June 30, 2014

SABR Baseball Analytics Conference 2014-Day 1

While I had a week-and-a-half break between the two, it felt like I essentially went right from the MIT Sloan Sports Analytics Conference in Boston to the SABR Analytics Conference in Phoenix:

I touched down on Wednesday at about 3:30. Having come from Minnesota, where we were having an exceptionally “warm” week of days above 30 degrees, the 85 degree weather was both very nice but inconveniently so. Because I had come from Minnesota, I had to wear two sweaters for the 28-degree weather, but didn’t have room in my bags to take them off. So I wore a dress shirt, tie, suit jacket, and sweater on my journey from the airport to the hotel I was staying at in Phoenix. So while I got there a bit sweaty, I did appreciate the view from my 20th-floor room:

I then set my things in the room and headed out to explore Phoenix. My initial plan was to go out and find a place to eat, but when I stopped in one of the tourism centers and the guy in there told me there was a Suns game that night, I decided that I couldn’t pass up going to a new arena.

When I woke up the next morning at about 6:30, I considered going to the Case Competition for the conference momentarily before deciding to wait until the panels themselves began since I had to do things before that like actually eat something since I had forgone that by going to the Suns game.

After the two main conference organizers, Vince Gennaro and Marc Appleman (in that order), made their opening remarks:

the first panel was entitled: Medical Analysis and Injury Prevention. The panelists were:

Dr. Stan Conte-

I actually knew at least *of* Conte from the MIT conference. I’m not sure if he was there this year (if he was, I missed his panel) but I recognized him immediately as medical director of sorts with the Dodgers. (His actual title is Vice President of Medical Services. Close enough, right?)

Glenn Fleisig-

Research Director at the American Sports Medicine Institute, who I don’t recall ever seeing before.

Chris Marinak-

Who I have definitely seen at MIT since the fourth picture when you google search him is one I took, and is the Senior Director of Labor Economics at MLB.

Buster Olney (moderator)-

I don’t think I have to introduce him to any baseball people, but he is a writer and TV personality for ESPN for you non-baseball people perusing this blog for whatever reason.

Buster Olney started right away in this panel (despite being the moderator who is typically simply a facilitator of the discussion) by pointing out that the new innings limits aren’t really as helpful as they are designed to be with the fact that Matt Harvey who was handled “perfectly” still ended up injured, but Tim Lincecum and Chris Sale who came right out of the gate pitching a lot remain uninjured.

Conte I wouldn’t say countered, but added that it’s not so much pitch count but change in pitch count that is more of a factor in looking at injury. He also added something that surprised me, which was the fact that the number of Tommy John Surgeries have stayed pretty consistently around 22 since the procedure came into existence; except for in 2012 when they doubled for one year inexplicably and fell back down to normal the next year.

Glenn Fleisig added that mechanics throughout the pitches is an important tool to evaluate pitch counts through in that a guy throwing x number of pitches with good mechanics is less at risk than a pitchers throwing x pitches with bad mechanics. It was pretty obvious to me, but I think he threw it in for the people who haven’t really looked at bio-mechanics of pitchers at all. I only know the fact about mechanics affecting injury back and forth because I was involved in high school baseball so much. Speaking of high school baseball, Fleisig added that pitch counts have a correlation to injury at the high school level but not at the MLB level.

Marinak came in with the fact that I’ve heard before, but that health evaluation is the next frontier in baseball in that it will be the next place where teams get their competitive advantages. I should mention that the reason (probably) that he was on this panel is that he actually founded a system for keeping track of players’ medical histories that was more or less standardize. Before his system, there was 1. Very little information about any players’ medical histories and 2. The information they had varied based on the thoroughness of the doctor when evaluating a player. And even then, different doctors could label things differently in their records:

I mean you couldn’t even look at something as basic as how many total injuries a given team had in a year. One remark that Conte gave that highlighted the absurdity of the situation was he said a team would say that they had a program that was reducing the number of total injuries, but they wouldn’t even know how to quantify that because they had no proof or indication if that was actually the truth; any truth to this statement was purely anecdotal.

One thing that teams learned was that even though hamstring injuries are the most prevalent in terms of number of injuries, pitcher-related injuries such as Tommy John Surgery were far more devastating in terms of the amount of time they decommissioned players for.

Speaking of Tommy John, Fleisig remarked that while *SOME* pitchers come back throwing faster than before the surgery, the fact that high school pitchers are getting an healthy arm operated on is absolutely absurd. The rest of the panel very quickly backed him up, saying that they had no clue why you would go through an unnecessary surgery just to add a couple of MPHs that won’t come since you weren’t repairing an injured arm.

On looking at injured pitchers/players, each of the three panelists had something to say:

Marinak:

You can’t standardize how long a player is going to need to rehab. This is a hard concept for many GMs to grasp, but because of many individual health factors, you actually can’t say that player x will be back from injury y after 5 weeks, because while that may be the trend throughout the league, player x’s body is individual to himself which may make him more apt to heal in 1 week or 10.

Feisig:

About 50% of the teams in MLB are looking at the biomechanics of pitchers.

Conte:

He and his team at the Dodgers look at Pitch F/X data; specifically the change in Pitch F/X data to determine whether a pitcher may be injured. If they see anything of note in the data, they then look into it.

Then both Fleisig and Marinak said that teams are very secretive about their data and Marinak added that is an impediment to innovation that this information about the players’ medical histories is not in the public domain.

Conte and Marinak then ended the panel with two of the things that I saw as most important:

Conte’s remark was that the 2012 and 13 seasons saw more time lost to injury in baseball than in any other two seasons. His point being that there’s still a lot of room to go in terms of improvements.

