Leveling Up: What I learned at the Sabermetrics, Scouting and the Science of Baseball Seminar

[Reverend]

Figured I’d start a new thread to try to create a more focused discussion to share and discuss what was learned at the seminar. It was a really great time and mad props again to Fris and jnai for putting on a great symposium—many persons of some significance in the baseball community were heard mentioning that they had stopped going to things like the Sloan conference but continued coming to this one because it was just better.
 
To begin, a big takeaway for me this year was a general sense of just how much more the more analytically progressive teams know—or think they know—about how to succeed at baseball than those clubs that lag behind and fan population in general. There was a heightened awareness on the part of insiders and of outsider presenters of the notion of the internal analysis the teams had and that they see certain things as key to getting better of which the public is not even aware of, or at least as lagging behind.
 
So there’s a bit of a distinction to be made between the “popular science” of baseball which encompasses fans and websites like FanGraphs, B-ref, B-Pro, THT and the like and the FOs and management of the teams and the different spheres of knowledge. Obviously, we knew this distinction existed, and certainly there is overlap, but it seemed to come across in high relief this year. A good example might be how the work on catcher pitch framing being done by our own jnai and others is bringing us, the fans, and understanding of the value there that is apparently understood by some teams (GO SOX!!) but not by others.
 
Anyway, I’ll now list some of what I remember and hopefully others will fill the gaps of what I missed due to working the merchandise table (awesome stuff available here, by the way—all proceeds go to the Jimmy Fund), fighting with my automobile, or, as Fris insists, sleeping through Farrell’s talk.
 
Also, for some reason many of the SoSHers there would not reveal themselves. Gave off a real Cold War spy vibe.
 
I’ll list the talks out separately to facilitate quoting posts for ongoing discussion.

 

[Reverend]

Dr. Aaron Seitz: See the Ball…
 
Really amazing talk about neuroscience and eyesight that went from “I’m not really a baseball guy” to “OMG THIS GUY CAN MAKE ALL OUR PLAYERS BETTER!!” in an instant. It was more detailed on brain imaging than I can do justice to but by observing brain patterns observed in various sight experiences, they learned that they can “re-program” the brain to see better through visualization therapy run on a computer and without anything invasive like surgeries—the subject spends time over a certain number of days/weeks (I don’t remember) looking at images that retrain the way the brain sees.
 
They hope to derive a variety of therapeutic benefits from this technology beyond sports to things like ADHS and glaucoma and all sorts of stuff. But they have worked with college baseball teams. The findings he presented showed a mean improvement in eye-sight for the baseball team from 20/13 to 20/10, and some people even got down to 20/7.5 which is, of course, nuts. Also, the team in question reduced their K% about 4-5% compared to a league reduction of 1% over the same period. He did acknowledge, however, some selection problems with the experiment as managers balk at randomizing their lineups. There was much muttering about Will Middlebrooks

 

[Reverend]

Baumer, Swartz, Gennaro: The Value of the Win
 
This crew talked about how to value the win. A big takeaway was that the value of a player to any given team in terms of his WAR and what the WAR is worth in the free agent market are very different things. Also, the value of a win is very context dependent as it is worth different amounts based on whether or the team is in the playoff hunt, or even on the value of the team’s TV deal.
 
Basically, what they told us renders the lion share of how people talk about the value of a WAR in the free agent market effectively meaningless.

 

[Reverend]

Jeff Luhnow, Astros GM: Let us do science to it!
 
Luhnow talked at length about the difficulties of getting buy in from players, coaches and managers and such for adjustments made from knowledge gleaned from the new analytics. His main example was using shifts and how players often resisted changing the way they had been doing things their whole life. An example he gave was how pitchers will get irritated if a ball gets through a shift-created gap but not give credit to the shift if a ball that would otherwise be a hit is picked up by the shift. There are also issues like players having to learn new ways of executing double plays and such.
 
Luhnow indicated that this sometimes meant not employing the optimal solution indicated by analytics to accommodate player comfort, which was an interesting insight into the interoersonal process. He also mentioned introducing a bunch of ideas to Tony LaRussa who responded with a variety of questions and potential problems that he had not initially considered which was interesting as well. Ultimately, he indicated that there was a significant issue in “selling” what they thought were compelling insights in how to adjust the game play to improve performance even when they felt they had compelling data. Of some note, he indicated that the Astros have in-house analytics indicating that they have saved a significant number of runs through use of the shift—yet they still get some pushback.
 
Some jerkwad asked about disappointing decisions like drafting a player with a high pick who then goes to college and he answered very diplomatically about making the decisions not with perfect information but with the information you have available; afterwards, jnai noted publicly that we love the Q&A process and requested of the audience they not be jerks.
 
