Applied Sports Science newsletter – September 9, 2016

I recently joined an applied research group at Georgia Tech, the Wearable Computing Center (WCC).

WCC is interdisciplinary and skilled in both technology development and communication. The group works with industry through contract services or on an ongoing basis. So if you are a sports team that isn’t getting desired results from athlete performance technology, the Center can create an educational workshop that gets your organization on the same page technically. WCC can also develop custom technology to help achieve unmet objectives. If you are a sports technology vendor, WCC can help with content, service designs, user interfaces and business models. Please get in touch if I can tell you more or if you have questions I can answer.

And I have gotten serious about adding content to the blog at http://sports.bradstenger.com where I am writing essays that work on making sense of the rapid and often technical advances in sports science. There is also a searchable archive of past Applied Sports Science newsletters dating back to April 2015. Last week there was a disruption in some newsletter mailings and if you missed any emails, those newsletters are available via the archive.

Thanks.
-Brad Stenger

… Routine is a major factor in baseball players’ lives, whether or not it actually has anything to do with how they perform. If a player thinks a routine is important, even if it is ridiculous and nonsensical, it’s important. And for some players, these routines can hurt more than they can help.

Think of the average starter’s routine. The average starter knows when he is going to pitch, so he knows he’s got at least two practice bullpen sessions (throwing to the catcher accompanied by a pitching coach, but no batter) between his last start and his next. During this time, he’ll work on what he wants to feature against his coming opponents, or fix errors he experienced in his last outing.

How those two practice sessions go can have a major impact on a pitcher’s confidence before his next start. Did he hit his spots? Did his slider break hard? Was his fastball down consistently? What does it all mean?

After former Celtics guard Evan Turner agreed to a massive four-year, $70 million deal with the Portland Trail Blazers in July, Boston assistant coach Jay Larranaga sent him a congratulatory text message.

Turner’s response was not about the contract or the season or even anything related to basketball. Instead, he thanked Larranaga for coming to his home in Columbus, Ohio, last summer. Larranaga had worked out with Turner and taken him and his mother to dinner and even gone to the theater with him to see “Straight Outta Compton.”

To Larranaga, the fact that the visit had resonated so strongly with Turner reaffirmed the effectiveness of the Celtics’ approach to the offseason.

“There’s value in getting to go to a player’s hometown, meeting their families, and just kind of growing our understanding of what their life is like away from the Celtics,” Larranaga said. “It’s building relationships.”

To improve communication between brain and muscle, you need to challenge your body in a way that demands such communication.

As the S.A.I.D. principle states: the human body Specifically Adapts to Imposed Demands.

In other words, you will need to stress the body neurologically if you want it to improve neurologically.

Three great ways for runners to achieve this are: speed work (flat), hill sprints and strength training.

… Determining if you are overtraining is a hard task. There’s no exam that can help you. The best you can do is to recognize the signs, try to catch the general symptoms early on and then rest and recover.

Overtraining affects musculoskeletal, immune, endocrine, cardiovascular, nervous, and hormonal systems. So, this is not just a physical phenomenon, you will also have psychological and emotional effects.

Common symptoms of overtraining include poor energy levels, poor performance, inability to finish a workout and persistent muscle soreness.
During overtraining, you may have a higher than normal heart rate while resting or sleeping and this can cause you trouble sleeping (insomnia). Your immune system will also be weakened and your susceptibility to infections will be higher.

Decision making in elite level sporting competition is often regarded as distinguishing success from failure. As an intricate brain-based process it is unlike other physical processes because it is invisible and is typically only evidenced after the event. This case study shows how an individual achieved great success in elite level professional football through consistent positive decision making on and off the field of play. Through prolonged interviewing, Gary Neville, a player from Manchester United Football Club, explored personal behaviours, the structure and activities of deliberate practice and his professional choices in match preparation. His career-long devotion to purposeful organised practice was focused on cognition, physical preparation, context-relative physical action and refined repetition to optimise his mental comfort while enhancing his match day performances. This approach was underpinned by diligent personal and collective organisation and by concerted action. Results provide an insight into the categorical nature of his deliberate practice, sport-specific information processing and match-based decision making. At the operational level, his process was mediated by a complex mental representation of ongoing and anticipated game situations; these representations were continuously updated during each match. Allowing for the limitations of the design, implications are provided for developmental and educational coaching, match preparation, deliberate practice activity and improved use of the performance analysis software packages in professional football.

A UC Santa Barbara researcher studying how the brain uses perception of the environment to guide action has a new understanding of the neural circuits responsible for transforming sensation into movement.

“Mapping perception to a future action seems simple,” UCSB neuroscientist Michael Goard. “We do it all the time when we see a traffic light and use that information to guide our later motor action. However, how these associations are mapped across time in the brain is not well understood.”

In a new paper, published in the journal eLife, Goard and colleagues at the Massachusetts Institute of Technology make progress in mapping brain activity in mice during simple but fundamental cognitive tasks. Although a mouse’s brain is much smaller than a human’s, remarkable structural similarities exist. The mouse brain is composed of about 75 million nerve cells or neurons, which are wired together in complex networks that unerlie sophisticated behaviors.

