Applied Sports Science newsletter – September 20, 2019

Dustin Byfuglien followed the normal path as training camp approached. He moved back to his NHL city, brought his family with him, enrolled his children in school. He participated in informal skates with teammates, readying for the real thing in the second week of September.

Physically, Byfuglien was in Winnipeg. Mentally, he was somewhere else.

Byfuglien’s 42 games in 2018–19 were his fewest since he became an NHL regular more than a decade ago. The biggest issue was an ankle injury that cost him 30 games and never fully healed. Even though he had eight points in six games during the first-round loss to St. Louis, Byfuglien never fully seemed like the dynamic wrecking ball who could singlehandedly change a series — one of the most fun players the league has to offer.

Now, as you’ve read, he’s away from the team, and what happens next is anyone’s guess.

It’s not as dramatic as the mock chart, but you can see subtle dips and rises with, especially around college age and post-retirement.

However, the binary color scheme limits the detail, and there’s of course a good amount of variation in how people sleep. So here’s a wider color scale:

Looking at a gradient, we see a more detailed picture of sleep schedules. Although it’s still not as dramatic as my mock chart.

In their motivation to improve player health and fitness, teams are starting to explore ways to meaningfully incorporate state-of-the-art sensors and monitoring equipment in their pursuit of excellence and a vital edge over opponents.

“As little as ten years ago our ability to track how a player moves on the court during games and training sessions remained technologically challenging,” Dr. David T. Martin, the former Director of Player of Performance Research and Development at the Philadelphia 76ers, tells the Leaders Performance Institute. “Whereas today, sport scientists can access fairly precise data streams that quantify how players move when they play basketball.” With those challenges being surmounted, data validity and reliability is leading to an evolution in the way teams approach practices, games as well as the notorious NBA travel schedule.

Player health and fitness, whether that means player availability or a return-to-play protocol for a convalescent player, are seen as the key areas of growth across the NBA. Chad Gerhard, the Applied Sports Scientist at the Orlando Magic, has witnessed this evolution and excitement from the inside. “There is so much data available right now and not just performance data but basketball-specific data,” he tells us. “We have a really robust analytics staff here that I work closely with and the information they pop out is very much absorbed by the front office, which is fantastic.”

Youth baseball players who participated in a weekly 10-minute prevention program experienced fewer shoulder or elbow injuries and improved performance compared to those that performed traditional warm-ups.

BACKGROUND:

Anterior cruciate ligament (ACL) injury prevention programs (IPPs) are generally accepted as being valuable for reducing injury risk. However, significant methodological limitations of previous meta-analyses raise questions about the efficacy of these programs and the extent to which meeting current best-practice ACL IPP recommendations influences the protective effect of these programs.
PURPOSE:

To (1) estimate the protective effect of ACL IPPs while controlling for common methodological limitations of previous meta-analyses and (2) systematically categorize IPP components and factors related to IPP delivery to assess the validity of current best-practice IPP recommendations.
STUDY DESIGN:

Systematic review with meta-analysis.
METHODS:

A systematic search of 5 electronic scientific databases was conducted to identify studies testing the efficacy of ACL IPPs. Studies were included if (1) the intervention aimed to prevent ACL injury, (2) the incidence rate (IR) or other outcome data that made it possible to calculate the IR for both the intervention and control groups were reported, and (3) the study design was a prospective randomized controlled trial (RCT) or cluster-RCT.
RESULTS:

Of the 2219 studies screened, 8 studies were included in the quantitative synthesis, and their analysis revealed a significant reduction in ACL IR when athletes received IPPs (IR ratio = 0.47; 95% CI, 0.30-0.73; P < .001). The majority of included IPPs tended to meet minimum best-practice recommendations and incorporated plyometric, strengthening, and agility exercises along with feedback on proper landing technique. However, the specific exercises included in each IPP and key factors related to IPP delivery were highly variable. CONCLUSION:

Despite limiting the analysis to only high-quality studies and controlling for time at risk and potential clustering effects, the study showed that ACL IPPs had a significant protective effect and reduced injury rates by 53%. However, significant variability in the specific exercises and the manner of program delivery suggests that ACL IPPs may be able to be designed within an overarching best-practice framework. This may allow practitioners the flexibility to develop IPPs that meet the specific characteristics of the target population and potentially increase the likelihood that these programs will be widely adopted and implemented.

