Phillies Acquire 1B Justin Bour from Marlins

[phillychuck]

Yeah I think it is odd the Phillies (if Gelb is right) rate Kingery higher defensively (and at SS specifically, it seems).

I don't know how this could be anything other than a reporter mis-interpreting a comment. Minor league teams play their best infield defender at SS, and Crawford has played SS throughout his minor league career. Kingery was tried at 3B his first year as a pro, then moved back to 2B because he struggled there. As far as I know he never once played SS in the minors despite playing on teams with no "prospect-level" SS on the roster. The truth probably is that Kingery is a better defender at 2B than Crawford is at SS, but the two positions have different defensive value.

[Philsmania]

Yeah I think it is odd the Phillies (if Gelb is right) rate Kingery higher defensively (and at SS specifically, it seems). There's just not much data either way, and a lot more minor league data on Crawford. If they really think that than andy is right, they're just not that into Crawford and would sooner play Machado or Kingery at SS next year. They were willing to try Crawford at 3B but Franco ended that for now.

But no question, at this point winning now means playing the best players and sending either of them to AAA for two weeks (that's all that's left) doesn't matter either way). In chat Gelb also predicted everyone will come up 9/1, helping the Pigs in the playoffs won't be a factor. It will just be interesting to see who of Kingery, Crawford and possibly Florimon can make the playoff roster as the bench SS.

It wouldn't entirely surprise me to see Crawford a part of a package traded to a rebuilding team for an in-his-prime outfielder like a Jose Abreu (just to throw an example out there).

I could see them moving Crawford before a big free agent signing and then, if they sign Machado, hastily trading Hernandez somewhere.

[andyb]

Yeah I think it is odd the Phillies (if Gelb is right) rate Kingery higher defensively (and at SS specifically, it seems). There's just not much data either way, and a lot more minor league data on Crawford. If they really think that than andy is right, they're just not that into Crawford and would sooner play Machado or Kingery at SS next year. They were willing to try Crawford at 3B but Franco ended that for now

I think part of it is they don't like Crawford too much. I also think it is very possible that they have a bunch of kids experimenting with new analytical data on defense and they don't really know what they are doing. Their mixed results on shifts are an indication of that.

They also don't care about defense that much. Cabrera clearly cost us one of the two runs on defense last night by not handling the throw on the SB.

Crawford clearly has not been his best at SS with the Phillies. He played mostly 3B last year (very well) and part of his SS time this year was littered with throwing errors in part because of a shoulder injury. If they are looking at SSS analytics from this year Crawford probably grades poorly. This is where you need a real scout to explain to the 24-year-old MIT guy that his data may not tell the entire picture.

[Lucky_Beagle]

This is where you need a real scout to explain to the 24-year-old MIT guy that his data may not tell the entire picture.

Why would a real scout need to explain anything to a data guy? Wouldn't the job of a data guy be to gather information and then give it to the right team personnel who would then decide what to do with the information?

[zambonir]

Well that's the question - is Gelb basing that on conversations he's had with people (including the GM and manager) or is it just informed speculation based on a small - but still accurate - set of facts? Those facts being that Crawford was on the verge of being benched before he got hurt the first time and that he couldn't win the SS job back from Kingery when he returned.

Short-term, it may be correct that Kingery, simply because he's been playing regularly, is a better defensive replacement at SS than the rusty, error-prone Crawford we've seen this season.

I still don't understand/recall what this shoulder injury was. He went on the DL with a strained forearm on April 29 after suddenly experiencing numbness. No indication he was playing hurt or struggling that month (except for the fact that he was actually struggling). Then he broke his hand.

In any case, it's a lost season for him, whether due to injury or just poor play, and whether he gets to stay up for the next 16 days or not. I don't think the front office - whether it's the GM or the data guys - need to be told this season is a SSS.

[teabag259]

I think both about and Santana could be trade bait in the offseason. Move Hoskins back to 1B and go sign a FA OF.

[andyb]

Why would a real scout need to explain anything to a data guy? Wouldn't the job of a data guy be to gather information and then give it to the right team personnel who would then decide what to do with the information?

Because the data guy is looking at small sample data and may not be experienced enough to know what is biased and what is significant. They are inventing stuff now and I imagine some of it is good and some of it is less good.

Of course they have a process where information from the scouting and data side get combined and then Klentak's people make a decision. People are less good at handling all this data than they make out to be.

My world (politics) has numerous examples of this. Why the Clinton campaign decided to skip campaigning in Wisconsin and Michigan is a classic example of this. They had "data" saying they would win those states easily. But they ignored non-data people, the ones on the ground talking to voters, who told them that was not necessarily true. In that instance the political equivalent of scouts were right and the data people were wrong. The data that the data people used was numerically correct. They were just asking the wrong questions and not seeing the bigger picture.

I am a data person. But I also have a healthy distrust of inexperienced data people who think they are always right. The Phillies seem to downgrade the importance of defense. I have pointed that out without judging whether it is right or wrong. I assume the Phillies have data people saying they will net more runs from this strategy and that is why they are pursuing it. But I imagine an honest answer from the Phillies would be they are "not sure" about this anti-defense strategy. They'll never say it, but it is likely true.

[phillychuck]

It's interesting, Andy, the idea I've seen you speculate about on the board that the Phils are downgrading defense. Going back to Moneyball, the first market inefficiency Beane exploited was, of course, walks and their contribution to OBP. Once his ideas became public knowledge, walks were no longer undervalued, and the rumor was that Beane now believed that defense was undervalued. I know I've read quite a few articles speculating that defense is the new market inefficiency. Wonder if the Phils disagree with that, or if they are actually sophisticated enough to believe that the defense market is now overvalued and the slugger who is limited defensively is now a market bargain?

