Friday, February 21, 2014

Technology, Information, & Expectations

I am realizing more and more that technology must be used with caution.  I've always been curious why triathlon coach Brett Sutton doesn't even use a stopwatch when training athletes. I gained some potential insight when I read What the Dog Saw by Malcolm Gladwell.  I've learned that technology can provide an overwhelming amount of information,  lead to poor predictions, and be demoralizing.  The issues occur when information, that may not be fully understood is used to make predictions.

Technology can lead to information overload and fail to provide clear answers to dilemmas. In What the Dog SawMalcolm Gladwell brings this issue to light through examining satellite images.  Over the years, as the images revealed more detail, more people and time was needed to analyze the detail.  Which still didn't solve the problem because the satellite had to be pointed in exactly the right place and that place wasn't always obvious. In a study on predicting horse races, those who were given some data performed better than those who were given none, but as the amount of data increased, performance actually began to decline.  

Technology can lead to poor predictions. As mammogram images became crisper, there became more to understand and predict.  Even with high resolution images, Physicians don’t know if what they're looking at has already metastasized, if it is a new lesion, benign, or what's hidden behind a duct or connective tissue.  All of which makes predictions difficult and why false positives are so high, 50-60%.  Surprisingly our sense of touch is very discerning, even more so than vision, and highly recommended for early detection of breast CA. 

In athletics, understanding of information is also difficult to interpret and predict future results. Lactic acid testing was thought to measure the build up of lactic acid, which supposedly impaired performance.  We now know it's actually a fuel source and high levels mean the body is making more than the muscles can use.  Unfortunately, even as technology advances it still does not give us a definitive answer on what to do.  A low heart rate could mean that one is exercising at an easy pace OR they're over-trained and can't elevate their heart rate.  A high power output could mean that one is over-exerting themselves OR their muscles are stronger.  Weather, sleep, or even a small suppression of one's immune system could also influence data. Brett Sutton says technology is "very misleading, we play the human game not the metronome training angle."  For Brett, data is not a good predictor of future performance.  He says it could also be potentially demoralizing or provide a false sense of security.  During a race great athletes push beyond what they are "supposed" to be able to do. They rise to a higher level.  Too much data could set up an artificial barrier.  Using data in training, as a predictor of future performance, could also be misleading.

Demoralization may be one of the worst drawbacks of technology because it has the potential to lead to rash decisions. If we try to hit a certain speed, wattage, or time then fail, we're often devastated. But what does the ability to go fast or push a certain wattage in February mean for one's race in June? Are we patient and stay the course OR do we rush to get fast OR do we slack off because we are already fast? The same could be said of quarterly profits predicting long term financial stability. We're devastated or we have a false sense of security, both of which could lead to poor decisions.

So what does one do? USE TECHNOLOGY CAUTIOUSLY. In the book How to Measure Anything, Douglas Hubbard says, our problems are not as unique as we think, we have more data than we think, and we need less data than we think. In other words, don't make it more complicated than it needs to be. Start with fully understanding your dilemma before trying to measure some obscure variable. Sometimes we ask the wrong questions and the answers are of little use. From there assess what you already know. We often know more than we think and can estimate a range with 90% confidence by eliminating absurd answers. Finally ask yourself, how much value will measuring X add to my knowledge about the future or influence my decision. Is it economically justified? Often we need 'just enough' info to reduce uncertainty and move forward. Rarely do we need 100% certainty. I am of the belief that we don't have time to do everything. Therefore focus on a few relevant things, that you are confident in, that you have control over, and do those to the best of your ability. Try new things, but don't spend an exorbitant amount of time deciding. Think, Decide, Do. In athletics, pain management is still and always will be the name of the game. Listen to your body, train accordingly, and you will excel.

No comments:

Post a Comment