Application Series 3

WINNING THE FIESTA BOWL

As the football season opens - both collegiate and professional - fans everywhere await the season's exceptional plays. SIR talked with two researchers at the University of Louisville about their involvement with tracking "the exceptional college football player."

Background

While much of the SIR user world is collecting data about such "weighty" topics as medical care, pharmaceuticals, and the space shuttle, two SIR users field telephone calls about weight and mass - from football coaches. Dr Roger Bell, Professor of Psychiatry and Ms Judith Stephenson, Biostatistician at the University of Louisville in Kentucky, began collecting and compiling athlete's measurements and statistics in 1990 - as a hobby.

One Sunday in August last year, Dr Bell considered new ways of managing athlete data. "I'm standing out there and watching them (members of the University of Louisville Cardinals football team) pump and lift and really getting into it. The coaches are out there meticulously marking all these things and taking thousands of measurements: weight, percentage of body fat, height, stretch and reach, 20/40 yard run, 2 mile run; weight lifting measurements: squat, bench seated military, 30 second chin; body measurements: weight, neck, chest and hip. And I say, 'are you putting that a computer?' They say, 'No, we handle it manually.' I ask, 'What do you look at?' And they say, 'Well, we compute it all by hand so not as much as we'd like to be able to look at.'" Bell's hunch was that compiling and correlating these measurements might indicate what makes an athlete exceptional. Since he was already a long-time SIR user, SIR was a natural choice.

Profiles in fitness

Preliminary development work suggests that player profiles could be developed, identifying key indicators for a given position. "One of the planned outcomes, now that we have a position record type, is to try to come up with a profile of what an average athlete looks like," Stephenson says. "It would help us answer the question of, for example, what are the qualities of an exceptional tight end and who doesn't meet those performance measures." Such information could help coaches isolate areas a player needs to emphasise in his training regimen.

Startling Information

To demonstrate the possibilities of this type of analysis, Bell and Stephenson asked coaches what they were looking for in an athlete. At that time, coaches wanted to know about jumping ability. The two began analysing all sorts of data about the athlete: body measurements, weight lifting capacities, performance measures on runs and calisthenics. What they discovered startled many: lean body fat is the single most powerful predictor of vertical jump. Whereas in the past, a regimen of squat thrusts was the trainer's rule of thumb for jumping, now diet and alternative conditioning may be the key factors. Their results have been submitted to a strength and conditioning journal.

Longitudinal View of Athlete

Having the data in the SIR database provide the ability to view an athlete's development over time. "The strength and conditioning people come up here and look through all these figures and immediately see the value in a longitudinal view of the athlete," Bell comments.

To illustrate, he discusses an athlete they've been observing for four years. "Look at this athlete. His weight has gone up, and his military press has gone up to equal his weight. At this point, he was really weak, but had huge gain which put him where he ought to be." Bell continues "His training picture is somewhat erratic. His press capacities fluctuate with the on and off season training."

Being able to see an overall training picture over time can help pinpoint specific training areas which need to be addressed, or perhaps identify inconsistent training habits - key information to strength and conditioning teams.

Bell shares another example. "There's been a steady progression in this athlete's body weight, from 190 lbs to 220 lbs. At the same time, his military presses have gone up - from 290 lbs to 450 lbs - and his squat thrusts have gone up from 350 lbs to 525 lbs. This man's body weight remains rather stable, while his strength capacity increases significantly. His body fat content remains relatively stable, too, at around 8 to 9 percent.

Steroid Use

Detecting steroid use is made a bit easier by using a longitudinal training perspective. "A consequence of steroid use is a decrease in body fat and a rapid increase in muscle mass - since steroids produce lean muscle mass."

Bell says, "There's also an increase in body weight, due to retention of water. So, if the graphs show a gradual growth curve, you're probably not looking at any steroid use. If you see a massive weight increase, steroid use might be the culprit."

Bell says, "We have this data set that starts in 1986 and goes through 1991. We now have records on 140 men, with an excess of 4900 entries that have multiple measurements for each one of those entries. We wanted SIR to manage that, because SIR is really good at doing multiple measurements and at working with different units of analyses."

"SIR was recommended to us for our own research by a biostatistician back in 1982," Stephenson reflects. "The problem we had back then is the same problem presented by the athlete/training project: dealing with different records of varying lengths and differing amounts of data. And SIR handles huge amounts of data and complexity very nicely."

Sweet Success

As for the University of Louisville's Cardinals, they did indeed win the 1991 Fiesta Bowl.

For more information on the use of SIR in sports statistics contact:

Dr Roger A Bell
University of Louisville School of Medicine
Department of Psychiatry & Behavioral Science
Louisville, KY 40292
USA

Back Sir Home