We also learned that some people don’t know how to ask a question requiring less than 3 minutes of speech.

 

[Reverend]

Dr. Stern: CTE, Concussion, Brain Damage
 
Getting hit on the head is bad for your brain. Despite what I had previously perceived as a lack of data as to whether or not sub-concussive hits can lead to CTE, he stated with conviction that they had cases of CTE where there was no record of concussion. Also, football appears to be a fundamentally stupid activity based on his data. I mean, I think we all knew that already, but he did science to it and it’s really, really bad.
 
Also: He is better looking that the guy who played him on CSI or whatever tv show it was.

 

[Reverend]

John Farrell
 
Farrell’s talk was colored by lots of “things haven’t gone as well as we hoped” stuff, but gave another wonderfully candid talk. Obviously, young player development was a big issue, and he said that he thought you could pretty much tell what sort of player you had after 600 PAs. He also indicated a preference for being able to observe a player at a given level for an entire season, which was interesting in juxtaposition with some of the promotion processes the Red Sox have used in recent history. 
 
He also indicated that he thought there was a significantly expanded strike zone this year, and that’s been a major factor in the decline in offense this year. He made a point of being very clear that he does not think this means the umpires are doing a bad job or anything, and that he didn’t know why exactly, be it because of more newer umps due to replay or an internal mandate to speed up games, just that it was bigger and he seemed quite certain of that in ways that suggested he was referring to their own internal tracking. It was interesting to note that implicit in his statements was the notion that there could be a shift in emphasis in umpiring that is not made public knowledge, even to the teams, as he suggested that in this case it could have a disproportionate effect on teams that try to take a lot of pitches and work the count.
 
Was the stupid Doubront question asked of Farrell or Cherington?
 
Beyond what was explicitly stated, perhaps even more so than last year there was a tenor to his responses that suggested that a big part of his role was to be able to straddle the line between ballplayers and the emerging science of baseball and to be adept at getting the ballplayers to do things they might find a bit weird on account of their newness.
 
Oh yeah—somebody asked why he pitched Koji only one inning instead of two given that he’d had two days of rest. He responded graciously and pointed out that Koji was 40 fucking years old and that while he hadn’t pitched for two days, he had pitched 6 of the last 9. It was a great moment, as one SoSHer put it, in terms of revealing how he took seriously his responsibility for thinking about the 162 game season and how we, as fans, can get hamstrung by looking at small windows of time.
 
Farrell also mentioned that he thinks the introduction of the radar gun is the worst thing ever to happen to pitching. He thinks it causes people to emphasize velocity over all else, often to the expense of being a good pitcher. He offered this in the context of his belief that it is over-extension and not number of innings that is the driving factor behind pitcher injury and the increase in surgeries.

 

[Reverend]

Ben Cherington
 
He more than anyone got me thinking of how differently he sees what makes for a successful team than not only the average fan but even other baseball insiders and ball players. There were two main things I took away from his talk and both concerned player development and prospects.
 
The first was in response to a question about how there was all this new stuff about the value of pitch framing by catchers and how the Red Sox seemed to have some catchers that were great at pitch framing and perhaps one of the best in the league and while this could be a coincidence, did the Red Sox have some sort of internal information on pitch framing and/or a sense of how it can be developed into the skill set of a young catcher.
 
Response: “We think it’s totally meaningless and it should never be discussed ever again.”
 
*laughter*
 
*long pause suggesting he wasn’t going to elaborate, leading to more laughter.”
 
“yeswedo”
 
The next big takeaway was how they view prospects. Someone asked who outside of their top ten prospects was someone the Red Sox valued highly. Cherington basically made a mockery of this question, said he didn’t know who the top ten were and asked if anyone else who worked for the Red Sox was in the audience. He found someone, who suggested a name, and Cherington repeated it. Someone else in the audience exclaimed, “That guy’s a Top 50 prospect in baseball!” to which Cherington all but responded:
 

Spoiler

 
Ultimately, between this and the next point I’m going to address, is the point that he was sorta snarkily suggesting that they don’t think about players and prospects so linearly but in terms of probabilistic projections. That is to say, for example, they may have a bunch of guys who they think have a 70% chance of panning out, some with a 50% and so forth. Also, there was a them across several talks about the difference between the “error bars” in projecting player performance and the notion of getting a player to perform at the ceiling of their potential. As such, there is significant overlap in the possibilities of what a player might turn out to be which, Cherington I felt implied, makes a rank ordering of the prospects not only not useful but may even obfuscate how they think about them.
 
