Statistical analyzation of testing scores

     The physical elements of testing set the baseline for upcoming micro cycles. What about the data collected? What is the purpose of collecting and analyzation of data? For starters, it can determine change in performance of an individual or a group or a performance based on similar individuals tested in the past. The NFL combine for example compares current NFL players to data derived from the combine. For example, player x tested like player Y. We could compare those scores to those of a group. Think positions on this one. If 10 corners ran a 4.4-yard dash than player A ran a 4.3 we have a baseline for comparison. We could also compare those scores to a local/national norm. If the baseline for collegiate football is to bench 300 pounds and player x benches 225 then we know that the player may not be strong enough to compete at a higher level. There are statistical outliers of course. Tom Brady looked gangly at the combine and his career turned out ok.

    We can read the data with descriptive statistics which summarizes large groups of data with different techniques. The mean or average takes all the numbers and averages them together. This is the most common method of collecting data. This can be used as a collective feature of the data; our average lineman can bench 350. Taking the median score is taking the middlemost scores of the collected data. When the data is an even number the average of the two middlemost scores are collected. This technique may be more useful than the average when outliers exist either on the low or high end. Using the mode technique is the score that appears the most; typically, this is the least useful measure of data.

    Within the collected data we would have a range of scores from the lowest to the highest. A more math driven exercise is to find the standard deviation. The formula is fairly complicated and can be found on calculators on the internet. Bottom line is that if the standard deviation is small then the data is similar. If the standard deviation is larger than a wider distribution of data exists. Another way is to measure against a percentile rank. For example, if the 60% baseline is a 315-bench press and an athlete bench over that amount then they are over the 60% threshold.