Background:The proposed model differs from the traditional performance analysis, which considers only the absolute performance (final placing on the competition). In the relative
perfor-mance index, proposed here, we consider how much an athlete was better than others competitors in terms of performance (measured by time). How to measure the relative performance of athletes in
different events, in different years, without only consider the final ranking? How to differentiate an athlete who ar-rives 1 second behind the first-place, from another that comes 1000 seconds behind
Methods:The 10 best Ironman athletes, who participated in the World Championship in Kona, HI, from 1981 to 2013, were analyzed. To enter in the sample, the basic criterion was
de-fined as having completed at least five IM Word Championship in the top 10 rank. For analysis of relative performance, we created a “P index” of relative performance, meas-ured by the athlete's
performance semicovariance, with the average performance of the top 10, divided by the athlete’s average performance semivariance. The use of semico-variance and semivariance was indicated to capture
only performances above the aver-age, since the goal was to measure superior average performance.
Conclusions:The relative performance index can be used to compare athletes who competed at dif-ferent periods. With the traditional model of absolute performance that is not possible,
because we have change in bicycle technology, training strategies and nutrition, for ex-ample. It can also be used to evaluate the relative performance of the disciplines of a competition (in the case
of triathlon, we can divide the index for swimming, biking and running), contributing to the coaches to target the on the training of their athletes. To the relative performance, the consistency of
results is the great goal to be reached. Although the present study has been done with triathlon, this index could be applied in other sports.