Finally, in response to a question about whether or not prospects are overvalued in the context of what went down at the trading deadline, he gave a sense of just how methodically the Red Sox are. He explained that the Red Sox had never been in the “seller” position at the trade line during the tenure of the decision makers. Therefore, they took an historical approach and looked at the history of such trades over the last 10 or so years, particularly with an eye to how the buyers v. sellers did in the long term. Apparently, they discovered that the sellers didn’t do as well, particularly with respect to the prospects they received for full-fledged major league players, as was conventionally believed. Really cool insight into the difference between the perceptions of how teams make out in a trade at the time as opposed to an understanding based on a longitudinal study of the process.
 
He also mentioned that the reduction in offense was in some measure "our fault" (addressing the audience) because the new information available from analytics favored pitching and defense over hitting.
 
It basically sounds like the Red Sox are run by a bunch of baseball scholars while, with a few other exceptions, many of the other teams spend their time throwing rocks at each other. It’s nice to root for an organization you can trust.

 

[Reverend]

Trackman Radar Demo
 
We all went across the street and managed to have nobody hit by the Green Line. A pitcher threw to a catcher and jnai tried hard, but in vain, to prevent people from standing behind the catcher where they might get hit by a rapidly moving baseball.
 
I couldn’t catch everything that was said, but looking at the monitor of the data tracked suggested to some that they might be getting data relevant to the new work being done on looking at “effective velocity” beyond just velocity and movement.

 

[Reverend]

Gennaro, President of SABR
 
Gennaro described the process of getting buy in for analytics from FOs, managers and players, particularly in the context of resistance to new ways of doing things. The talk had a heavy entrepreneurial bent as opposed to illuminating anything about baseball per se. That said, given that Farrell, Luhnow and Cherington have indicated that that’s a not insignificant part of their jobs in trying to build a winning ball club, it’s an interesting part of the process of how to win beyond see the ball, hit the ball, throw the ball.

 

[Reverend]

Goldbeck, Some Baseball Analytics Company
 
Goldbeck described new innovations in biometric analysis of baseball players, including measuring things like torque on the elbow and stuff and possible correlations between different arm slots and breaks or stride length as a function of height and the velocity on the ball. Basically, looking for information to create a science to inform the engineering of baseball mechanics. Crazy. He and Gennaro both stated that they don’t even know the degree to which teams are using this kind of stuff—more insight into how much of baseball analytics is in the black box.

 

[Reverend]

Alan Nathan
 
Explained the physics of hitting a baseball in terms of the collision between the bat and the ball and how different forms of contact (e.g. spots on the bat) generated hit velocity. It was really cool but there’s not good way to do the talk justice. Big takeaway was that bat speed is way more important than pitch speed and hitting the ball at the sweet spot of the bat is way more important than I had realized—even given that I thought it was really important—due to energy lost to bat vibrations off the sweet spot.
 
He also demonstrated the sweet spot of the bat by using a plastic cup glued to the end of bat as a “speaker.” Alan Nathan is cool.

 

[Reverend]

Dan Brooks
 
Talked about Fight Club.

 

[Reverend]

Cross, Glaser, Pavlidis, Lichtman, Gassko: Projection Systems
 
These guys talked about projection systems. Interesting points were made about how much—or little—value you could be extracted from improving even relatively simple projection systems. Also, there was a lot of variance to how much time people thought analytics departments should spend on projections. Whether or not luck was a key component to getting to MLB and the degree to which that was a complicating factor came up. Also, they seemed to have confidence in pitching projections than with batters, The most interesting takeaway for me was the weirdness of trying to project players who might be playing in minor league ballparks with really strange dimensions vis-à-vis major league parks.

 

[Reverend]

I think it’s really interesting that, given the Red Sox’s situation, how much the discussion mirrored projecting the future rather than what it takes to win now—much like difference between the conversations on this board compared to this time last year.
 
OK, that’s what I’ve got off the top of my head from the sessions I attended; hopefully others can fill in the gaps of my memory, and one way or another, there’s a lot here to process.
 
Thoughts?

 

[OttoC]

I'm not sure why using Lasik surgery to artificially improve eyesight far beyond what average people have is not considered on par with PED use.
 
I have read that the average sight for major league players nowadays is 20/12. A person bron with 20/12 eyesight is one thing but a person born with, say, 20/17 eyesight who uses surgery to get his eyesight to 20/12 has done so artificially. And if you are going to scream about players using PEDs to get an edge, shouldn't you do the same about the artificial eyesight improvement? This is not a condemnation of Dr. Seitz's efforts as I was not there for his talk but pushing a healthy body past its normal limits through elective surgery, such as fathers wanting Tommy John surgery for their youngster in the hopes of improving his fastball speed need to be talked about.
 
