Noah Syndergaard Has a Torn Lat
The Washington Nationals are the best team in the NL East. The second best team in the NL East might be the Mets Disabled List. Already consisting of Yoenis Cespedes, Steven Matz, Lucas Duda, David Wright, Wilmer Flores, Seth Lugo, and Brandon Nimmo, the injured Mets are now going to add Noah Syndergaard to the list, as the Mets announced his MRI this morning revealed a torn lat muscle.
Update on Noah Syndergaard. #Mets pic.twitter.com/n8JVxIBBid
— New York Mets (@Mets) May 1, 2017
While there’s no official timetable, this isn’t going to be a short DL stint. Matz missed two months with a similar injury back in 2015, and that was diagnosed as the lowest grade lat tear. At this point, it’s probably unlikely that Syndergaard is back before the All-Star break.
While the Mets theoretically had a lot of pitching depth before the season started, no team can really sustain the loss of three starting pitchers that easily, and there’s no replacing Syndergaard. This probably costs the Mets a win or two even if Syndergaard gets back in July, and if this lingers beyond that, it could be closer to three or four wins. This is a huge blow, on par with the Giants loss of Madison Bumgarner, and puts the Mets 2017 season in some legitimate jeopardy.
The NL Wild Card race might really end up being first-to-87-wins-gets-it. This doesn’t end the Mets chances of making the postseason, but they’re going to need some things to turn around in short order. They can only dig so big a hole before it becomes overwhelming.
Dave is the Managing Editor of FanGraphs.
Funny how all the projections said that the Giants and Mets were likely going to be the wild-cards. But injuries to your ace…that will do a number on your projections.
How DARE those confounded projections not account for potential dirt bike shenanigans!
It’s just another reminder that there are huge, huge uncertainties around projection systems…especially teams that have a couple of real star-level players.
Of course there are. That’s why projections shouldn’t be treated as anything more than a best guess, a weighed average of many possible outcomes.
I know there are people out there who like to pretend that a projection is a prediction (it isn’t) and furthermore that there’s some sort of institutional bias against the reader’s own personal preferences … but that’s the readers fault, not the projection’s.
Actually it was the dumb stupid humans who projected Syndergaard and Bumgarner for a full season of innings.
Would a projection system that simply looks at past time on the DL and some generic age-based injury risk be better at filling out the depth charts? Maybe. All we really need a human for is to account for a team’s specific plans; in other words, to select the intended starters, or to identify the platoons and stuff.
Yeah, but in this case it wouldn’t have worked either. You’d just have to assume that everyone gets 15% fewer PAs/IPs than we think, which would reward depth. I would guess that would bunch projected records even closer together.
I am just really skeptical of any output that gives me a mean and no confidence intervals.
For team projections, I have to think even this model would do better than just assuming everyone is healthy.
Unless they are hidden somewhere, I believe the current projections also only list a mean? And this should at least be a better mean.
It should be a better mean, but I would also think that there would be even more teams hovering around .500 in the projections. There wouldn’t be as much differentiation there. And that is unsatisfying.
I suspect this is also the reason why no one bothers to report the expected range out of outcomes and only reports the mean. The expected range of outcomes (at whatever interval you pick) would almost certainly be wide enough that it wouldn’t be fun to talk about.
My preferred solution is to build an interval (not really a confidence interval, just some error) around the projection that explicitly takes into account possible injuries to different players, which would take depth into account. But this would be horrible to model and could potentially ruin the fun talk around projections.
I enthusiastically await the dirt-bike related algorithms in future STEAMER models.
var @HoldMyBeer = Math.Random();
I guess is my point is that a computer model may have built in a 10% chance of ((dirt bike accident or other freak accident)) whereas a human is likely to give a guy like Bumgarner closer to a 0% injury risk by projecting him for like 210 innings.
Per this comment and in no way intended to be snarky;
I consider the preseason projections to be kind of unrelated insofar as they were projecting a different team (one with Syndergaard or Bumgarner on it).
The inured ace doesn’t change what the team (on paper) at the beginning of the season would likely do in the context of all the other “paper teams,” in a theoretical 2017 season.
The injured ace does a number on construction of the team. The newly constructed team (distinct from the one constructed over the offseason) then has a new set of projections in a new context (relative to changes to the rest of the league and what has already happened).
Even still, all of this occurs in theory. To paraphrase Derek Jeter, they still have to play the games.
To use an analogy, “In the Blink of an Eye” is an excellent film theory book. Written by an excellent filmmaker. It is not an excellent film. Because it consists of ideas about films.
The projections at any point for any teams are ultimately theoretical and can help inform us how a team will do (even when they lose their ace).
Player projections are (in my mind) essentially ideas that can lead to decision making for managers/gms/fantasy baseball players the way Walter Murch’s ideas can lead to creative decisions in filmmaking. They can also give us new ways of looking at the game, the teams and the players. (The way one might “watch films differently” after reading about film theory).
Projections give us a pretty good idea of teams but (as is mentioned here there and everywhere) they are not totally predictive.
All that said, I don’t think anyone really needs math of any kind to conclude, “If a team loses it’s best player it will be worse.”