2024 State Finals: Projections vs. Reality

Finally caught up on sleep, time to start looking back on the State Finals.

The first installment are how the projections (D1, D2, D3, D4) fared against the cold-hearted truth of REALITY.

I noted here that I expected to be correct on 7 of the 8 champions, ~13 of the 16 podium teams, and (I guess I didn’t write this) 74 of the 80 top-10 teams. In reality, the projections nailed all 8 champions, 10 of the 16 podium teams, and 71 of the 80 top-10 teams.

In particular, they struggled with 2nd place teams, my theory being that teams in contention, but not favorites, press so hard that they risk suboptimal performances. This allows the patient, well-paced teams to swoop in and earn a trophy. On my end, I think I have to increase the uncertainty to adjust for uncertain conditions and outstanding performances. Although many Michigan kids race in large invites, nothing quite compares to a race track with towering stands, looming over a state title environment. This reflects too in the projected place ranges, where 82% of teams were within the bounds listed beforehand. I’d hope to be more around 90% in the future.

One metric that I track from year to year is the place difference between the projections and reality. In 2023, I had the lowest absolute difference to that point, a 2.23 place difference per team. This year’s version improved on that, standing at an average difference of 2.06 places.

Bad news for you guys, these are getting a little more predictive, so I’m gonna keep doing this in 2025.

There will be more to come on the great individual and team performances of the day.