As an aside to this, are pitchers getting Lasik surgery? If not, perhaps that is one of the factors in why so many of them do not hit well.

 

[pokey_reese]

As someone who was not able to attend the conference, thank you so much for this recap, Rev.  A lot of really interesting stuff to consider.

 

[Laser Show]

One of my favorite presentations was from Tony Blengino, who had developed a metric that measures how well pitchers manage contact. If I understood it correctly, it was set on a 100 point scale, where 100 was average, above 100 was below average, and below 100 was above average. He went through some of the best and worst pitchers of all time according to his metric. It should come as no surprise that Greg Maddux was the all-time best at 59 (I think) and was clear of #2 Tim Hudson by roughly 20 points. I have no idea the metric's effectiveness and I wish I could remember more off the top of my head, but it was very interesting.
 
EDIT: Oh and Javier Vasquez was the 3rd worst all-time, which explains a lot.

 

[joe dokes]

As someone who was not able to attend the conference, thank you so much for this recap, Rev.  A lot of really interesting stuff to consider.

 
+1

 

[ScubaSteveAvery]

I'm not sure why using Lasik surgery to artificially improve eyesight far beyond what average people have is not considered on par with PED use.
 
I have read that the average sight for major league players nowadays is 20/12. A person bron with 20/12 eyesight is one thing but a person born with, say, 20/17 eyesight who uses surgery to get his eyesight to 20/12 has done so artificially. And if you are going to scream about players using PEDs to get an edge, shouldn't you do the same about the artificial eyesight improvement? This is not a condemnation of Dr. Seitz's efforts as I was not there for his talk but pushing a healthy body past its normal limits through elective surgery, such as fathers wanting Tommy John surgery for their youngster in the hopes of improving his fastball speed need to be talked about.
 
As an aside to this, are pitchers getting Lasik surgery? If not, perhaps that is one of the factors in why so many of them do not hit well.

Dr. Seitz' method is non-surgical. It was more or less about how targeted training of the brain can lead to unexpected, but highly beneficial results. The kicker for him was that the impacts of his "training" or "experiment" were long lasting. There were even initial, but unconfirmed, gains made for pitchers.

 

[Reverend]

Dr. Seitz' method is non-surgical. It was more or less about how targeted training of the brain can lead to unexpected, but highly beneficial results. The kicker for him was that the impacts of his "training" or "experiment" were long lasting. There were even initial, but unconfirmed, gains made for pitchers.

 
Excellent point--I forgot to mention that he said that the "training" stopped three months before the season they examined the results in 
 
 

As someone who was not able to attend the conference, thank you so much for this recap, Rev.  A lot of really interesting stuff to consider.

 

 
+1

 
Hey, nothing to it--this place is best when we share what we (think we) have learned.

 

[ScubaSteveAvery]

 
Ben Cherington
 
Ultimately, between this and the next point I’m going to address, is the point that he was sorta snarkily suggesting that they don’t think about players and prospects so linearly but in terms of probabilistic projections. That is to say, for example, they may have a bunch of guys who they think have a 70% chance of panning out, some with a 50% and so forth. Also, there was a them across several talks about the difference between the “error bars” in projecting player performance and the notion of getting a player to perform at the ceiling of their potential. As such, there is significant overlap in the possibilities of what a player might turn out to be which, Cherington I felt implied, makes a rank ordering of the prospects not only not useful but may even obfuscate how they think about them.
 
 

 
This seems to be a really interesting point about how the organization views players and their performance.  Tom Tippett also started his presentation by discussing this.  They certainly view players in probabilistic ranges and try to project if their performance will improve or not.  They also look at the league average and replacement level to see if the player is truly hurting them, or just playing below their expected level of play.  If you think about Bogaerts and look at PECOTA - he's playing between the 10th and 20th percentile.  So his play is not outside the bounds of what was possible when he started the year.  His WARP for the year is 0.0, so even though is play is way below what we all expected, he isn't seriously hurting the team relative to what they could pick up off the scrap heap.  I say this only as an example of the way the team was looking at players evaluation, and how we can possibly calibrate how we look at it in order to understand why the team does what it does. 

