Article: 2017 Twins Player Predictions: Jason Castro

[Seth Stohs]

The Minnesota Twins offseason is generally considered to have been very quiet. It would be difficult to dispute that too much. However, the Twins new front office was very aggressive in their pursuit of one free agent. By mid-November, just weeks after the new regime began, it was clear that the Twins were a serious player in negotiations for catcher Jason Castro. By the end of the month, Castro had signed a three year, $24.5 million deal with the Twins.

 

As has been written about, ad nauseum, much of Castro’s value comes from the his work behind the plate. No need to go over there much here, but Castro has great pitch framing stats. He presents the ball well, but he’s also touted for working well with pitchers and calling a good game. His numbers indicate that he is average at controlling the running game. He also seems to give up a fair number of passed balls.However, we are here to provide predictions for the offensive side of the game. Castro was the Astros first-round pick in 2008 (10th overall) out of Stanford. He was an All Star in 2013 when he set career-highs in most statistical categories. The last three years have not been great offensively, particularly if you only look at batting average. He has hit .222, .211 and .310. However, he generally appears to have a good idea at the plate, and he has had ten or more home runs each of the last four years.

 

 

So let’s get to it. Below you will find my predictions (or guesses, if you prefer) for Jason Castro’s 2017 season. Hopefully you will consider posting your thoughts and your predictions into the comments below. It’s always fun to take a look at the end of the season and see how our predictions look.

 

KEY NUMBERS

 

30 - No, Castro isn’t going to hit for average, and he won’t be an on-base machine, but the last two years he has had 30 extra base hits and 11 home runs each year. He will occasionally show some pop in his bat.

 

.149 - Castro only hit .231 against right-handers last year, but he hit just .149 against lefties. The American League Central Division has several quality left-handed starters (Carlos Rodon, Jose Quintana, Danny Duffy, to name a few), which is why it is important that the backup hit right-handed. Aside from Opening Day against Duffy, hopefully Castro will have limited plate appearances against southpaws.

 

 

PREDICTIONS

 

Jason Castro: 390 at-bats, .228/.302/.372 (.674), 20 doubles, 0 triples, 12 home runs.

 

---------------------------------------------------------------------------

 

I think that Castro will play around 120 games, and obviously he’ll bat in the lower third of the lineup. It’s unlikely that Castro will hit even .250, but I think he can hit for a better average if he is used appropriately. He’s got a smooth swing, and he should continue to hit plenty of extra base hits.

 

Now, a .674 OPS for a catcher would be pretty solid. I think coupled with his defense, it would be a successful season for the 29-year-old.

 

YOUR TURN

Your turn. Share your thoughts on and predictions on Jason Castro in 2017. We can take a look back at the end of the season and see how we did.

 

Click here to view the article

[jsteve96]

I think if they use him well and mix in Giminez or Garver vs lefties that a .674 OPS is definitely low balling. I could see a .700 OPS if he gets limitied LvL AB's.

[HitInAPinch]

Just watching the way he moved behind the plate in SP showed how big of an upgrade Castro was.

 

Limit his AB's against lefties:  Universal agreement on that, me thinks!

[tarheeltwinsfan]

Since I pitched in high school (yeah, I know, "big deal") I have always been interested in the selection of pitches. Think about it. A good pitcher can pitch up, down, middle of the plate, fast, slow, slower, inside , outside, rising, sinking, moving horizontally right or left. Just using these 12 variables, how many different pitches can a major league pitcher throw? Some math major reading this may answer this question for us by saying it is 12 squared, which is 144. However I say the answer is infinitesimal. I know control factors in, but I'm just referring to pitch selection. The last 4-5 years I have been very interested in pitch selection. I assume the catchers have been calling the pitches (I don't know that, I'm just assuming that). I have been disappointed with he lack of variety of consecutive pitches being called by the Twins catchers. Pitching is largely deception. The pitcher tries to fool the batter. A lack of variety of pitches makes it easier for the batter to know what is coming, and therefore conceivably easier for the batter to hit what he knows is coming out of the pitcher's hand. I told Jason Castro in spring training several weeks ago that I was glad to have a Stanford University graduate calling the pitches for the Twins. I am interested to see if the variety between pitches changes and if that helps.

[markos]

Steamer .224 .303 .374
ZiPS  .221 .299 .374
PECOTA  .239 .313 .405

I'm a little worried overall about Castro. There are enough warning signs with his profile (particular the increasing amount of Ks) that it wouldn't be shocking to see his offensive output crater. Therefore, I'm tentatively going to take the under on a .675 OPS.