[Austinfan]

The problem with most data people these days is they're poorly educated, too much statistical training is either mathematical proofs or hands on how to use stat packages.
I focused on econometrics in graduate school not to be an econometrician (I wasn't that good), but to understand the intuition behind statistical models.

My favorite econometrician was George Box, and I love his famous quote, "all models are wrong, some models are useful." The problem is most people who use data don't understand the intuition behind the models, apply them mechanically, and don't know how to properly interpret the results. Gee, it passed the significance tests.

I realized early that all significance tests are seriously biased with real world data (experimental data is another world altogether, because you have more control but other ways to bias your results), because left out variable error is ubiquitous. It's an issue that's ignored because it's too ugly to solve.

There's nothing wrong with analytics, as long as the analyst has the proper humility to appreciate the limits of the data, and is vigilant in searching for other data, including qualitative data (i.e. scouting input). My belief is you need three things, a model, data and a story. The story doesn't accept causality as presented by the analytics, rather, it attempts to identify the mechanism of causality, and by doing so, hopefully will expose the limits of the analysis and avenues for further research to verify or reject the story. That is, it's not enough to say the data suggests causality runs from Variable A to Variable B if you can't explain HOW A causes B.

[andyb]

I think they feel that some stats like defensive WAR are overrated because there are fewer balls in play and even among balls in play the defense is less important because of the shifts they do. All of that could be correct (though I note that shifted defenders are still playing defense so skill matters there too even if we have no idea how to quantify how well an individual in a shifted defense plays.

Their theory that defense is overrated may be true. Whether they can quantify correctly (or better than other teams) is a different and unanswered question.

[phillychuck]

Andy, is the "devalued defense" idea speculation based on evidence, or is it supported in quotes from the administration? Either is OK, I'm just curious...

[andyb]

Just actions and evidence. Klentak is pretty good at saying nothing to the press though I guess we can see they are now using defensive substitutions for Cabrera and Hoskins more often, so they kind of realize what they are doing offensively has a defensive consequence.

The whole Santana move out of the box this winter showed the importance they placed on an offense that works counts. I think they believe there is an added benefit that does not show up in individual WAR stats that represents the cumulative effect of a patient offense on team WAR. And that cumulative positive effect was greater than the defensive hit they would take with Hoskins in the outfield. Their quotes at the time glossed over the latter of course, saying Hoskins would get better with experience. Then the TV crew gushes this line at every opportunity. The reality though is that Hoskins is pretty bad out there. No idea how much the patient offense helps, as teams often see a benefit of going to the bullpen earlier anyway (no 3 times through the lineup). I'd venture to say that this strategy probably works against teams with bad bullpen depth but may not work against the better bullpens.

[zambonir]

I think both about and Santana could be trade bait in the offseason. Move Hoskins back to 1B and go sign a FA OF.

Options are always good. But right now it is more likely that if they trade Santana they won't move Hoskins back to 1B, Bour would play 1B. Santana wouldn't be easy to trade without eating money. But I think they are quite happy with him, low BA aside. In the clubhouse as well, I know we often scoff at that stuff but he's been given a lot of credit for mentoring Franco. He's just miscast as a clean-up hitter.

They can certainly afford Bour is they like him. Putting a LHB at 1B 40 times a year to try and make Santana more effective (both platoon-wise and rest-wise) also not a bad thing.

[allentown]

I think putting Bour at 1B 40 times a year to have the LHB at 1B and make Santana more effective is problematic if they stick with the 4-man bench. Counting Hoskins, Santana, Cabrera, and Bour that's basically 4 guys who should play 1B or DH. Of course, Cabrera may not be back, but he seems to have more flexibility than Bour. It will be good to have all of these guys for September and the post-season, but then 40-man spots need to be opened.

[zambonir]

Likely true but you just never know. Maybe the 4-man bench doesn't stick next year. But Cabrera is probably gone. For the sake of this speculation he definitely is, whereas Bour is controlled until 2021 (and costs less money, hard to say what Cabrera might rate as a FA).

But we won't really know anything until they do or don't get Machado and do or don't decide what Kingery and Crawford's future is. Obviously if they do non-tender Bour, or trade him for nothing much, six weeks of him now is still a good return for six weeks of Howie Kendrick last year.

[ButlerBulldog]

Per the Phillies Twitter feed, Eflin is being optioned to Lehigh Valley to make room for Bour. He’ll be back before his next start. I’d rather Kingery had been set down so that he can play every day, but I guess Kapler wants him available for late-inning substitutions.

[zambonir]

Clever move to expand the bench until the doubleheader though.

[el-pietro]

He is due to pitch again in the Double Header on the 16th of August. I'm sure this was clarified the last time but is he allowed to come back up again as the 26th man in that scenario? Would have have to go down again immediately after? His next start would be due on the 22nd.

[zambonir]

I imagine he will come back as the 26th man, go back down, and then they will have to make a real move on the day of his next start, at which point his 10-day clock will have expired.

And of course he's not leaving the team.

[julio]

He is due to pitch again in the Double Header on the 16th of August. I'm sure this was clarified the last time but is he allowed to come back up again as the 26th man in that scenario?

The answer here is "yes." A player on optional assignment can be brought up as the 26th man, and this does not reset his 10-day option clock... so optioning Eflin now, with the intent of bringing him up as the 26th man in a DH, and then recalling him for his following start... just works. And frankly, kudos to the front office for realizing that the 10-day option limit, and the 26th man rule, can be used this way; it's a creative way to maximize roster space.