An_apple wrote:
What data is used for these race seedings? How does he get figure out times for specific disciplines? Some athletes have already raced multiple times this year while others haven’t yet raced.
https://www.trirating.com/what-is-thorstens-ttr/ - have a read.
He uses all non-drafting race (>99km) data and reduces the weight of results more than one season ago.
"The weighting is done as by assigning each result a weight based on how long ago the race was . . . for each year that goes by the weight is reduced by a factor of 0.85."
Think
you could, with a bit of effort and interest, work out an algorithm for amalgamating 18km > 21.1km > 42.2km run time and biking 80/90/180km data.
Clearly the environmental, water and road conditions on the day will have an effect, but that will be a relatively uniform factor on every athlete's performance. He also rates races, when there is years of history it's more reliable/useful, to recognise that some races are 'fast' (Dubai, Bahrain, Busselton, Florida, Roth ottomh), and applies that factor to his 'seedings'.
So it's not so much the actual times but the differential between athletes that is worth looking at. He does not factor in emotion or overweight how an athlete's immediate previous race went and discards 'fails' from an athlete's data.