Purpose

To describe individual sleeping patterns and nocturnal cardiac autonomic activity of National team female soccer players during an international tournament.
Materials and methods

Twenty elite female soccer players (aged 25.2±3.1 years) wore wrist actigraph units and heart rate (HR) monitors during night-sleep throughout 9 consecutive days (6 day-time training sessions [DT], 2 day-time matches [DM], and 1 evening-time match [EM]) of an international tournament. Training and match loads were monitored using the session-rating of perceived exertion (s-RPE) and wearable 18-Hz GPS (total distance covered [TD], training and match exposure time, and high-speed running [HSR]) to characterize training and match loads.
Results

Individually, s-RPE, TD, exposure time, and HSR during training sessions ranged from 20 to 680 arbitrary units (AU), 892 to 5176 m, 20 to 76 min, and 80 to 1140 m, respectively. During matches, s-RPE, TD, exposure time, and HSR ranged from 149 to 876 AU, 2236 to 11210 m, 20 to 98 min, and 629 to 3213 m, respectively. Individually, players slept less than recommended (<7 hours) on several days of the tournament, especially after EM (n = 8; TST ranging between 6:00–6:54 h). Total sleep time coefficient of variation (CV) ranged between 3.1 and 18.7%. However, all players presented good sleep quality (i.e., sleep efficiency ≥75%; individual range between: 75–98%) on each day of the tournament. Most of the players presented small fluctuations in nocturnal cardiac autonomic activity (individual nocturnal heart rate variability [HRV] ranged from 3.91–5.37 ms and HRV CV ranged from 2.8–9.0%), while two players presented higher HRV CV (11.5 and 11.7%; respectively). Conclusion

Overall, this study highlights the substantial individual variability in sleep and HRV measures, suggesting the adoption of an individual approach to monitor sleep, training and match loads and recovery, to better understand how players cope with highly demanding competitions. [full text]

Change of direction (CoD) has been indicated as a key mechanism in the occurrence of anterior cruciate ligament (ACL) injury during invasion sports. Despite these associations, assessments of knee function in athletic populations at the time of return to sport following ACL reconstruction (ACLr) have often focused on strength and single-leg hop tests, with a paucity of evidence to describe the CoD characteristics. Therefore, the aim of this narrative review was to describe the movement strategies exhibited following ACLr during CoD tasks and to critically analyze the range of tests that have been used. Specifically, we examined their ability to identify between-limb deficits and individuals who display a heightened risk of secondary injury and/or reductions in their level of pre-injury performance. MEDLINE, PubMed and SPORT Discuss databases were used and 13 articles were identified that met the inclusion criteria. Examination of the available literature indicates that current field-based practices are not representative of relevant sport demands and are unable to effectively assess knee function following ACLr. Laboratory-based studies have identified residual deficits and altered movement strategies at the time of return to sport, and this in part may be related to risk of re-injury. However, these assessments exhibit inherent limitations and are not practically viable for monitoring progress during rehabilitation. Consequently, alternative solutions that are more-aligned with the multitude of factors occurring during CoD maneuvers in chaotic sports environments are warranted to allow practitioners to ‘bridge the gap’ between the laboratory and the sports field/court. This approach may facilitate a more informed decision-making process with the end goal being, a heightened ‘return to performance’ and a lower risk of re-injury. [full text]

Charlie Young has been fascinated with both astronomy and computers for as long as he can remember, so when he started applying to colleges, he jumped at the chance to combine the two topics in a blended arts and sciences program at the University of Illinois—Urbana-Champaign. The Naperville, Illinois, native applied and was accepted to the school in 2016, joining its CS+X program, which allows students in various disciplines to earn a single degree that incorporates classes in computer science into their chosen major.

Young, who will graduate in May 2020, plans to use the “CS” portion of his undergraduate training to pursue a career in baseball analytics. While at the U of I, he has secured internships doing everything from web development to data mining for the Houston Astros, Baltimore Orioles and Cincinnati Reds.

… The new facility is quite impressive. The practice courts have facial recognition technology running at all times, meaning individual players can go back and look at their shot charts after shootaround.

