If it's helpful, please consider donating to the fundraiser I am doing for the Pediatric Brain Tumor Foundation (a link is on the top of each sheet, too).
I adjusted the algorithm for these, as it is harder to do the predictions when the field is stacked. For the other psych sheets, looking at previous IM / HIM finish places was enough to predict who most of the top finishers would be. However, with most everyone in Kona having finished in the top-5 at some point in the past year, I took a different approach.
I'm not currently looking at consistency across events or quality of fields with respect to finishing places, although those would be good to add in the future. For now, I calculated each competitor's best "score" as follows, using their best finish place within their division multiplied by a certain factor:
1. Regular Ironman: 1 + 0.5 * number of years before 2017
2. Regular Half Ironman: 2 + 1 * number of years before 2017
3. Ironman World Championship: 0.1 + 0.2 * number of years before 2017
After the equivalent of around 20th place at an Ironman, I factor in USAT scores if they exist.
These numbers are just made up by me, but the thinking was that a top Kona result is worth a lot more than a regular Ironman result.