Improvement is a Weak Signal

The cliche goes, “You can’t improve what you don’t measure.” But data-enabled improvement is not automatic, especially when what you would like to get better at is everything, and everything is multi-faceted, but the data measures one thing at one point in time.

It helps to have a plan and a goal. The plan should be flexible enough so that it can evolve to take advantage of new technologies and practices create alternative paths that merit exploration. The goal should be ambitious enough to keep participants interested but still be focused enough that the different facets all make sense in the big picture.

Carnegie-Mellon University has a big inter-disciplinary brain science initiative, BrainHub, that pulls from lots of university research groups. Initial goals focus on developing probes to record neuron activity, effectively turning brain activity into constituent data elements that can be analyzed, decoded and then re-integrated. By starting work at this lowest possible common denominator, progress will be easy to track. Better probes, better data, better analysis and ultimately a clearer understanding of brain fundamentals like the connections between neurons. Eventually there will be a complete map for the brain, 100 billion total neurons made up of 10,000 different neuron types.

Ben Lindberg wrote a recent Grantland article on how the Pirates are improving the team’s ability to integrate data analysis with its on-field play. The plan has been to embed team analyst Mike Fitzgerald with the field coaching staff. And the goal, as far as I can tell, is to make more in-game decisions evidence-based, evidence equalling analytics. So baseball isn’t brain science but things seem to be working out in Pittsburgh.

The Dallas Mavericks are planning a bigger data integration project for this NBA season and it will be interesting to see what the team is able to achieve. In the article Mark Cuban promises that nuances will be the result, whatever that means– “The game is changing,” Cuban said. “If you’re going to look for things, look for little nuanced changes into how the game is played this season.”

Teams win and lose their games, giving them a regular (but imperfect) benchmark on improvement. Individual athletes don’t have such mileposts, yet can be driven to pursue goals with too fast timeframes and little workable planning. Mike Reinold and John Davis have blog posts how this easily lead to overtraining and jeopardize improvement in any sort of timeframe, fast or slow. <i>Men’s Journal</i> has an article on marathon training programs that advises doing the majority of training at an easy pace, and in the process, improving raceday speed.

Applied Sports Science helps to make plans for athlete development that work and seem to be getting better at reading the weak signals to show athletes’ improvement. The field is getting better at getting better, which is what you expect when the science is done right.

Leave a Comment

Your email address will not be published.