 

[ScubaSteveAvery]

I liked the media panel in part because all of the writers for large, traditional papers (Peter Abraham, Alex Speier - The Globe & Evan Drellich - The Houston Chronicle) talked about how they know more about stats than they are allowed to print.  There is a struggle between educating the public and increasing readership.  In a lot of ways, we already knew this, but sometimes a column that may reference a stat that nobody takes seriously anymore is due more to the editor than the writer.  It was Abraham who even mentioned that Red Sox press releases even use fielding percentage, which is a garbage stat, but goes to show how even they cater to stats that normal consumers of information understand.  

 

[Eddie Jurak]

Dr. Aaron Seitz: See the Ball…
 
Really amazing talk about neuroscience and eyesight that went from “I’m not really a baseball guy” to “OMG THIS GUY CAN MAKE ALL OUR PLAYERS BETTER!!” in an instant. It was more detailed on brain imaging than I can do justice to but by observing brain patterns observed in various sight experiences, they learned that they can “re-program” the brain to see better through visualization therapy run on a computer and without anything invasive like surgeries—the subject spends time over a certain number of days/weeks (I don’t remember) looking at images that retrain the way the brain sees.

I will add some
explanation here as I used to work in vision research.

To simplify, you can think of vision as having two components: the optical component, where light enters the eye, strikes the retina, and activates photoreceptors; and the signal processing component, which involved the retinal neurons, optic nerve, and brain. Most of what humans do to improve their vision is on the optical side, for example, glasses, LASIK, cataract surgery, even squinting.

Dr. Seitz' work is interesting because it is focused completely on the signal processing side - what the brain can get out of the visual information it receives.

If you think of the photoreceptors on the retina as pixels, there are some visual tasks that humans can do to much greater precision than one would predict based on pixel size alone. An example of this, hyperacuity or Vernier acuity, involves aligning to fine lines (or other similar tasks). Humans can do this at an order of magnitude greater precision than one would expect based on photoreceptors alone. Clearly, we can only do it because of some kind of analysis algorithm running in the brain.

The interesting part of Seitz' presentation is that he pretty much just used a bunch of standard visual tasks used in vision research for his training program. None of this was really designed to help baseball players hit. One would expect that he should see better results if he can devolop some training tasks that are more related to baseball tasks. Hitting is probably the big area where training should be beneficial, tracking fly balls may be another.

On the other hand, I would venture to guess that part of what makes a big league hitter a big league hitter is the ability to do this kind of stuff. So the room for improvement could be lower.

 

[twothousandone]

Thanks, Rev

It basically sounds like the Red Sox are run by a bunch of baseball scholars while, with a few other exceptions, many of the other teams spend their time throwing rocks at each other. Its nice to root for an organization you can trust.

Rev: how did you come to this conclusion? The few others doesn't include the Cubs, Astros or Yankees, right? Are you just thinking of GMs, or was there a vibe at the conference that a few teams don't and won't get it?

 

[richgedman'sghost]

Figured I’d start a new thread to try to create a more focused discussion to share and discuss what was learned at the seminar. It was a really great time and mad props again to Fris and jnai for putting on a great symposium—many persons of some significance in the baseball community were heard mentioning that they had stopped going to things like the Sloan conference but continued coming to this one because it was just better.
 
To begin, a big takeaway for me this year was a general sense of just how much more the more analytically progressive teams know—or think they know—about how to succeed at baseball than those clubs that lag behind and fan population in general. There was a heightened awareness on the part of insiders and of outsider presenters of the notion of the internal analysis the teams had and that they see certain things as key to getting better of which the public is not even aware of, or at least as lagging behind.
 
So there’s a bit of a distinction to be made between the “popular science” of baseball which encompasses fans and websites like FanGraphs, B-ref, B-Pro, THT and the like and the FOs and management of the teams and the different spheres of knowledge. Obviously, we knew this distinction existed, and certainly there is overlap, but it seemed to come across in high relief this year. A good example might be how the work on catcher pitch framing being done by our own jnai and others is bringing us, the fans, and understanding of the value there that is apparently understood by some teams (GO SOX!!) but not by others.
 
Anyway, I’ll now list some of what I remember and hopefully others will fill the gaps of what I missed due to working the merchandise table (awesome stuff available here, by the way—all proceeds go to the Jimmy Fund), fighting with my automobile, or, as Fris insists, sleeping through Farrell’s talk.
 
Also, for some reason many of the SoSHers there would not reveal themselves. Gave off a real Cold War spy vibe.
 
I’ll list the talks out separately to facilitate quoting posts for ongoing discussion.

