Let’s winnow the wheat from the chaff
The final SnakePit award is, of course, the most prestigious, even in a shortened season such as we’ve seen already. Despite there only being 60 games, I am initially inclined to continue the usual process, by which the winners of Rookie, Unsung Hero and Pitcher of the Year awards get automatic nominations, leaving the other slots (probably two or three) open for consideration. I really do not think that we’ll need all that many candidates. But even though the collective performance of the Diamondbacks was underwhelming, there were still some individuals who were worthy of praise. So, on that basis, Tim Locastro, Daulton Varsho and Zac Gallen are already on the final poll.
To look at the rest and figure out who should join them, I’ve taken a similar approach to the one used for Pitcher of the Year. Below are the bWAR and fWAR figures for all non-pitchers who were above replacement value in 2020 [There’s a whole argument about whether pitchers should be eligible for the MVP, but if there’s going to be anyone, it’ll obviously be Gallen, and he’s already nominated. So I decided not to run the numbers for anyone else] The players are listed in descending order of the average of bWAR and fWAR.
In most cases, fWAR and bWAR are fairly close. But in the case of Ketel Marte and Carson Kelly, they’re quite disparate, especially considering this was over only 60 games. If we pro-rate Ketel’s numbers above to 162 games, you get a bWAR of 4.1 and fWAR of 1.1 – three wins difference, which is pretty significant. For Carson, the pro-rated gap is about two wins, but he only appeared in 39 games last year. I asked Jack to weigh in on this discrepancy, and he said as follows:
Typically the major differences between Fangraphs and Baseball Reference WAR boil down to defensive metrics. Not 100%, but the bulk of any gaps usually can be attributed to the defense. This is certainly the case with Carson Kelly’s 2020 WAR metrics
A couple of quick notes:
- BR does not show rounded numbers, whil FG shows to 1st decimal point
- BR breaks out Baserunning and DP’s, but FG puts the DP’s into either their batting runs or the baserunning (it’s not clear which immediately)
- Positional and Replacement runs are typically almost exactly the same
- FG includes pitch framing in their catcher defense calculation, while BR does NOT include framing. When I asked Sean Forman about this, he said he still wasn’t 100% confident in the metric to include it.
So as you can see, with Kelly he has very close numbers in batting runs, Positional, and Replacement. But the fielding runs gap of 6.3 runs accounts for almost all of the difference of 0.7 WAR between the two metrics
Ketel Marte’s comparison is more confusing and frankly I don’t quite understand it.
There is a 3.2 runs difference in batting, which is unusual but not altogether crazy, and while the gap of 6 runs in fielding is quite large for such a small sample, again, it’s not that unusual either. But what’s really weird is the positional adjustment. I don’t know why BR has +2 and FG has just +0.6
This is a good example of why it’s probably best to average the two metrics whenever possible. It’s extra work, but I think creates a more balanced view, as opposed to picking the metric that favors one’s personal bias. If you need a quick and dirty, and are just referencing, using one or the other is fine. But if diving deeper, averaging is suggested.
And that’s what we’ve done here for MVP purposes. 🙂 Anyway, we’ll adopt a similar approach to picking the other candidates as we did for the pitching award. Post up to three names in the subject line of your comment, that you want to see make the final ballot. You can use the body of the comment to make the case for any of your candidates, as appropriate. I’ll then count up the “votes” and add the most popular candidates to the final ballot on Thursday.