[Brock Beauchamp]

Castro is a .750 OPS player against RHP and a .500-.530 guy against LHP.

 

Castro took 26% of his plate appearances against LHP in 2016.

 

I did a quick spot check and the league average for LHB against LHP is just a tick over 30%.

 

So Castro was already seeing his plate appearances against LHP minimized by the Astros.

 

There may be a bit more wiggle room in there as you push those plate appearances toward 20% but I'm skeptical the Twins, even with aggressive platooning, will be able to get it below that number.

 

So, maybe there's room to improve there but not a lot. I think Seth's .675 OPS prediction is pretty fair. Maybe the Twins could try to put that toward .690 but large gains will not be found through platooning.

[mikelink45]

A catcher does not have to hit a lot to be valuable.  I remember Del Crandall anchoring the Milwaukee Braves in the 1950s when they were a really good team, but he did hit 260s with a strong arm and a good behind the plate presence.  Castro hitting in the 220s is way to low for my liking, especially on a three year contract.  I hope we get Garver up and splitting time before the season is over. 

[Respy]

Offensively, I'm most interested where his batting average will sit.  .210 is painful.  .230 is tolerable.  His 2013 season at .276 would be amazing.  His BABIP numbers give a good indication of where he projects.  MLB average BABIP is about .297-.300.  The year he hit .276, his BABIP was .351.  A very lucky year for him.   The last three years where he hit, .222, .211, and .210, his BABIP was .294, .280, .297, respectively.  A bit on the unlucky side, perhaps.  Using these three years, and MLB average BABIP of .297, would say he projects to hit at .219.  Again, I would be happy if he hits over .230 and hits over 10 homeruns.  

 

Regarding those homeruns, I looked at his batted balls chart against RHP from 2014-2016 on Fangraphs (cannot figure out how to paste chart here).  20 HR to right field.  2 to center field. 10 to left field.  6 of them to left field were really short (short porch in Houston).  Those 6 would not be homeruns at Target Field.  You might ask about deep flies to Tal's Hill in Houston that would be Target Field homeruns?  It looks like Castro doesn't hit many deep fly ball to center.  So, we might see a regression in homeruns this year.

 

 

[Respy]

 

Steamer .224 .303 .374
ZiPS  .221 .299 .374
PECOTA  .239 .313 .405

I'm a little worried overall about Castro. There are enough warning signs with his profile (particular the increasing amount of Ks) that it wouldn't be shocking to see his offensive output crater. Therefore, I'm tentatively going to take the under on a .675 OPS.

 

Regarding his strikeout numbers, you can look at his minor league numbers and his first two MLB seasons (2010 and 2012) and he had a walk rate of around 15%ish.  But, he only averaged 6 homeruns per 400 plate appearances.  I think at some point around 2012-2013 someone told him that if he is going to have any value whatsoever as a hitter, he needed to start popping some dingers.  His strikeout rate doubled to about 30%, and his homer rate about doubled to 12.3 HR per 400 plate appearances.  Has that made him more valuable as a hitter?  I don't know.  I think modern analytics say probably yes.  I don't think he'll change his approach given his successful offensive year in 2013.

[Intramural Legend]

Sports Illustrated has an article with an anonymous scout talking about each team.  The scout had this to say about Castro

 

"They gave Jason Castro a lot of money to be their front-line guy. They had a good defensive catcher in Kurt Suzuki and let him walk. I would rather have Suzuki than Castro.."

 

Needless to say, this scout is not very good.

[Mike Sixel]

IIRC, he's one of the guys that decided swing plane and elevating the ball was more valuable than hitting down on it, and that's why he changed, but that could be a lack of sleep thing causing me to imagine things.

[Doomtints]

It would be ideal if he could just hit like Henry Blanco did when he was a Twin.  Blanco did not have a great batting average, but he banged out doubles and home runs when it mattered.  He looked like a superstar at the right moments.  

 

If Castro can play defense as some people suspect and he can swoop in offensively with a superman cape when the situation is dire like Blanco did, all will be fine. It's OK if he's Clark Kent the rest of the time at the plate.

Edited March 28, 2017 by Doomtints

[Respy]

 

It would be ideal if he could just hit like Henry Blanco did when he was a Twin.  Blanco did not have a great batting average, but he banged out doubles and home runs when it mattered.  He looked like a superstar at the right moments.  