Hi Rev...  I want to second the handshakes and fistpounds that are due to Jnai and Fris. They did an absolutely masteful job coordinating all the conference and arranging all the speakers. This was my first seminar and I firmly intend to go next year. Regarding the bolded, I bought one of those wondeful polo shirts on Sunday. However, the gentleman I bought the polo shirt from was not wearing any sort of religious frock or collar so I assumed it was not you. :buddy:   I did get to meet and talk with Chuck (Fris)  on Saturday and he is an  even better person than on the computer. Anyway, thank you, Rev,  and the rest of SOSH for putting together such a fine seminar.

 

[Eddie Jurak]

The jackass who asked Luhnow about Brady Aiken was either trying to ask very very delicately or he was not aware of the medical backstory.

It would have been funnier if he asked him if, when push came to shove, he would have considered trading Luke Harrell for anything less than Luke Giolito. B

 

[ehaz]

Also, for some reason many of the SoSHers there would not reveal themselves. Gave off a real Cold War spy vibe.

 
Figured we'd have a special sticker on our name tags.  I asked the catching question to Ben.  

 

[JGray38]

I learned what to look for in the images if I google a player.

I had a blast, learned a ton, and it was fun to discuss everything with Rev and Scuba Steve during the breaks in the action.

I found Cherington's proactive attempt to avoid the Aiken and Koji type questions that his predecessors got entertaining. "I mean, I guess we can talk about specific transactions if you want, I'd like to talk about what's happening in the game."

Next year we need a code word or secret handshake to identify each other. Or we could just meet by the stairs at lunch or something.

 

[SoxLegacy]

Rev, I have to add my thanks to those posted above. I was unable to attend, but truly appreciate your posts here to share the information you picked up--it sounds like a terrific time with a tremendous amount of interesting material.

 

[maxotaur]

Dan Brooks
 
Talked about Fight Club.

What about it? As it relates to baseball? Seriously. If it wasn't pure irreverence I'm curious.

 

[Jnai]

What about it? As it relates to baseball? Seriously. If it wasn't pure irreverence I'm curious.

 
I actually talked about our catcher framing model published at BP earlier this year. =)
 
[And GIP, which is my new stat that I think will be on BP later this month.]

 

[OttoC]

 
....
 
[And GIP, which is my new stat that I think will be on BP later this month.]

 
Unless GIP stands for Games in Progress, could you dis-acronymn it, please?

 

[Jnai]

 
Unless GIP stands for Games in Progress, could you dis-acronymn it, please?

Gip really needs to be experienced to fully understand it.

 

[StupendousMan]

The first rule of GIP is that you do not talk about GIP.

 

[joe dokes]

 
This seems to be a really interesting point about how the organization views players and their performance.  Tom Tippett also started his presentation by discussing this.  They certainly view players in probabilistic ranges and try to project if their performance will improve or not.  They also look at the league average and replacement level to see if the player is truly hurting them, or just playing below their expected level of play.  If you think about Bogaerts and look at PECOTA - he's playing between the 10th and 20th percentile.  So his play is not outside the bounds of what was possible when he started the year.  His WARP for the year is 0.0, so even though is play is way below what we all expected, he isn't seriously hurting the team relative to what they could pick up off the scrap heap.  I say this only as an example of the way the team was looking at players evaluation, and how we can possibly calibrate how we look at it in order to understand why the team does what it does. 

 
I think this hits a much larger point that even well-educated (baseball-wise) fans should take into account more often.  I think there is often an attempt to fit the actions we see taken by the FO into our own understanding of the game and what we see. I see it as similar to anthropomorphizing our pets' behavior into things we understand about humans. Sure there's some obvious stuff. But it probably only *looks* obvious and there's more to it. Given the Vazquez discussion, maybe the decision with Saltalamacchia is a good example. *We* could see a guy who was hitting his prime, offense on the upswing, seemed to be the type of "leader" a team wants at catcher, and adequate at defense. The Sox presumably saw something different. (Or, you (impersonal you) might just think the FO isn't all that smart/talented -- or not as smart/talented as you are -- in which case none of this discussion really matters to you).
 
This isn't to say -- as I'm often accused of believing -- that the FO is always right. It's more that, in line with the bolded phrase, that when the FO does something "surprising,"  I prefer to search for a reason that I'm not seeing, but that the FO is, rather that going on the presumption that what *I* see is evidence that puts the "burden of proof" (for lack of a better term) on the team.  "Well, since player X had all these things going for him (that I can see), the team obviously screwed up (because they aren't telling me what they saw)." 
 
Did anyone ask, "so, can any of you give us a quick-and-dirty analysis of 'what the fuck Ruben Amaro is doing.'"?

 

[Snodgrass'Muff]

A big "thank you" to everyone involved in bringing this event to life and everyone contributing to sharing the results with those of us who could not attend. It's always great to get a glimpse into the insights that were shared and I hope that one of these years I'll be able to make the trip to enjoy it in person.
 
