Friday, February 16, 2018

Neuroscience Can Project On-Base Percentages Now | FanGraphs Baseball

Image result for Neuroscience Can Project On-Base Percentages Now | FanGraphs Baseball

Once this type of data can be incorporated into scouting and player development, there will be less draft mistakes and better hitters. 

Subject: Neuroscience Can Project On-Base Percentages Now | FanGraphs Baseball

Neuroscience Can Project On-Base Percentages Now | FanGraphs Baseball

Neuroscience Can Project On-Base Percentages Now

I have an early, hazy memory of Benito Santiago explaining to a reporter the approach that had led to his game-winning hit moments earlier. "I see the ball, I hit it hard," said Santiago in his deep accent. From which game, in what year, I can't remember. Also, it isn't really important: it's a line we've heard before. Nevertheless, it contains multitudes.

We know, for example, that major-league hitters have to see well to hit well. Recent research at Duke University has once again made explicit the link between eye sight, motor control, and baseball outcomes. This time, though, they've split out some of the skills involved, and it turns out that Santiago's deceptively simple description involves nuanced levels of neuromotor activity, each predictive of different aspects of a hitter's abilities. Will our developing knowledge about those different skills help us better sort young athletes, or better develop them? That part's to be determined.
A team of researchers spread across Duke ran baseball players from two full professional organizations through a battery of nine tests on Nike Sensory Stations to measure different aspects of a player's sensory motor abilities. After creating something similar to Major League Equivalency lines for each player, the researchers were able to test the effect of each of the scores against real-life baseball outcomes.
"If you have a 23-year-old, completely average outfielder, the model predicts that his on-base percentage in the major leagues would be .292," explains Kyle Burris, one of the researchers on the project. "The model would expect a similar player who scores one standard deviation higher on the perception span task to have an OBP of .300."
The high-level, easy takeaway from their study is that these skills, taken as a whole, are predictive of good plate discipline. There was no link to slugging percentage, though, so we're not quite yet predicting full batting lines from your neuromotor scores.
But if you drill down a bit into these new findings, you'll see that there is a great deal here to get excited about. Here's a profound image that shows how each subsection of the larger skill set was linked to baseball outcomes. Darker colors denote a stronger relationship between the skill and the baseball statistic.
A table of findings reprinted with permission from Kyle Burris, Kelly Vittetoe, Benjamin Ramger, Sunith Suresh, Surya T. Tokdar, Jerome P. Reiter & L. Gregory Appelbaum "Sensorimotor abilities predict on-field performance in professional baseball" in Scientific ReportsTake a look at the row labeled "perception span," in particular, and you find an interesting story. That task was linked to good on-base percentages and strikeout rates, but not necessarily good walk rates.
"It's kind of like a game of Simon," says Burris as he tries to explain the perception-span task, "but for a split second, it gives you shapes that appear in various aspects of your peripheral vision, and you have to determine was there a square there, or a pentagon there, and it flashed at you in a split second and you have to try and remember what the shape was."
When we asked players what they see when the ball is released, a good portion of the responses detailed how little is ultimately visible to the eye. And there's that study of cricket which suggests that cricket players get more from information they gather before the release of the ball than after. This finding fits right in: players who are good at noticing things on the periphery — like the way a forearm might look different on a breaking ball, or the way the body might drag on a changeup — are better at making contact.
Hidden within the other differences between the tasks and their links to outcomes is a similar story: both the ability to suss out quickly the difference between shapes seen both near and far, and also to capture a target quickly were both good for making contact. That makes sense.

But why would hand-eye coordination be better for player's walk rate than his strikeout rate?
Partly, this could be because players have to start their swing before they know if they want to swing — a requirement velocity puts upon them — and hand-eye coordination helps them to better stop that swing if the pitch is a ball.
Partly, this could be a result of the limited capacity for actually testing hand-eye coordination. The particular task linked to that number requires respondents to tap baseballs as they appear on a screen, testing how fast they can do so.
"I'm not sure that it actually goes and tests hand-eye coordination," admitted Burris, who is headed to Cleveland for a summer internship with the Indians. "There is a little bit of hand-eye coordination in that you have to see it and then immediately translate that to a motor response, but I'd say that that was almost response-time-esque."
If you look at the separate reaction-time outcomes, you'll see a similar link to walk rate, so maybe that's the key skill in taking walk. Reacting quicker.
Or there's another way to separate the skills. You could consider the first three tasks — visual clarity, contrast sensitivity, and depth perception — as "hardware." They're linked to outcomes, of course, because there's a decent part of the game that requires good eye sight. But they're the sort of thing with which you're born.
"There will never be a blind ballplayer," said co-author Gregory Appelbaum.
Those other six tasks, though? They represent the software of our neuromotor system. They represent our ability to take the visual information given to us and process it. Software is more malleable, subject to updates. Software can be changed for the better.
"There is evidence that these processes can be improved," agreed Appelbaum. "There have been demonstrations of neuroplasticity in these processes."
Appelbaum pointed to two interesting studies that pointed to the fact that our neuromotor system's software could be trained. A study from 2015 of which he was part showed that "significant learning was observed in tasks with high visuomotor control demands but not in tasks of visual sensitivity," for one.
A 2014 study at the University of California-Riverside found that actual baseball outcomes could be improved by using a "perceptual learning program." In that study, players reported improvements such as being able to see further, and having eyes that felt stronger and didn't tire as quickly.
Appelbaum is ready to find out what these visual training technologies will look like as we go forward. He's helping launch the Duke Vision Sports Center, a clinic and lab where researchers will use sensory stations, immersive reality, and more, in order to pursue this line of thinking.
When it comes to new stats coming out of Statcast, I've personally seen a change in how players assess the numbers. Early distaste has given away to curiosity, as more players — Yonder Alonso and Andrew Heaney, for example, in my own experience — now speak up at the end of interviews to ask me about launch angle, exit velocity, and how they can use that data to train and improve.
So, while the Boston Red Sox have long been using the link between neuromotor skills and baseball outcomes in their minor leagues in an effort to bring "neuroscouting" to their own organization, these new findings offer a different use for neuromotor study. Instead of sorting players, there's major potential to use these activities to develop players and get the most out of them.
There may never be a blind baseball player, sure. But that's just hardware. Let's see how we can make the most out of our favorite player's software.