 

If Castro can play defense as some people suspect and he can swoop in offensively with a superman cape like Blanco did, all will be fine. It's OK if he's Clark Kent the rest of the time at the plate.

 

I know analytics does not tell the whole story, but I'd point out that Henry Blanco's 'Clutch' stat was -0.65 with the Twins in 2004.  FYI Jason Castro's average yearly MLB Clutch value is -0.2.  Interpretation chart below (via Fangraphs):

 

Rating                Clutch
Excellent               2.0
Great                    1.0
Above Average     0.5
Average                0.0
Below Average    -0.5
Poor                    -1.0
Awful                   -2.0

[Respy]

Slightly off-topic, but for fun, I thought I'd also share a projected Twins lineup, with 2016 clutch values shown:

 

Polanco (0.1)
Mauer (-1.8)
Dozier (-2.3)
Sano (-0.5)

Buxton (-1.5)
Kepler (-1.1)
Park (-0.9)
Castro (0.1)

Rosario (-1.0)

 

Well that's depressing.
 

 

[Mike Sixel]

 

Slightly off-topic, but for fun, I thought I'd also share a projected Twins lineup, with 2016 clutch values shown:

 

Polanco (0.1)
Mauer (-1.8)
Dozier (-2.3)
Sano (-0.5)

Buxton (-1.5)
Kepler (-1.1)
Park (-0.9)
Castro (0.1)

Rosario (-1.0)

 

Well that's depressing.
 

 

wowsers. 

[Brock Beauchamp]

 

Slightly off-topic, but for fun, I thought I'd also share a projected Twins lineup, with 2016 clutch values shown:

 

Polanco (0.1)
Mauer (-1.8)
Dozier (-2.3)
Sano (-0.5)

Buxton (-1.5)
Kepler (-1.1)
Park (-0.9)
Castro (0.1)

Rosario (-1.0)

 

Well that's depressing.
 

There's a reason the team finished a whopping 12 games under their expected sequencing.

 

The entire team was awful in run-scoring situations.

 

At one point, the team had something like 17 solo home runs and 3 home runs with runners on base.

 

The real kicker to that stat? With the bases empty, the team had one of the best OBP in baseball at the time (2nd or 3rd IIRC). With runners on base, they were dead last in OBP.

 

That kind of statistical anomaly is incredibly difficult to repeat.

[Deduno Abides]

390 ABs > 2015 or 2016 PAs. More ABs -> more ABs vs. LHP -> lower BA & lower SA. Prediction: unless he builds on his 2016 improvement in % of hard hit balls, OPS < .640.

 

 

Two alternatives. (1) High: despite predictions that it would never happen, he rediscovers his 2013 bat and becomes the free agent steal of the year. (2) Low: he starts at .200 with a .275 OBP and weaker SA due to playing at Target Field instead of Minute Maid, while Garver gets called up mid-season, plays OK and takes many of Castro's ABs, making him a bad free agent investment.

[crapforks]

 

Castro is a .750 OPS player against RHP and a .500-.530 guy against LHP.

 

Castro took 26% of his plate appearances against LHP in 2016.

 

I did a quick spot check and the league average for LHB against LHP is just a tick over 30%.

 

So Castro was already seeing his plate appearances against LHP minimized by the Astros.

 

There may be a bit more wiggle room in there as you push those plate appearances toward 20% but I'm skeptical the Twins, even with aggressive platooning, will be able to get it below that number.

 

So, maybe there's room to improve there but not a lot. I think Seth's .675 OPS prediction is pretty fair. Maybe the Twins could try to put that toward .690 but large gains will not be found through platooning.

I know that lineup position has largely been proven to not profoundly impact runs scored, but in some lineups the 7,8,9,1 is Park, Castro, Buxton, Dozier. I think that the opposition would rather not deploy a LOOGY to get Castro out in that case when they could save them for Kepler, Mauer, Rosario. Surrounding Castro with LHH would only increase his PA vs LHP. 

[Brock Beauchamp]

 

I know that lineup position has largely been proven to not profoundly impact runs scored, but in some lineups the 7,8,9,1 is Park, Castro, Buxton, Dozier. I think that the opposition would rather not deploy a LOOGY to get Castro out in that case when they could save them for Kepler, Mauer, Rosario. Surrounding Castro with LHH would only increase his PA vs LHP. 

