Back

Is a fair comparison between doping tested and doping untested powerlifters possible?

Khudayarov, A. G.

2021-05-12 sports medicine
10.1101/2021.05.07.21256806 medRxiv
Show abstract

PurposePowerlifting is clearly divided into doping tested amateur competitions and doping untested pro competitions with pro competition money prizes ranging from a few thousand USD to over $120 000 USD in a single competition. However, the results of both pro and amateur powerlifting competitions are then collected and put into the same result database (although there is the possibility to filter the results using doping control status). The powerlifting results are compared by scores to have a fair comparison between different weight class athletes. This study plays with the thought of comparing doping tested and doping untested athletes, and creates a coefficient for the comparison. MethodsThe powerlifting results (noted in kg) of the 10 top ranked athletes per weight class and separately for men and women, and for doping tested and doping untested categories were collected from the openpowerlifting.org database. A weight adjusted model was fit to these results separately for men and women with the doping control status used a binary factor. ResultsDoping control status was a significant factor when modelling the powerlifting results, p<0.001 for both men and women. Mens doping untested results were 1.0725 times as large as the doping tested results. Womens doping untested results were 1.1208 times as large as the doping tested results. These separate and precise factors could be used as coefficients for scoring doping tested and doping untested athletes powerlifting results. Moreover, the coefficients approximate the effect of doping on an elite level athletes results.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.