Separate names with a comma.
Discussion in 'Red Sox Forum' started by Trotsky, Apr 26, 2018.
Earnest question - what difference does height make to a baseball player?
Ya, I mean Altuve and Pedey have both won MVPs. Is there extensive research on aging shorties?
Taller and heavier players are generally better hitters and hit with more power, and also project to hit better and for more power going forward.
It's a really interesting article, but Betts is a 5 tool guy, and that doesn't address the other tools and positional weight since the article is 12 years old before that stuff was mainstream. There's no evidence his overall value would erode more rapidly than a guy at the other extreme like Aaron Judge, who almost certainly will need to change positions before Betts will. His fielding baseline is just way lower.
The conclusion from the article using two hypothetical guys at different size extremes:
"Due to his size alone, Barry is projected to be more than seven runs better than Larry, despite equal performances in the previous season! That’s a pretty large effect. He’s expected to hit five more doubles and six more home runs, and have only six fewer singles and two fewer triples. But in this example, we’re going from about the 7th percentile to the 93rd in terms of player size. For 95% of all players, the effect of size will be capped at a total difference of less than 10 runs. Is that significant? Certainly. But it’s not that much."
Slight cherry picking but between his 2016 season and this one so far (Judge doesn't have enough full seasons) Betts he has been worth 89 runs above average on offense (including base running). Aaron Judge has also been worth 89 runs above average between last season and this one. The defensive difference is significant enough to wipe out the hypothetical offensive decline and then some.
Defense and base running decline more reliably and inevitably than hitting, generally almost immediately by reaching the big leagues, but certainly after age 25. According to fangraphs, Mookie's have already begun to do so this season. Judge's UZR/150 this season is 15.2, Mookie's is 14.8. By b-ref dWAR, it's Judge 1.0 (in admittedly more games), and Mookie .5.
I'm not going to get drawn into any further comparison between Mookie and Judge. I'm a Red Sox fan, and I like Mookie better. But over the next ten years the great majority of both of their value will be at the plate, and I think you're overstating the defensive delta between them.
does anyone believe that Mookie is now a 1.0 dWAR (0.5 through 78 player this year) when he was a 2.9 & 2.6 dWAR player in 2016 & 2017?
He's slowed down at 25?
I think everyone should believe that a half season of defensive data is highly unreliable.
I don't believe Mookie is worth precisely 1.0 dWAR any more than I believed he was worth precisely 2.6 last season. But has he slowed down at 25? Almost certainly. Pretty much everyone does. Every year.
Betts's sprint speed the last 3 seasons: 28.0 27.9, 28.1... So, not yet.
The sprint speed measurement is a pretty cool tool, but it doesn't really tell us too much with respect to things like first step/jump on defense and much of baserunning (particularly stolen bases). It's a peak speed measurement, and it's not surprising that a player will largely retain that peak speed in his 20s. The important part for most baseball actions is getting up to that peak speed. For example, we don't really think that McCutchen has not lost a step over the past three or four years, or that JBJ or Mookie have below league average effective speed for a CF and RF respectively, as pure reliance on Statcast sprint speed would suggest, do we?
Also, Mookie was at 28.7 ft/second in 2014. That's a fairly significant decline by that measure.
I agree that top sprint speed isn't the be-all and that first-step, reaction time, route-efficiency, acceleration, etc are all critical to variables that aren't necessarily related to sprint speed. But, the empirical evidence we have suggests he's as fast as he's been the last 3 years, with his 28.7 sprint speed at age 22 in the first year they recorded the stat as an acknowledged data point. Maybe he's slower than he used to be, but there's no evidence that he's slower than he was last year. And, if you want a proxy for acceleration, maybe something like home-to-first times is a better proxy, but for the life of me I can't find those stats online.
Anyway, as others have pointed out, I think the simplest explanation for Betts's "dip" in his age-25 season on defense is that the data set is too small and his stats are still fluctuating quite a bit. Of course, in 2-3 years we can look back on this conversation and have a better sense if Mookie's defense was in decline at this point.
Well, I don't disagree with the point that we're working with a small defensive data set that can be misleading within a season. But you could say the same thing for your reliance on the sprint speed data for your conclusion that "there's no evidence that he's slower than he was last year." One might even theorize that a player's recorded sprints would be expected to be slower in the later stages of a long season, which could bring Mookie's number (now based on only 127 runs) down below last year's, which was averaged from 314 runs. Let's see where that number is at the end of the year. I'd bet it won't be going up.
Anyway, he's gone 28.7, 28.0, 27.9 in his three full seasons. That's not inconsistent with the average decline of a little more than an inch per season. I wasn't saying he's fallen off a cliff, or is unusual. There's no reason to believe it won't be gradual, but Father Time is undefeated.