That's a valid point.

 

And I'm not sure lineup position in regards to handedness has been proven irrelevant. Most conversations I've seen revolve around stuff like whether Dozier should hit second or fifth and whether that matters (ignoring Dozier's home runs, just speaking generally about any player, really).

 

I think handedness matters quite a bit. If you stack three lefties in a row, those lefties are going to suffer in the last third of the game as managers throw lefty relievers at them at least once a game, possibly multiple times a game if possible.

 

If you're a lefty facing LHP 10% more often than a typical lefty bat, your season statistics are going to suffer and the difference won't be negligible. Many LHB hit LHP at a .150-.200 lower OPS. That's not a small difference, that's the kind of thing that can move a player's overall season OPS down .020 or more points.

[Respy]

Anyone know if opposing teams tend to deploy pull-shifts for Castro?  He seems to me to be fairly pull-happy, especially on infield grounders.  Perhaps there could be a benefit here to putting a speedy guy like Buxton in the lineup in front of Castro, so that his steal threat and first-to-third threat makes the other team less likely to shift?

 

On that note, does anyone know if a team is less likely to shift when there's a speedy runner on first?

[Brock Beauchamp]

 

Anyone know if opposing teams tend to deploy pull-shifts for Castro?  He seems to me to be fairly pull-happy, especially on infield grounders.  Perhaps there could be a benefit here to putting a speedy guy like Buxton in the lineup in front of Castro, so that his steal threat and first-to-third threat makes the other team less likely to shift?

 

On that note, does anyone know if a team is less likely to shift when there's a speedy runner on first?

He's definitely a pull guy (most are nowadays though) but not obnoxiously so.

 

http://www.fangraphs.com/spraycharts.aspx?playerid=8722&position=C&type=battedball

 

In comparison, he's no Brian Dozier, that's for sure.

 

http://www.fangraphs.com/spraycharts.aspx?playerid=9810&position=2B/SS&type=battedball

[Willihammer]

 

That's a valid point.

 

And I'm not sure lineup position in regards to handedness has been proven irrelevant. Most conversations I've seen revolve around stuff like whether Dozier should hit second or fifth and whether that matters (ignoring Dozier's home runs, just speaking generally about any player, really).

 

I think handedness matters quite a bit. If you stack three lefties in a row, those lefties are going to suffer in the last third of the game as managers throw lefty relievers at them at least once a game, possibly multiple times a game if possible.

 

If you're a lefty facing LHP 10% more often than a typical lefty bat, your season statistics are going to suffer and the difference won't be negligible. Many LHB hit LHP at a .150-.200 lower OPS. That's not a small difference, that's the kind of thing that can move a player's overall season OPS down .020 or more points.

Of course they *might* suffer later in the game but they will score more runs early in games (assuming the starter is opposite-handed) over time, all else being equal.

The other thing is, conventional bullpen managment says to save your best relievers for leads, particularly when on the road. Take a tie game or +1 lead into the 7th and you might never have to worry about facing a lefty or, at least, not the best lefty in some cases.

[Brock Beauchamp]

 

Of course they *might* suffer later in the game but they will score more runs early in games (assuming the starter is opposite-handed) over time, all else being equal.

The other thing is, conventional bullpen managment says to save your best relievers for leads, particularly when on the road. Take a tie game or +1 lead into the 7th and you might never have to worry about facing a lefty or, at least, not the best lefty in some cases.

That's true about starters but lefties get that advantage either way. It doesn't matter if they're split up or in a row. Yes, bunching them together might impact run-scoring a bit but I was thinking of individual OPS performance, which it won't affect.

 

As for leads/deficits late in games, I suspect the lead/deficit would have to be pretty substantial for a manager not to say "aw hell, three lefties are coming in a row, I'm just going to throw a lefty out there to make sure nothing bad happens". Every pen has one lefty reliever and lots have two of them. Chances are, one of them isn't amazing so why not put him out there to face a trio of LHB?

[Brock Beauchamp]

Anyway, my general point is that LHB are traditionally horrible against LHP in comparison to their RHP performance. Bunching lefties together in the lineup makes the opposing manager's decisions a lot easier from the sixth inning onward.

[Willihammer]

 

That's true about starters but lefties get that advantage either way. It doesn't matter if they're split up or in a row. Yes, bunching them together might impact run-scoring a bit but I was thinking of individual OPS performance, which it won't affect.

