Applied Sports Science newsletter – April 28, 2017

Play up top, play wide, but in U.S. Soccer, you don’t play high through the middle. How Lavelle develops could change that.

… Coach D’Antoni’s not gonna play games with you. He’ll give it to you straight.

So when he told me in November that he was taking me out of the starting lineup, all he said was, “I’m moving you to the bench.”

Bam. No leading into the conversation. No softening the blow. Just like that, I’d lost my starting spot. We were 11 games into the season.

I’ve been a starter my whole life, from rec league in the Indianapolis JCC to high school to Indiana to the NBA. Coming off the bench was completely foreign to me.

… Allardyce appears to have finally ended Palace’s poor defensive record and Townsend is adamant the 62-year-old’s vigorous fitness regime is paying dividends.

“For the first time in my career I’m putting in a good shift defensively and helping the team out,” Townsend told Sky Sports.

“I don’t think there has ever been a time when I’ve played as well as I am now.

“Fundamentally I wasn’t fit enough in the first half of the season, I wasn’t able to get up and down the pitch as much as I would like.

When Dr. Robert Butler joined the Cardinals to help steer them into baseball’s next frontier, one of the priority things his staff wanted to change wasn’t physical, medical, or even habitual, but architectural.

The team’s outdated facilities in Jupiter, Fla., had a floor plan that distanced the athletic training staff, medical staff, and several of their rooms. His office wasn’t even on the same floor. Charged with unifying all those departments, Butler found “nothing impedes a conversation like cinder blocks.” The Cardinals’ increasingly claustrophobic weight room had equipment where it fit, not where it belonged, and there was little room for growth, either from the players or the program. A wall stood in the way.

“We needed breathing room,” says Butler, a Cardinals official.

That wall came down on Feb. 1.

The building continues.

… Whether the designers and leaders of such programs describe their work as building character, promoting positive development, or fostering social and emotional learning, they are eager to learn about promising practices used in other settings, evidence of effectiveness, and ways to measure the effectiveness of their own approaches.

This publication from the Board on Testing and Assessment summarizes a workshop held in July 2016 that reviewed the research and practice relevant to the development of character, with a particular focus on ideas that can support the adults who develop and run out-of-school programs. [pdf download]

PURPOSE:

To assess the impact of match start time and days relative to match compared to the habitual sleep characteristics of elite Australian Football (AF) players.
METHODS:

45 elite male AF players were assessed during the pre-season (habitual) and across four home matches during the season. Players wore an activity monitor the night before (-1), night of (0), one night after (+1), and two nights (+2) after each match and completed a self-reported rating of sleep quality. A two-way ANOVA with Tukey’s post hoc was used to determine differences in sleep characteristics between match start times and days relative to the match. Two-way nested ANOVA was conducted to examine differences between competition and habitual phases. The Effect size ± 90% confidence interval (ES ± 90% CI) was calculated to quantify the magnitude of pairwise differences.
RESULTS:

Differences observed in sleep onset latency (ES=0.11 ± 0.16), sleep rating (ES=0.08 ± 0.14) and sleep duration (ES=0.08 ± 0.01) between competition and habitual periods were trivial. Sleep efficiency (%) was almost certainly higher during competition than habitual, however this was not reflected in the subjective rating of sleep quality.
CONCLUSION:

Elite AF competition does not cause substantial disruption to sleep characteristics compared to habitual sleep. Whilst match start time has some impact on sleep variables, it appears that the match itself is more of a disruption than the start time. Subjective ratings of sleep from well-being questionnaires appear limited in their ability to accurately provide an indication of sleep quality.

Phase 2 pulmonary oxygen uptake kinetics (ϕ2 τVO2P) reflect muscle oxygen consumption dynamics and are sensitive to changes in state of training or health. This study identified an unbiased method for data collection, handling and fitting to optimize VO2P kinetics estimation. A validated computational model of VO2P kinetics and a Monte Carlo approach simulated 2 x 105 moderate intensity transitions using a distribution of metabolic and circulatory parameters spanning normal health. Effects of averaging (interpolation, binning, stacking or separate fitting of up to 10 transitions) and fitting procedures (bi-exponential fitting, or ϕ2 isolation by time removal, statistical or derivative methods followed by mono-exponential fitting) on accuracy and precision of ϕ2 τVO2P estimation were assessed. The optimal strategy to maximize accuracy and precision of τVO2P estimation was 1-s interpolation of 4 bouts, ensemble averaged, with the first 20 s of exercise data removed. Contradictory to previous advice, we found optimal fitting procedures removed no more than 20 s of ϕ1 data. Averaging method was less critical: interpolation, bin averaging and stacking gave similar results, each with greater accuracy compared to analyzing repeated bouts separately. The optimal procedure resulted in ϕ2 τVO2P estimates for transitions from an unloaded or loaded baseline that averaged 1.97±2.08 and 1.04±2.30 s from true, but were within 2 s of true in only 47-62% of simulations. Optimized 95% confidence intervals for τVO2P ranged from 4.08-4.51 s, suggesting a minimally important difference of ~5 s to determine significant changes in τVO2P during interventional and comparative studies.

