Hey everyone. While I feel that the intent is there, there are several things that I see that worry me about the study. Sorry for the long post

   1. BMI is not the best way to track analyze someone’s health. A person who is 6′3″ and weighs 200 pounds has a BMI of 25 (which is the cutoff for healthy/unhealthy) regardless of whether or not the person has a fat percentage of 7% (healthy athlete) or 32% (obese).
   2. BMI doesn’t factor in level of hydration at several important times for this study. Hydration is not considered at weigh in time or at time of injury. Both of these are incredibly important when evaluating health and evaluating injury risk factors. Dehydration has been shown to be an injury risk based through several mechanisms in the body including changes in tensile strength and neuromuscular pathways. BMIs will be fluctuate severely based on the amount of water present in the body.
  3. The heights and weights are not measured at the same time of the day in order to standardize the process.
  4. If we assume that BMI is a proper way to evaluate health in athletes, can we really trust that the heights and weights on the rosters are accurate? Using the Red Sox as an example since I follow them the most, can we safely assume that Brad Penny, David Ortiz, and Aaron Bates are all 6′4″ and 230 pounds while someone like Masterson weighs in at 260. Even if the weights are correct can we assume that Pedroia is 5′9″.
   5. This study doesn’t explicitly remove players who have suffered multiple injuries. If a player suffers two DL stints of 25 days each to the same body part (think hamstring, shoulder) in a very short time (say 55 days from beginning to end of the 2 DL stints), it changes the average days lost to a lower number rather than counting it as 50 days when in fact it is the same injury.
   6. Categorizing someone as healthy versus hurt based only on DL transactions is ignoring a lot of data and also ignoring a lot of injuries. What about those pitchers who get their start skipped and don’t go on the DL or have their start pushed back a couple of times? Based on a 35 start season, every start counts roughly as 3% of his season. If he has 2 starts skipped over the 6th month season, that’s 6% that’s lost but not counted in the DL transactions.
   7. The 120 IP mark does not take into account those pitchers who came up to the bigs for 3-4 starts, went back down to the minors and was hurt in the minor leagues. If it does take those pitchers into account, it does not explicitly say so.
   8. It also appears that the 120 IP mark doesn’t factor in injuries for that season. If a player is used to throwing 180 IP, gets hurt that season but still ends up throwing 125 IP and then goes on the DL in year +1, he counts just the same as a guy who throws 125 IP for 3 years and gets hurt in year + 1.

I guess the biggest issue I see is using BMI as one of the variables.
  

I disagree on the importance of player size. I think it’s much more important to measure someone’s health/fitness level rather than size. My reasoning why i think a players size has little to do with injury prediction is below.

1. Modern medicine is fairly advanced now in taking the stress of additional weight off of a players lower body & spine. This can be done through crutches, walking boots, running in the pool rather easily in your example.
2. It’s been shown that a person’s level of fitness and health prior to an injury plays close to as much a role as the rehab after the injury/surgery. This is why players routinely wait several weeks to a month performing rehab before going into a surgery. The same applies to everyday injuries. If the player happens to be heavier but is in better overall shape and health, that heavier person will most likely be able to rehabilitate faster than a lighter person who is not in as good shape.

 Also, a person who is a better athlete (even if heavier) will more likely react to the comebacker in such a way that it will lessen the force transmitted into the body. A heavier person can have a reaction time that is much better than a lighter person and will therefore be able to protect themselves more and decrease that chance of serious injury.

It’s been shown multiple times that correct training of any body tissue will alter that tissue to the demands placed on it. Thus, If a pitcher correctly works on his strength, reaction times, visual training exercises, and reflexes they are less likely to get injured in the first place and when an injury does occur it is likely to be less severe.

3. Other than the issues I raised above the problem I see of using BMI is the issue of being precise versus being accurate. From “Choosing and using statistics” by Calvin Dytham:

“Accuracy is the closeness to the the real value…..Precision is the closeness of repeated measures to the same value. It is possible to have data that are very precise but very inaccurate. For example your balance gives exactly the same value for repeated measures of the same object but they are all overweight because the balance was not calibrated properly.”

Trying to measure a person’s health with BMI is like using a balance that is not calibrated properly as it doesn’t factor in age, sex, body fat%, bone mass, hydration, and fitness level. Can anyone possibly determine how healthy an athlete is without evaluating these traits?

4. When making predictions, I want them to be able to stand up in a court of law….beyond a reasonable doubt, being accurate and precise. I don’t want to convict someone based on an inaccurate eyewitness account, even if that person repeatedly told me the same thing.