While pointing to changes along the Lake Michigan shoreline, Mark Dantonio quietly hinted at things that could be in Michigan State’s future.

As in improved football facilities.

Alan Haller, MSU’s deputy athletic director, told the Free Press on Thursday that he and Dantonio were part of meetings last week to discuss planning for upgrades to help the program, in the short- and long-term.

We propose marathon running as a novel domain for recommender systems and machine learning. Using high-resolution marathon performance data from multiple marathon races (n = 7931), we build in-race recommendations for runners. We show that we can outperform the existing techniques which are currently employed for in-race finish-time prediction, and we demonstrate how such predictions may be used to make real time recommendations to runners. The recommendations are made at critical points in the race to provide personalised guidance so the runner can adjust their race strategy. Through the association of model features and the expert domain knowledge of marathon runners we generate explainable, adaptable pacing recommendations which can guide runners to their best possible finish time and help them avoid the potentially catastrophic effects of hitting the wall.

Researchers at Carnegie Mellon University have developed a computer model that can translate text describing physical movements directly into simple computer-generated animations, a first step toward someday generating movies directly from scripts.

Scientists have made tremendous leaps in getting computers to understand natural language, as well as in generating a series of physical poses to create realistic animations. These capabilities might as well exist in separate worlds, however, because the link between natural language and physical poses has been missing.

Louis-Philippe Morency, associate professor in the Language Technologies Institute (LTI), and Chaitanya Ahuja, an LTI Ph.D. student, are working to bring those worlds together using a neural architecture they call Joint Language-to-Pose, or JL2P. The JL2P model enables sentences and physical motions to be jointly embedded, so it can learn how language is related to action, gestures and movement.

… Players might change direction a few times per point and more than a thousand times per match. To quantify the physical impact of how often each athlete reverses course, IBM and the USTA developed a metric called Red Steps, or redirect steps. This metric is the foundation for IBM’s enhanced player workload monitoring in the latest iteration of its Watson-powered tennis tool. Coach Advisor tracks both overall physiological load and shorter-burst mechanical intensity. Players returning serve generally “are working harder because they’re running and tend to have less control of the game,” says Elizabeth O’Brien, IBM program director for sports partnerships.

The sheer quantity of redirections is not the only metric that matters. “It sounds weird, but it’s the direction of the change of direction,” says Dan Kant-Hull, Marquette’s assistant director of sports performance. Kant-Hull used to oversee the use of Catapult GPS devices with the NFL’s Cleveland Browns and, for the past three seasons, did the same with Marquette’s men’s and women’s tennis programs. Right-handed tennis players, he learned, are more apt to go from left to right because they want to set up their forehand. Knowing the split can inform training programs. Another takeaway: “There’s more of vertical component than we ever give tennis credit for,” Kant-Hull says. While rotational movements dominate the game, tennis players get airborne on serves and forehands more than one would naturally expect.

Conversion of raw data into insights and knowledge requires substantial amounts of effort from data scientists. Despite breathtaking advances in Machine Learning (ML) and Artificial Intelligence (AI), data scientists still spend the majority of their effort in understanding and then preparing the raw data for ML/AI. The effort is often manual and ad hoc, and requires some level of domain knowledge. The complexity of the effort increases dramatically when data diversity, both in form and context, increases. In this paper, we introduce our solution, Augmented Data Science (ADS), towards addressing this “human bottleneck” in creating value from diverse datasets. ADS is a data-driven approach and relies on statistics and ML to extract insights from any data set in a domain-agnostic way to facilitate the data science process. Key features of ADS are the replacement of rudimentary data exploration and processing steps with automation and the augmentation of data scientist judgment with automatically-generated insights. We present building blocks of our end-to-end solution and provide a case study to exemplify its capabilities.

Heading into the season to mark the silver anniversary of their championships, the Rockets come off the best three-year regular season record in franchise history without having broken through for a return to the NBA Finals.

They have been retooled again, this time with the rotation largely returning intact around one enormous change, with every intention of contending for a title.

With Russell Westbrook replacing Chris Paul next to James Harden in the backcourt, the core of the rotation all under contract for at least two more seasons and the Western Conference considered even deeper with contenders than during the Warriors’ reign, Rockets general manager Daryl Morey addressed some of the questions heading into next week’s start of training camp.