These simulator runs compare different ability spreads in the fleet, and how to best assess the outcomes of events with such ability spreads.

The first graph shows the five ability spreads compared using the 'egalitarian' discrepancy weighting method.  In this method, all boats weight equally, the also-rans given equal weight with the top boats.  This graph shows that, if equal weight is to be given to all boats regardless of ability, then the more promotions per heat the better to obtain better (lower) discrepancy scores, and that is particularly true the more closely spaced the majority of the boats in their abilities.

The second graph shows a linear weighting method, where the top boats are given a higher weighting.  The result is similar to the egalitarian weighting method, but where boat abilities are reasonably spread through the fleet, then there is not such a need to have promotions above 4 or so per heat.

The last graph shows the effect of weighting the top boats very highly, and more or less disregarding the also-rans.  Because the top boats do tend to do well in the simulation, this weighting method gives low discrepancy scores in general, and does not distinguish particularly well between different fleet ability spreads.

©2024 Lester Gilbert