Sent from my iPhone

Post a Comment

Giants Top Minor League Prospects

  • 1. Tyler Beede 6-4, 215 RHP from Vanderbilt projects as top of the rotation starter when he works out his command/control issues. When he misses, he misses by a bunch.
  • 2. Chris Shaw 6-3. 230 1B Lefty power bat, limited defensively to 1B, Matt Adams comp?
  • 3. Bryan Reynolds 6-2, 210 OF Switch hitter with average speed and polished hitting approach. Fits Giants mold of high-floor, low-ceiling prospects.
  • 4. Stephen Duggar 6-1, 170 CF Another toolsy, under-achieving OF in the Gary Brown mold, hoping for better results.
  • 5. Sandro Fabian 6-0, 180 OF Dominican signee from 2014, shows some pop in his bat. Below average arm and lack of speed should push him towards LF.
  • 6. Aramis Garcia 6-2, 220 C from Florida INTL projects as a good bat behind the dish with enough defensive skill to play there long-term
  • 7. Heliot Ramos 6-2, 185 OF Potential high-ceiling player the Giants have been looking for. Great bat speed, early returns were impressive.
  • 8. Garrett Williams 6-1, 205 LHP Former Oklahoma standout, Giants prototype, low-ceiling, high-floor prospect.
  • 9. Heath Quinn 6-2, 190 OF Strong hitter, makes contact with improving approach at the plate. Returns from hamate bone injury.
  • 10. Seth Corry 6-2 195 LHP Highly regard HS pick. Was mentioned as possible chip in high profile trades.
  • 11. Jacob Gonzalez 6-3, 190 3B Good pedigree, impressive bat for HS prospect.
  • 12. C.J. Hinojosa 5-10, 175 SS Scrappy IF prospect in the mold of Kelby Tomlinson, just gets it done.
  • 13. Shaun Anderson 6-4, 225 RHP Large frame, 3.36 K/BB rate. Can start or relieve
  • 14. Garett Cave 6-4, 200 RHP He misses a lot of bats and at times, the plate. 13 K/9 an 5 B/9. Wild thing.

2018 MLB Draft - Top National HS Players

  • 1. Ethan Hankins 6-6, 215 RHP Forsyth Central HS (GA) Mi 90's FB tops at 96-98, plus breaking ball. Vanderbilt commit.
  • 2. Kumar Rocker 6-5, 250 RHP North Oconee HS (GA) Heavy 98 FB, sharp mid 90's slider. Vanderbilt commit.
  • 3. Matthew Liberatore 6-5, 200 LHP Mountain Ridge HS (AZ) High 3/4 arm slot, 91-93 FB tops at 95, with good feel for pitching. Arizona commit.
  • 4. Slade Cecconi 6-4, 195 RHP Trinity Prep HS (FL) High 90's FB tops at 97, with mid 80's breaking ball. Miami commit.
  • 5. Carter Stewart 6-6, 200 RHP Eau Galle HS (FL) Highest spin rate breaking ball in draft. Mississippi State commit.
  • 6. Luke Bartnicki 6-3, 210 LHP Walton HS (GA) Low 90's FB with command, workable slider. Georgia Tech commit.

2018 Top MLB College Draft Prospects

  • 1. Brady Singer 6-5, 200 RHP Florida Sergio Romo-esque slider from whippy low 3/4 arm slot. Mid 90's FB, sharp slider and change-up. 3.4 K/BB rate.
  • 2. Casey Mize 6-3, 210 RHP Auburn Forearm issues, 96 FB with split/slider mix, 6.2 K/BB ratio.
  • 3. Logan Gilbert 6-6, 205 RHP Stetson Loose arm action, 3 pitch mix, 93-96 FB 3.2 K/BB.
  • 4. Ryan Rollison 6-3, 200 LHP Mississippi Smooth delivery from 3/4 arm slot, 89-93 FB tops at 94/95. Late 1st, early 2nd rounder. 2.8 K/BB rate.
  • 5. Shane McClanahan 6-1, 175 LHP South Florida Thin build, 3/4 arm slot, tall and fall delivery. 93/96 FB range. 3.0 K/BB rate.

2018 Top MLB HS Draft Prospects in Tampa Bay Area

  • 1. Connor Scott 6-4, 180 OF Plant HS (FL) Florida commit.