Marinak’s point was that nowadays a “medical guy” is what a “stat guy” was ten years ago in the front office in that it is the cool and innovative thing to have.

Then it was time for questions, and I won’t bore you with…well, I’ll bore you with the question I got to ask them and their answer to it plus the tangent it sparked in their conversation.

Mateo’s (admittedly-stupid) Question: “With Tommy John surgeries staying more or less consistent despite advancements in medical knowledge, what is the goal of researching it?”

Conte’s A: To get the number of surgeries down to zero. We are headed in the right direction, but the thing is medical data is dirty data. We just need to get to the root cause, but the goal is still no injuries. While old players say they pitched every day and never got injured, it goes with the saying: “the older you are, the better you were,” and we have no data to say that there were less injuries.

Marinak:

There are today an average of 20 more pitches per game than there were 40-50 years ago because of the emphasis on walking.

There are no new injuries in baseball;it’s been played the same way since it began. There are just injuries that were wrongly listed in the past.

Overhand throwing is also not an unnatural motion. What is unnatural is throwing overhand at 95 MPH 90 times in three hours.

If you would like to watch it for yourself, though, here is a video of the full panel:

END OF PANEL

Now in the interest of your sanity and mine as well as the brevity of these entries, I will only introduce the panel or presenter and give the five most interesting things that they had to say during the panel/presentation. If you want to know more about the panels, let me know in the comments. I took notes to this detail for almost every panel/presentation that I attended. Additionally, I can refer you to an audio recording of each of these.

If you’ve read my longer-form entries for the MIT Sloan Sports Conference in past years, you may have picked up that there were always multiple things going on at one because the conference is so big. Well this conference is a little smaller, so one of the products of that is that while there were sometimes multiple things going on at once, for the most part, the agenda for the conference was not an elective process. Therefore, between the Injury Analytics Panel and this next presentation, the only thing that I had to do was sit in my seat and wait for the next presenter to get on stage. Who was the next presenter and what was he presenting? Well I’m glad you asked, conveniently-inquisitive voice in my head. It was Vince Gennaro, the president of SABR, doing a presentation entitled: “In-Depth Study of Team Chemistry.”

Through Vince’s admission, though, I will say that the presentation was a misnomer. It might be that the end product will indeed be an in-depth study of team chemistry, but what this presentation was was an update at about the half-way point in a process of interviewing former players and coaches. Here are the five most important things from that presentation:

There’s an instinct in the analytics community to disregard something if it isn’t quantifiable.
The conclusion that a baseball player is more self-reliant compared to other sports because of the one-on-one nature of it is an incorrect one. If anything, a baseball player is more reliant on his teammates in baseball.
Once in combat, social cohesion in US troops trumps patriotism by a huge margin in terms of motivational force. This relates to baseball because similarly to troops, they see their teammates more than their families.
The culture in today’s players is that they don’t only want to be told what to do but also want to be told why they are doing it. (Which drives some coaches/managers nuts.)
Gennaro was looking more at the ingredients of team chemistry moreso than the payoff of good team chemistry.

Here is a video of his presentation, if you’re interested in how exactly he went about evaluating team chemistry:

END OF PANEL (Well presentation, but you know what I mean.)

Panel: Numbers Athletes Love and Hate

Eduardo Perez:

Manny Acta:

Kirk Goldsberry:

Aaron Boone:

SABRDay1 Aaron Boone

John Walsh:

I didn’t really take very detailed notes for this panel. If you want, you can listen to the full panel, but I did ask a question towards the end of the panel, and got an interesting response from the panel.

Mateo’s Question: “What would you say is the perception of statistics in the Hispanic players that come over to the Majors Leagues?”

Eduardo Perez: They keep it old school- He said that he had actually just the day before asked around amongst Hispanic players at the Spring Training complexes what their favorite statistic was. All of them answered very basic statistics. He shared an anecdote of him asking Alejandro De Aza. De Aza answered Batting Average, but Perez was befuddled since De Aza was at the time a lead-off hitter for the White Sox and should probably then be more focused on On-Base Percentage.

Manny Acta added onto Perez’s question by agreeing that Latino players were indeed way behind in terms of statistical literacy is concerned, but added that there was a reason for that, saying, “We have to have someone teach us how to say water before we can learn to say On-Base Percentage.”

That was a pretty hilarious moment as the room burst out laughing. If you want to see that or other hilarious moments that occurred during the panel, go ahead and watch the full panel here:

Finally, Brandon McCarthy and Brian Bannister finished out the day with the “Analytics from the Players’ View” panel hosted by Jon Scambi. While I could describe the whole panel in text, here’s a five-minute highlight video of the panel that I put together which ends with them responding to a question that I asked them myself:

And after that, I went to the conference-wide networking reception and met some pretty nice and cool people who were pretty much all from Texas/going to Rice. We ate and talked and then I left to write the first 2,000 words of this entry which as I am writing these words was over three months ago. I’m sorry, but life and other work has taken over. I have no clue when I will get out the entries from the next two days of the conference as well as the two games I have gone to this year at Target Field but they will be out sometime on the future. Until then, I have been doing a weekly baseball/ballhawking podcast that is live every Sunday at 10:30 PM EST/ 7:30 PST. It then goes onto my YouTube Page (linked here) starting fifteen minutes after it ends until the rest of eternity for anyone who cannot watch the podcast live. For the links every week, just tune into my Twitter account (that’s in the sidebar over there –>) and I tweet out the link 10-30 minutes before the show starts.