance in losses is worthless, nor do I think that a player on a losing team had minimal value.

But a lot of people believe that argument. Most of the people who have MVP votes believe that argument. That argument is going to be the reason Miguel Cabrera runs away with his second straight MVP award, and Mike Trout probably won’t even finish in the top two. But, I would challenge people who believe that performance in a losing season does not matter to also not consider performance in losing games. If you can’t be valuable on a losing team over 162 games, how can you be valuable on a losing team in a single game? Let’s be consistent. Either performance in wins is the only thing that matters, or overall individual performance matters. Pick one and run with it.

Dave is the Managing Editor of FanGraphs.

Newest Most Voted
Inline Feedbacks
View all comments
Jacob Smith
10 years ago

As an extension of this idea, wouldn’t you still say Cabrera’s contributing less overall to his teams’ wins, given there’s a better team around him in Detroit? Davis is doing far more to keep his team in contention, relatively, than Cabrera is. The Tigers might still be winning the central if Cabrera hadn’t played an inning this year, while the Orioles most certainly wouldn’t be, given their division.

Jacob Smith
10 years ago
Reply to  Jacob Smith

Couldn’t, rather than wouldn’t in first sentence above.

10 years ago
Reply to  Jacob Smith

The answer is no, and this is the fundamental point of the development of context-neutral statistics. Win or lose for the Tigs, Cabrera has contributed what he’s contributed. Cabrera (or any player) can only control (some of) what happens when he’s at the plate or when a ball is hit to his side of the IF/shallow OF/foul territory. It doesn’t make sense to judge any player’s overall value based on the circumstances that he doesn’t control.

That doesn’t mean there aren’t good reasons to use context-dependent stats. Quite the opposite. Numbers that rely on context may be helpful to making in-game decisions or off-season signings if a player is truly found to excel in certain situations that a baseball club knows they can repeatedly put that player in. This can help a team find the right player that fits their price range and can have his value maximized if his strengths fit the club’s specific deficiencies.

10 years ago
Reply to  Nathan

“It doesn’t make sense to judge any player’s overall value based on the circumstances that he doesn’t control.”

Taken to that logical extreme, it would not make sense to gauge a player on their natural genetic talents (which is clearly out of their control). Obviously, this is a continuum rather than a black and white issue.

For my part, I would be somewhat interested in seeing how well a player hit against good pitching in considering their MVP stats. While it is context, there is a significant difference between piling extra runs onto a bad pitcher compared to scoring a single run against an ace. While we treat “run environment” as being season long, it is actually game-long.

So, while Dave was trying to beat up on a straw man, there is something to be said for examining individual games rather than the season stats. However, the way I would do it would be to adjust the “replacement level” for each game based on the opposition (e.g., aggregate run prevention for those pitchers with that defense). So, basically, you get extra credit for hitting a home run off of 1999 Pedro but reduced credit for hitting one off of 2013 Ryan Dempster.

10 years ago
Reply to  B N

“, I would be somewhat interested in seeing how well a player hit against good pitching”

Why? Is killing bad pitching and struggling against good pitching somehow less valuable than performing more evenly? I could see some arguments that it might be, but I don’t think its an obvious fact by any means.

Free Bryan LaHair
10 years ago
Reply to  Jacob Smith

could, yes, but the author’s analysis is based on numbers. your conclusion is based on the eye test and how you feel about Cabrera/Davis after watching them play.