One of the reasons milesplit changed it to that was due to the shear number of prs set. They had over 85% of the people that raced at this race set either lifetime or season pr. They have NEVER had that happen before without the course not being legit. There is a lot of years worth of data that they can go back to and they looked thru to see if this has ever happened before (nationwide not just ohio) and it had not.
Lets just say they would have allowed the times as legit and all rankings are set and all of a sudden ohio athletes jump into the national rankings as the top few and then we find out the course is short or a turn was missed during the race that no one noticed. How does that make milesplit look as to validity of rankings?
At the risk of receiving more text messages accusing me of maligning Dublin Jerome (honestly folks, I teach several of their athletes, I like their coaches, I do not dislike the school) I am going to weigh in again.
I usually don't worry overly much about course length. I have a regression model I use to compare one course on a particular day to another course on a particular day. Essentially it computes a course adjustment, so that you can compare, say the Celtic Clash in 2020 to the 2018 Midwest Meet of Champions or the 2018 Lancaster Bob Reall Invitational. One of the things it does is to toss out outliers. Like the kid who was sick for the first meet and ran 20:15 then runs 17:25 the next meet on a similar course. The model drops that athlete as an outlier, at least until they have run more races and we can figure out which time should be dropped as the outlier. As a coach, I know that sometimes an athlete can have a huge breakthrough, so it very lenient in terms of allowing times. Even still, that model tossed out over a third of all the results from the Celtic Clash because they were so extreme. After I forced it to consider all of the results I could tell that no matter the length of the course, a lot of athletes ran really fast times. Team spreads were remarkably low and the number of athletes with unusually slow times was really low. Some teams, like the Olentangy Orange boys and Olentangy Liberty girls, clearly ran great races. They are effectively being penalized because no one is talking about their races, just about the times.
For what it's worth, since I was asked, when I measured the course I did not use the line at all (it was a single line down the middle of the course), I followed the (very clear on Sunday morning because of trampled grass) path the runners actually used, using straight lines point to point. I got it to be just a hair under 3 miles, well inside the accuracy for a 3 mile course. For reference, most courses I have measured with a laser rangefinder come up 40-50 meters short of 5000 m. Which is fine because it is within the acceptable margin of error and because athletes never actually run that straight. If a course is that close using the straight lines most of the athletes will run close to or over 5000 meters. So I think the 2.x miles on Milesplit is incorrect, but as Finishtiming pointed out, all they have to go on is the extreme improbability of over 80% of the athletes running PRs and having the PRs be by such large amounts.
A retired coach I know asked this question, if the course is an official length 5K, are we prepared to say that Orange and Jackson are two of the best teams in the history of Ohio? And for what it's worth, Mileslpit's decision will have absolutely no bearing on whether any kid who ran there gets a scholarship or not. Unless you have multiple performances of similar quality you are not going to get a scholarship based off one high school cross country performance. Yes, some schools do have XC time standards for walking on. But they are going to offer scholarships without more evidence of talent. If you have a best time of 15:48 and a second best career time of 16:35, and your best 1600 is 4:34, you are not going to get a scholarship that you wouldn't have also gotten for just the 16:35 and the 4:34.
Instead of arguing about length, how about this? The model I use suggests that to compare a time from 2021 Celtic Clash and 2021 Lancaster Bob Reall Invitational, take a boy whose "baseline" time is 16:30. At the Clash he would run 15:47 and at Lancaster he would run 16:41.
[Technical note: My model is based on my team's races and the races of other athletes who have run against us. So baseline is weighted toward the relatively flatter courses prevalent in Central Ohio. We have run at Pick North, the DeWine meet, Watkins, Michigan State and Lancaster this season.]