 

As for leads/deficits late in games, I suspect the lead/deficit would have to be pretty substantial for a manager not to say "aw hell, three lefties are coming in a row, I'm just going to throw a lefty out there to make sure nothing bad happens". Every pen has one lefty reliever and lots have two of them. Chances are, one of them isn't amazing so why not put him out there to face a trio of LHB?

From an individual perspective there are costs too. How do you split up 3 lefties? Either bump them up, down, or take one of them out of the order altogether. Two out of three choices will reduce the number of PAs against RH starters. Since most starters are RH and most bullpen arms are RH too. If anything I would assume the overall impact of conventional L-R-L-R staggering is to decrease OPS, not increase it.

[crapforks]

 

That's a valid point.

 

And I'm not sure lineup position in regards to handedness has been proven irrelevant. Most conversations I've seen revolve around stuff like whether Dozier should hit second or fifth and whether that matters (ignoring Dozier's home runs, just speaking generally about any player, really).

 

I think handedness matters quite a bit. If you stack three lefties in a row, those lefties are going to suffer in the last third of the game as managers throw lefty relievers at them at least once a game, possibly multiple times a game if possible.

 

If you're a lefty facing LHP 10% more often than a typical lefty bat, your season statistics are going to suffer and the difference won't be negligible. Many LHB hit LHP at a .150-.200 lower OPS. That's not a small difference, that's the kind of thing that can move a player's overall season OPS down .020 or more points.

Is the impact of the lineup construction great enough to move players around in the order to potentially protect platoon hitters? In the 7,8,9,1 of Park, Castro, Buxton, Dozier that I used as an example, that's assumes they belong in those spots. If Buxton is hitting like he did in September, I hope he is not in the 9 spot. I really have no point to all of this except it would be ideal to employ platoon hitters near each other while breaking up their handedness if possible. I would really like to keep Castro and Rosario from slotting back to back in the lineup.

[Brock Beauchamp]

 

From an individual perspective there are costs too. How do you split up 3 lefties? Either bump them up, down, or take one of them out of the order altogether. Two out of three choices will reduce the number of PAs against RH starters. Since most starters are RH and most bullpen arms are RH too. If anything I would assume the overall impact of conventional L-R-L-R staggering is to decrease OPS, not increase it.

True, but that knife cuts in both directions. If you group LHB against righties, that means you're also bunching RHB against righties. The effect isn't as great (usually, anyway) but you're comprising parts of your lineup either way.

 

And in those later innings, the opposing manager doesn't even have to think about it. You've created the perfect storm for him to exploit with left/right pitching matchups.

[Willihammer]

 

True, but that knife cuts in both directions. If you group LHB against righties, that means you're also bunching RHB against righties. The effect isn't as great (usually, anyway) but you're comprising parts of your lineup either way.

 

And in those later innings, the opposing manager doesn't even have to think about it. You've created the perfect storm for him to exploit with left/right pitching matchups.

Well the caveat is "all else being equal." You wouldn't bat Castro in front of Dozier just because he's LH. But you wouldn't swap Park for Mauer against a RH starter just to break up Kepler and Rosario. Even if the opposition brings in a LOOGY later on you only traded one PA versus what, two, three against the starting pitcher?

[ashbury]

Therefore, I'm tentatively going to take the under on a .675 OPS.

So did Seth.

[mlhouse]

I personally am very skeptical of the "pitch framing" analysis.  

 

Two things are really suspect in the analysis.

 

1.   I think it is impossible to truly isolate pitcher quality in these measures.  Without being able to isolate out the impact of pitching staffs it is almost impossible to compare the relative values of "pitch framing".   Even more important, what is called a strike for top level pitchers is going to be different than what is called a strike for a Twins pitcher or another member of a poor pitching staff.  

 

2.  I am also highly skeptical of the run value these analysis assign to each incremental strike.  The .113 run/strike seems to be to high of a value.  I am not going to say that it is impossible, it just seems that it could be a number that is significantly exaggerated.  And this is very believable.  Consider what it takes to measure every pitch a catcher catches, evaluate if it is a true strike/ball, and then compile all that info.  You are going to want it t be important!!!.   

 

But even if we assume that the value of an incremental strike is .113 runs/strike, Jason Castro's pitch framing created 0.92 strikes/game.  At .113 runs/strike that is 0.10 runs per game.  It might be "real", but it is almost immaterial within the error levels of a normal baseball game.