Great work, everyone.

 

[rmurph3]

The jackass who asked Luhnow about Brady Aiken was either trying to ask very very delicately or he was not aware of the medical backstory.

 
To be fair, the questioner was a high school kid.

 

[Laser Show]

  
Did anyone ask, "so, can any of you give us a quick-and-dirty analysis of 'what the fuck Ruben Amaro is doing.'"?

They were kind of the running joke for the weekend. Someone at the end asked "so we all like to make fun of the Phillies, but just how bad is it?" After some laughter Dave Cameron asked if anyone who worked for the Phillies was there and of course there wasn't. He said he's always wanted to interview Philly's stat guy. "So what do you do all day? Continuously beat your head into a wall? Do you wear a helmet?"

 

[seantoo]

 
This seems to be a really interesting point about how the organization views players and their performance.  Tom Tippett also started his presentation by discussing this.  They certainly view players in probabilistic ranges and try to project if their performance will improve or not.  They also look at the league average and replacement level to see if the player is truly hurting them, or just playing below their expected level of play.  If you think about Bogaerts and look at PECOTA - he's playing between the 10th and 20th percentile.  So his play is not outside the bounds of what was possible when he started the year.  His WARP for the year is 0.0, so even though is play is way below what we all expected, he isn't seriously hurting the team relative to what they could pick up off the scrap heap.  I say this only as an example of the way the team was looking at players evaluation, and how we can possibly calibrate how we look at it in order to understand why the team does what it does. 

I find it interesting that JBJ has about a 1.5 WAR while Xander's is -.02 (Based on Fangraphs today). JBJ's defense more than offsets his offense and Xander's offense is not much better but his defense is not good either. How you get a positive # is irrelevant, and yet most of the attention has been on JBJ, it's as if some don't understand that overall value is what matters most. Don't take this out of context other than to compare the amount of attention each player has received.
The other thing is rookies struggling should be expected, this is what they do and the Trout's of the world are the exception that verifies the rule, not break it. Why is anyone that surprised by rookies struggling?

 

[joe dokes]

joe dokes, on 19 Aug 2014 - 09:32 AM, said:
 
Did anyone ask, "so, can any of you give us a quick-and-dirty analysis of 'what the fuck Ruben Amaro is doing.'"?

 

They were kind of the running joke for the weekend. Someone at the end asked "so we all like to make fun of the Phillies, but just how bad is it?" After some laughter Dave Cameron asked if anyone who worked for the Phillies was there and of course there wasn't. He said he's always wanted to interview Philly's stat guy. "So what do you do all day? Continuously beat your head into a wall? Do you wear a helmet?"

 
Ha. I was about 75% tongue-in-cheek. That's great.

 

[seantoo]

I will add some
explanation here as I used to work in vision research.

To simplify, you can think of vision as having two components: the optical component, where light enters the eye, strikes the retina, and activates photoreceptors; and the signal processing component, which involved the retinal neurons, optic nerve, and brain. Most of what humans do to improve their vision is on the optical side, for example, glasses, LASIK, cataract surgery, even squinting.

Dr. Seitz' work is interesting because it is focused completely on the signal processing side - what the brain can get out of the visual information it receives.

If you think of the photoreceptors on the retina as pixels, there are some visual tasks that humans can do to much greater precision than one would predict based on pixel size alone. An example of this, hyperacuity or Vernier acuity, involves aligning to fine lines (or other similar tasks). Humans can do this at an order of magnitude greater precision than one would expect based on photoreceptors alone. Clearly, we can only do it because of some kind of analysis algorithm running in the brain.

The interesting part of Seitz' presentation is that he pretty much just used a bunch of standard visual tasks used in vision research for his training program. None of this was really designed to help baseball players hit. One would expect that he should see better results if he can devolop some training tasks that are more related to baseball tasks. Hitting is probably the big area where training should be beneficial, tracking fly balls may be another.

On the other hand, I would venture to guess that part of what makes a big league hitter a big league hitter is the ability to do this kind of stuff. So the room for improvement could be lower.

I remember an exercise that everyone's favorite player, Arod, used to do to train his eyes. Focus on an object far away then close up, repeatedly. This appears in short time to increase the speed at which your eyes focus on an object. I've heard that several players, during the off-season, take batting practice with ping pong sized balls as well.

 

[Jnai]

Some of Eric's great photos have been featured in Grantland's article about the event:
http://grantland.com/the-triangle/2014-saber-seminar-mlb-sabermetrics-gets-soft/

 

[Eddie Jurak]

He said he's always wanted to interview Philly's stat guy. "So what do you do all day? Continuously beat your head into a wall? Do you wear a helmet?"