Do we need a group to certify that IoT products support some baseline security and privacy standards? If so, what are the standards and who is the group?

I get at least one if not handful of pitches a week from people who want to write guest articles about the lack of security in the Internet of Things. They all start the same:

“By 2020 there will be 10 gazillion IoT nodes connected…” The best ones make a few specific recommendations about using authentication and encryption.

… In this blog post I am going to argue that people often apply deep learning models naively to computer vision problems – and that we can do better. I think a really good example is with some of my own work from the first year of my PhD. PoseNet was an algorithm I developed for learning camera pose with deep learning. This problem has been studied for decades in computer vision, and has some really nice surrounding theory. However, as a naive first year graduate student, I applied a deep learning model to learn the problem end-to-end and obtained some nice results. Although, I completely ignored the theory of this problem. At the end of the post I will describe some recent follow on work which looks at this problem from a more theoretical, geometry based approach which vastly improves performance.

The Library Archives at Georgia Tech has a football helmet from the 1930s. It’s made of leather, with a paltry amount of padding and no face mask. Helmets, and other sports equipment, have changed considerably over the years. This evolution of sports equipment is taught as part of a class called Materials Science and Engineering of Sports (MSE 4803R).

The idea for the class came from the Georgia Tech Athletic Association, where Jud Ready is on the board of trustees.

“During exit interviews with student athletes, they said they wanted academically rigorous sports related courses,” said Ready, deputy director of the Institute for Materials who has a shared appointment between the Georgia Tech Research Institute and the School of Materials Science and Engineering. Ready seized the opportunity to create an innovative course that combines his love of sports with his passion for materials.

… The meldonium scandal is not unique. In the early Noughties there were similarly hundreds of positive tests for another substance, an anabolic steroid, nandrolone, across a range of sports from football to tennis, from athletics to martial arts to baseball. Arguably the two most famous positive tests (famous now, at least) belonged to a then football player, now manager, Pep Guardiola.

The respective journeys of Sharapova and Guardiola through the anti-doping prosecution process attest to the uphill battle anti-doping organisations can face. One admitted to the use of the drug, the other didn’t. One was cleared of intent, the other was cleared entirely.

Their stories illustrate how ‘drug cases’ in sport are often far from light and dark. A substance can be completely legal one day, totally illegal the next; legal in one situation, prohibited in another. Sportspeople need to be 100 per cent responsible for any substance in their system, yet are allowed to argue mitigation on why they might not be responsible.

… “The salary drop,” as it’s informally dubbed, spawns a litany of chatter and media content. We see pieces containing various combinations of phrases like “most overpaid,” “best values,” and the biggest “steals” in one direction or another. Though it’s the players’ choice to release this info, it inevitably leads to some awkwardness in one form or another. That includes some not-so-flattering exposure for those perceived to be falling short in performances vs wages.

So how did we get here? Why does this happen in the first place? MLSPU executive director Bob Foose was kind enough to chat with me about the salary drop and related topics this week. The following is a selection of highlights from that conversation.

… Manchester United currently sits in fifth place in the Premier League with 63 points after 32 games. On Thursday, they face Manchester City in the much-anticipated Manchester derby, with a chance to pass their rivals in the standings and stake a claim on fourth place. It’s a chance to salvage what has seemed to be a disappointing season with a momentous win and a real shot at the top four.

While Manchester’s defense has been excellent, the attack has struggled to keep pace. The team only concedes 0.77 goals per game, the second-best mark in the league.1 Only Tottenham Hotspur has been stingier. The attack is averaging 1.55 goals per game, though, only seventh best in the Premier League this season. It’s a testament to United’s strong defense that they’re even as close to the top as they are. The funny thing is that there’s nothing wrong with United’s attack. If instead of looking at the team’s goals, we look at its expected goals, a metric that estimates how many goals a team should have scored on average given where they were taken, the team looks much better. United’s expected goal total is a more respectable 1.74 goals per game, the fourth highest in the Premier League.

A new study provides some clues about when and why some men show signs of respect after a fight.

Researchers from Oakland University said their findings indicate that humans may have evolved psychological mechanisms to signal evaluations of fighting performance. The study, published in the peer-reviewed Personality and Social Psychology Bulletin, examined what factors could predict signals of respect, such as shaking hands with an opponent.

One of the biggest challenges in data analytics is presenting results in a way that’s meaningful to people who aren’t data scientists. As MIT Sloan Master of Business Analytics student Souhail Halaby pointed out, there’s a model that shows that the winner of the Super Bowl can predict the next year’s stock market performance.

While this is interesting, it’s little more than the butterfly effect, Halaby said. “You need human intuition to determine whether the correlation is important,” he said.

Halaby, and more than a dozen other students who enrolled in the new master’s program, learned how to apply a human touch to mathematical models in a five-day class offered during the Sloan Innovation Period last October.