Before Apple’s first smartwatch went on sale last year, Wall Street analysts had high hopes for the sharp looking wearable device.

On average, they expected Apple would sell over 20 million in 2015 and 33 million this year. Revenue estimates were $10 billion for 2015 and $17 billion for 2016.

It hasn’t turned out nearly so well. The first Apple Watch suffered from limited battery life, slow apps, and a somewhat confusing user interface, leading to mixed reviews and slow sales after an initial burst. Apple shipped only 11.6 million watches last year and just 3.1 million in the first half of this year, according to estimates by International Data Corp. (Apple doesn’t disclose its watch sales)

By contrast, sales of cheaper and more focused fitness trackers have continued to boom. Fitbit is by far the leader in that segment, trailed by the likes of Jawbone and Garmin. Fitbit, which unlike Apple discloses sales every quarter, sold 21 million devices last year and 10.5 million so far in 2016.

Up against Serena Williams in the US Open quarterfinals later Wednesday, Simona Halep will be relying on a piece of cutting-edge technology to get an edge over the 22-time major winner.

A smart court.

The fifth-seeded Romanian has improved her serve by lowering her ball toss after her coach, Darren Cahill, introduced Playsight’s SmartCourt video and analytics technology in practice in March.

This paper proposes a new approach to automatically quantify the severity of knee osteoarthritis (OA) from radiographs using deep convolutional neural networks (CNN). Clinically, knee OA severity is assessed using Kellgren & Lawrence (KL) grades, a five point scale. Previous work on automatically predicting KL grades from radiograph images were based on training shallow classifiers using a variety of hand engineered features. We demonstrate that classification accuracy can be significantly improved using deep convolutional neural network models pre-trained on ImageNet and fine-tuned on knee OA images. Furthermore, we argue that it is more appropriate to assess the accuracy of automatic knee OA severity predictions using a continuous distance-based evaluation metric like mean squared error than it is to use classification accuracy. This leads to the formulation of the prediction of KL grades as a regression problem and further improves accuracy. Results on a dataset of X-ray images and KL grades from the Osteoarthritis Initiative (OAI) show a sizable improvement over the current state-of-the-art.

Many parents may be following outdated recommendations when caring for a child who has a concussion, potentially making the symptoms worse, according to the results of a survey commissioned by UCLA Health.

Of the 569 parents polled across the country, 84 percent supported restricting children with concussion symptoms from all physical activity, 77 percent would likely wake a child throughout the night to check on him or her, and 64 percent would take away a child’s electronic devices – including cell phones – if symptoms lasted more than one week.

What is a concussion? Probably not what you think it is. In this talk from the cutting edge of research, bioengineer (and former football player) David Camarillo shows what really happens during a concussion — and why standard sports helmets don’t prevent it. Here’s what the future of concussion prevention looks like.

Nathan Eovaldi stood by his locker at Yankee Stadium the other day, all dressed up with nowhere to pitch. Eovaldi, the Yankees right-hander, was in full uniform for the official team photograph but will not pitch for the rest of this season, or next. He underwent his second Tommy John operation last month.

The scourge of pitching injuries struck again on Wednesday night in Washington, when Stephen Strasburg winced in pain and left his start in the third inning with elbow discomfort. Strasburg had Tommy John surgery in 2010, and a strained flexor mass was found in a magnetic resonance imaging exam on Thursday. The Nationals said the injury was not season-ending. But until he returns, Strasburg, who is 15-4 and signed a $175 million contract extension in May, is another damaged pitcher in an industry full of them.

“I don’t know, I really don’t,” Eovaldi said, struggling to explain why so many pitchers seem to break down. “There’s guys that don’t throw as hard and they need it. There’s guys that don’t work as hard as others, and they’re fine. Other guys work hard, and they need it. I don’t know what the good combination is. It’s just one of those things, I feel like — good mechanics and staying healthy, it’s hard to do.”

Where do 2016 college football players come from? Jake Sharpless shows us on the blog Rukkus, which is filled with some cool, interactive maps that paint a picture of college football’s geography entering this season.

Moneyball was a game-changer in raising general awareness of the possibilities for data analytics in elite sport.

Always remember that Moneyball is only “based on a true story” and does not provide an authentic representation of how data analytics developed at the Oakland A’s.

The conflict between scouting and analytics is exaggerated for dramatic effect.

The real lesson of Moneyball is the value of an evidence-based approach. This goes beyond the immediate context of player recruitment in pro baseball to embrace all coaching decisions in all sports.

What does “causality” mean, and how can you represent it mathematically? How can you encode causal assumptions, and what bearing do they have on data analysis? These types of questions are at the core of the practice of data science, but deep knowledge about them is surprisingly uncommon.

If you analyze data without regard to causality, you open your results up for the possibility of enormous biases. This includes everything from recommendation system results, to post-hoc reports on observational data, to experiments run without proper holdout groups.

I‘ve been blogging a lot recently about causality, and wanted to go through some of the material at a more technical level. Recent posts have been aimed at a more general audience. This one will be aimed at practitioners, and will assume a basic working knowledge of math and data analysis. To get the most from this post you should have a reasonable understanding of linear regression and probability (although we’ll review a lot of probability). Prior knowledge of graphical models will make some concepts more familiar, but is not required.