Apparently, Fangraphs reworked their WAR numbers over the break. They now have José Ramirez, Trout and Mookie tied at the top at 6.5 (had been Trout, Ramirez and Mookie, separated by a couple of tenths each ). Looks like Mookie’s defensive component (now 3.9) was adjusted upward. It looked too low by the eye test all year. Of course, Mookie would probably be leading the WAR thing easily if he hadn’t missed over two weeks’ games. Keep on keepin’ Mook.
That's interesting. I wonder what specific adjustments Fangraphs made. Bref still shows Trout at 6.8, Ramirez 6.6, Betts 6.3.
It's amazing how close the three of them are.
Mookie actually has the most runs above replacement for fielding -- 7.0. It's just he loses the most runs for positional adjustment.
Pretty sure it was in the defensive area like I said, but what parameters within that, don’t know. Googled “fangraph changes in calculating WAR” for a while and got nothing. As some SOSHers often say, defensive WAR is much less of a science, or takes longer to “believe” than the offensive side. So, the positional adjustment for right field is -7.5, same as for left field. That’s a joke at Fenway of course.
Hits for the cycle on 8/9 in Toronto tonight.
Single, triple, double, homerun with a walk in the plate appearance before the home run.
He was on pace for 10 WAR before today's game with 20 games missed. That goes up to 12 if he doesn't miss games. The Sox have had 8 10 WAR seasons, 3 by Ted, 2 Yaz, 2 Speaker, 1 Petrocelli.
How crazy is his season if he doesn't miss time when he's scorching hot in May?
Video of the his cycle for anyone who missed tonight's game (or just want to watch his greatness over and over).
Edit: that query got jacked up
Lets try that again
Lou Clinton was the last sox to hit for the cycle and walk in 1962
Last non sox was old friend Pablo Sandoval!
Players who hit for the cycle and lost:
Back over 1.100 in OPS. Back in front of Trout and everyone else in OPS.
So incredibly fun to watch. We are witnessing greatness (obviously). I’m constantly in awe of a guy his size doing the things he does. It’s almost hard to appreciate it in real time.
With last night's game, Mookie bridges much of the gap that opened up between himself and Trout/Ramirez. He's nearly tied in fWAR (Trout-7.6, Ramirez-7.5, Betts-7.5) and just a bit behind in bWAR (Trout-7.8, Betts-7.5, Ramirez-7.2). Given that team success matters in the MVP process (debatable though that may be), Betts may get a meaningful edge if the race ends up this close in the end.
I hear people on MLBN talking JDM for MVP, largely because of his triple crown possibility.
This may be a redux of Trout vs. Cabrera a few years ago. If JDM gets the TC does that necessarily mean he had a better year than Mookie when factoring in defense and baserunning?
1) JDM 4.57
2) Betts 4.44
3) Beni 3.63
4) Jose Ramirez 3.47
5) Bregman 3.45
Probably makes sense to give Betts and Martinez MVP Awards this year -- issue 2 in the AL and none in the NL.
That would mean that JDM made up a deficit of .016 to Mookie in the batting title race, which in the last 2 months of the season is not easily done. So in that case, it is extremely probable that other statistical measures like WAR would also shift in JDM's favor. As a result if JDM wins the TC, it is extremely likely he wins the MVP, even more so when it would be only the 2nd TC in the last 50 years (hard to believe that Yaz's 67 was 51 years ago).
I think the triple crown produces the MVP. Helping JDM also is that there is a lot of talk about his self improvement study habits, helping other Red Sox players, great clubhouse guy, etc. Big question is can he win the batting title. I would bet no.
I would guess voters tune in to that stat in roughly 2025 after it starts to go out of style, but it really ought to have a lot of weight when considering the offensive side.
Mookie Betts saving a 5th inning run when the Red Sox win by 7 is much less important than him hitting a game tying single in the 8th inning.
I know Bill James greatly prefers those types of metrics instead of WAR.
What if JD wins Triple Crown and Jose Ramirez goes for 50HR/40sb? There's a chance Betts doesn't even finish 2nd in the MVP voting.
I started to laugh at this, but then I went and looked. Wow. The top 4 in the NL in bWAR are all pitchers, and of the top 2 position players, one is heavily buoyed by his defensive numbers (Cain). I suppose if Carpenter keeps mashing the way he is, he'll separate himself from the field by the end of the season, but man there are really no standouts in the NL this year.
I am fully aware of Jose M. I also keep an eye on Francisco Lindor. So, there is kind of a big 5 in the AL MVP race: Trout, Martinez, Betts, Ramirez and Lindor.
Big 5? Martinez...
Manny's not here anymore? Thanks, fixed.
Long odds of it happening, but Ramirez has a chance to lead the league in both stolen bases and home runs (he's leading in SBs and only 2 HRs behind J.D. right now). SBs aren't so trendy lately, but there may be some voters who would think a feat like that as a testament to his all-around game, giving him a small boost. I can't think of anyone who ended up leading the league in both of those categories.