Maybe this is why Fris invited the CTE expert to give a talk?

 

[Snodgrass'Muff]

I find it interesting that JBJ has about a 1.5 WAR while Xander's is -.02 (Based on Fangraphs today). JBJ's defense more than offsets his offense and Xander's offense is not much better but his defense is not good either. How you get a positive # is irrelevant, and yet most of the attention has been on JBJ, it's as if some don't understand that overall value is what matters most. Don't take this out of context other than to compare the amount of attention each player has received.

 
Less than a full season of UZR or dWAR is incredibly unstable. Stats like WAR, which build on defensive metrics are inherently flawed because of that. You can't just throw out "JBJ is worth 1.5 wins above replacenent and Bogaerts is worth -.02" as though that tells the whole story. Without three full seasons of data, defensive metrics don't really tell us much about a player's value, so WAR doesn't work very well, either. Besides, this ignores all of the scouting reports that mark Bogaerts as the much better player overall and as far more likely to reach his projections.
 
There are good reasons for people to be worried about Bradley but not Bogaerts.

 

[absintheofmalaise]

 
Less than a full season of UZR or dWAR is incredibly unstable. Stats like WAR, which build on defensive metrics are inherently flawed because of that. You can't just throw out "JBJ is worth 1.5 wins above replacenen and Bogaerts is worth -.02" as though that tells the whole story. Without three full seasons of data, defensive metrics don't really tell us much about a player's value, so WAR doesn't work very well, either. Besides, this ignores all of the scouting reports that mark Bogaerts as the much better player overall and as far more likely to reach his projections.
 
There are good reasons for people to be worried about Bradley but not Bogaerts.

Good points all.
 
Also, when a player has split time between two positions in a season like XB has, you also should see how those UZR numbers break down by position because FG combines the UZR numbers in WAR. You should also look at how DRS measure the defense and compare that to UZR to see how big of a gap there is between the two systems. One other thing, FG doesn't even list XB on the UZR leaderboard for 3B or SS because he hasn't played enough innings at either position to even make the list. 
 
He currently has 626.1 innings at SS. His UZR is -1.2 and DRS is -7. He has 385.1 innings at 3B. His UZR is -7.7 and DRS is -7. Both measure a player against the average player at that position that season.
 
You also need to look at the numbers for positional adjustment since those are in the formula. CF is +2.5. SS is +7.5 and 3B is also +2.5 

 

[OttoC]

Gip really needs to be experienced to fully understand it.

 
Undoubtedly, but I still have no idea whether it applies to hitting, pitching, fielding, running, throwing, a combination of some or all of those things, or something entirely different, such as which way the manager "dresses."

 

[Frisbetarian]

 
Undoubtedly, but I still have no idea whether it applies to hitting, pitching, fielding, running, throwing, a combination of some or all of those things, or something entirely different, such as which way the manager "dresses."

 
Yes!

 

[Reverend]

Yes!

The GIP that can be expressed is not the eternal GIP.

 

[DaubachmanTurnerOD]

Good points all.
 
You also need to look at the numbers for positional adjustment since those are in the formula. CF is +2.5. SS is +7.5 and 3B is also +2.5

 

And let's not lose sight of probably the most important numbers: 21 and 24 - X and JBJ's respective ages.

(Also - I'd like to add my thank you to Rev for the summaries, and congratulate Fris, Jnai, and everyone for what by all accounts was a tremendous success. I really hope to make the trip someday).

 

[Harry Hooper]

 

 
Farrell also mentioned that he thinks the introduction of the radar gun is the worst thing ever to happen to pitching. He thinks it causes people to emphasize velocity over all else, often to the expense of being a good pitcher. He offered this in the context of his belief that it is over-extension and not number of innings that is the driving factor behind pitcher injury and the increase in surgeries.
 

 
 
Thanks for the recap. Interesting given his recent comments in the Globe:

 

Uehara, a 39-year-old former starter, has said he is comfortable with his workload. But for Tazawa, 28, this is something new.
 
“Very conscious of him,” Farrell said. “You monitor the reaction of the hitter. You monitor the stuff in which he’s throwing. He hasn’t lost velocity. There might be a little bit less consistency to the overall command.
 
Tazawa has been healthy the last two seasons and charts show his velocity has indeed been consistent. But there is a line Farrell has to watch.
 
“You begin to prioritize rest and health,” he said. “That’s not saying that we’re not looking to win every game. But I think of prioritizing and not pushing a guy too far.”