I think you missed the point. If Jose Ramirez went 50/40, he would be the only player ever to do so. That would definitely be more than a small boost.
So, if we replace the b-r method of calculating offense above average (Rbat) with WPA, then we get the following WAR leaders in the AL (I only looked at the current top 10 in WAR excluding Judge)
JD Martinez 5.4
The movements are really only due to Trout having a much worse WPA than Rbat (3.3 vs. 4.9).
In my opinion, talking about WPA really only serves to demote Trout, it doesn't really bolster JD Martinez.
Edit: It brings Benintendi & Bregman into the outskirts of the list with 5.4 & 5.2 WAR, respectively.
Edit again - similar story starting with fWAR as your base and converting it to WPA offensive:
JD Martinez 5.2
Using this method, instead of Trout being the leader with a Betts/Ramirez nipping at his heels, you have Betts as the leader with Ramirez & Lindor close and no one else really in the conversation.
It's a good way to argue for Ramirez, really - get Trout out of the conversation and argue for the eye popping HR/SB numbers to make up the half win difference in WAR, a non-precise metric.
Actually, I was trying to complement your point by offering another possible scenario in which Ramirez's accomplishments in power and speed would suppress votes for Mookie. You may easily convince me that the effect would be more than small, however--in either case.
WPA is an interesting statistic and one that seems tailor-made to evaluating the most *valuable* player in the league, at least if you interpret that stat to mean the player that resulted in the most wins for his team. That said, I do see a couple shortcomings when using that stat to determine the *best* player in the league, which, if we're being honest, is *sometimes* what people mean when they refer to an MVP. Consider the following:
If you're a great player on a shitty team, your WPA will always suffer, since you'll often find yourself on the losing end of blowouts where your greatness barely moves the needle. Call this the Mike Trout Effect.
If you're a great player on a great team with other great players, your WPA will also suffer, since you'll often find yourself on the winning end of blowouts where, again, your greatness barely moves the needle. Let's call this the JDM-Betts Effect. In this way, players like JDM and Betts would cannibalize each others' WPA by hitting bombs and putting the game out of reach. For example, how many games did Mookie take a big bite of WPA with a HR or a hit+SB before JDM ever got to the plate? How many times did Betts come to the plate in his 2nd AB with a 4-0 lead instead of a 1-0 lead because JDM hit a 3-run bomb?
Anyway, this is probably all par for the course for someone that is familiar with WPA but I'm just now sinking my teeth into it.
Crazy — seven of the top nine are either Red Sox (4) or Indians (3). If vote-splitting is a factor, maybe that gives a boost to Trout and Bregman?
Nobody has ever even gone 50/30 -- which he could possibly do.
Interestingly, of the 4 players who have gone 40/40, three of them didn't seriously threaten to win the MVP. Soriano in 2006 was 6th in the voting, Arod in 1998 was 9th in voting, and Bonds in 1996 was 5th in voting. Only Canseco (1996) won the award.
But I guess that's all messed up by steroids. 40/40 would probably be viewed as more of accomplishment today than it would have been back then given that the 40/40 guys were competing in an era where hitting 50 wasn't as special.
The Silver Slugger in theory ought to "reward" Trout's seasons since WPA as you mentioned can't have as much of an impact for a player on a shitty team. But of course the Silver Slugger is basically the equivalent of the European Championship WWE belt. Just pointing out that there is a an award for what Trout does best since the Angels are perennially defeated by Mike Scioscia.
Though the Betts/JDM analogy still has left plenty of meat on the bones for one another this season.
Soriano was better than the actual MVP winner, Ryan Howard, but so were 8 other guys. He had the 5th highest bWAR in the NL and was nearly 2.5 behind Pujols.
ARod was THREE AND A HALF WINS better than Juan Gonzalez, and also had the highest bWAR in the AL - he should have won the MVP in 1998 (Clemens was the only guy close, about a half a WAR behind ARod).
Bonds was THREE WINS better than winner Ken Caminiti, and 1.5 wins better than the 2nd best bWAR in the NL, Bernard Gilkey - he should have won the MVP in 1996.
Canseco was a win worse than Wade Boggs, but at least he had an MVP caliber season.
Silver Slugger is only about hitting (aside from being awarded to each position). The definition of MVP is very much up for debate.
WPA/LI is intended to address the difference in opportunities:
JD Martinez 4.2
So if you use WPA/LI, then the leaderboard looks something like:
JD Martinez 4.7
But really, at some point this becomes a bit of p hunting (or whatever those statisticians call it). Edit: It's called "p-hacking."
Somewhat surprisingly, Mookie Betts is now one of only three Red Sox in history to accumulate both 100 HR and 100 stolen bases -- and somewhat less surprisingly, the only one to get there while still in his twenties. (The other two are Pedroia and Yaz.)