2019 girls Foot Locker predictions and discussion!

Tuohy would have been in a battle for 3rd today.

:)

Don’t think so...Touhy ran 20:11 for 6k, that’s going through the 5k at 16:49 pace—and she did it on a muddy + hilly course.

Methner looks so relaxed while hammering it is spooky.

speed ratings wrote:

astro wrote:

Just to follow up on how blazingly fast the course was running today, three girls managed sub-17, something which had only been done once in 25 years.

The midpoint finishers were about 10 seconds faster this year so a first guess at speed ratings would be around a 166 for Dudek.

169 (same scale as last year)

I would put it around 162-163. I appreciate the speed ratings as a tool but they always seem to fall apart with the Footlocker race. The 180+ rating from years back was way off for example. Between the restricted field and the course they end up inflated. Three runners ran sub 17 and 11th place for example ran equivalent to 4th last year. That is a super fast course.

astro wrote:

I would put it around 162-163. I appreciate the speed ratings as a tool but they always seem to fall apart with the Footlocker race. The 180+ rating from years back was way off for example. Between the restricted field and the course they end up inflated. Three runners ran sub 17 and 11th place for example ran equivalent to 4th last year. That is a super fast course.

Although the times look fast this year, and the ratings high I am not a big fan of a big variance in speed rating vs time for courses where there does not seem to be an obvious factor. NXN in Portland is clearly a good example of where things need to change.

If you look at some of the history, especially at FL, there have been some big variations that don't always seem to make sense on their face.

I think there are just some races where the field runs slow, but it is not because conditions are tough. But that is just me.

That not withstanding, they are a great tool for assessing how runners from different regions might fare against each other. But don't use just any one race result to draw a conclusion.

this shows the problem with speed ratings and using midpoint runners as the benchmark. For a course like footlocker where weather doesn't materially change in a given year, how can it run way faster than prior years? sometimes 20th place or whatever was just a much better run than 20th place the prior year, instead of the course "running super fast". If the speed ratings for a given time at footlocker ever deviates materially from the identical time in a different year (like more than 5-10 seconds), you know there's a problem with the formula.

Foot Locker National Championship (Dec 14) ... Girls - Zofia Dudek (169 speed rating) wins, Marlee Starliper (2nd, 168) and Abby Vanderkooi (3rd, 166) .... Boys - Josh Methner wins handily (201 speed rating)

xc watcher wrote:

this shows the problem with speed ratings and using midpoint runners as the benchmark. For a course like footlocker where weather doesn't materially change in a given year, how can it run way faster than prior years? sometimes 20th place or whatever was just a much better run than 20th place the prior year, instead of the course "running super fast". If the speed ratings for a given time at footlocker ever deviates materially from the identical time in a different year (like more than 5-10 seconds), you know there's a problem with the formula.

They actually turned out the way you thought they should. It is the exact same scale as last year despite the midpoint runners being 10 seconds faster and the winner being 15 seconds faster. So from your perspective there was no problem at all.

And the method is not just based on the midpoint runners. It considers all the runners. Looking at the midpoint runners was something I did as a first approximation of what the final answer would be.

The thing is those ratings have been adjusted to take the overall speed of a race into account. So that is a term that the person who formulates these ratings uses and has even explained over the years why certainly ratings seem to skew lower than the times. That was the explanation when Tuohy was given 167 for running 16:37 in 2018 but 172 for running 16:44 in 2017 at nxn xc nationals.

So a near course record at FL would seem on its face to warrant a high 160s-low 170s SR. But three girls ran sub 17. So either 3 girls ran historically fast races in the same race on the same day or the course and conditions were "fast".

I guess it is above my pay grade. :)

Appreciate the explanation, and makes sense. I was more so responding to the other guy saying the course ran super fast this year simply because times were faster (which on the face of it, clearly doesn't make sense - imagine if someone tried to apply the same analysis to championship track meets saying it "ran fast" or "ran slow" because overall/winning times deviated from prior years).

This was the explanation that was provided in 2018:

"... a speed rating indicates how fast a runner ran relative to other runners.

To derive speed ratings, final times and the time differences between runners are correlated with the known speed ratings of the individual runners and known speed of specific races AND groups of runners (of known quality and speed) ... graphical and statistical comparison are made to determine the speed of the particular race being evaluated ... subjective estimations are commonly necessary in the early season, and speed ratings typically improve in accuracy as the season progresses ... Statistically, it is partly an iteration process. Good historical data is applied and helps when needed. Statistical iteration involves guessing, but experience helps to make better "guesses" to arrive at an acceptable solution."

So if I have this straight we are comparing time differences between runners and comparing that to the "known speed" of a specific race. So if a race runs fast because of course and conditions, which does happen (the course looked like a hard dirt track in the broadcast), that is going to get compared to the course's supposed historic speed rating and inflated.

One more point, but I think important. When we talk about reprodicibility of a statisical result, one of the most important has to be reproducibility in the results of the particular athlete. As these ratings seem to be based on the Beyer speed ratings in horse racing, to use an example, we can trust the speed ratings for Secretariat because they happened over and over and correspond to race records that still stand in all three triple crown races.

With Tuohy we see again repeated comparable performances and the course records to back it up that gives us a measure of confidence in the ratings she achieved.

With the Trotter and Fairchild ratings the issue has always been that they are such outliers per that particular athlete's individua performances that something seems amiss.

So let's take this race's ratings. Before today the highest SR either athlete achieved was 162, and that in a FL regional against what turned out to be a very beatable opponent. Dudek only had a 159.

But now all of the sudden we get a 169 and 168 where they finished within one second of each other where three girls ran a sub 17 in the same race. This again seems to be the definition of a result that is not reproducible.

Anyway.

Anybody realize that the guys started 30 minutes after the girls finished and they did not run fast times? The girls overlapped with the 40 best guys in the country and some of you are trying to figure out if they ran faster than normal. On a normal day, they wouldn't be within 2 minutes of those guys. Tuohy would have battled for 3rd place today. She would have never led the race.

Footlocker fixer wrote:

Anybody realize that the guys started 30 minutes after the girls finished and they did not run fast times? The girls overlapped with the 40 best guys in the country and some of you are trying to figure out if they ran faster than normal. On a normal day, they wouldn't be within 2 minutes of those guys. Tuohy would have battled for 3rd place today. She would have never led the race.

Anyone (else) realize Footlocker Fixer is a fool?

fastTuohy wrote:

If Starliper or Hart feel good, they might be the ones to push the pace. Anyone else feeling good will go with them. Otherwise I think it will be a pack. I do not see Masciarelli, Vanderkooi or Dudek taking the lead early and forcing the pace. After 2 miles, anything goes.

I stand by this

It's still pretty close though

Last year was 17 minutes 164 speed rating

So by the same scale 1645 should have been a 169

So that's what it worked out to

Honestly bill is really good and he's using some pretty good statistical analysis and has a lot of data so I think he's pretty accurate

Can you give a speed rating for Katelyn Tuohy a club cross 6K

Even if it won't be as precise can we get some sort of barometer for where she was at

If she ran a 170 at feds with no competition jogging to the Finish looks like she's still capable of at least a 172

Not sure if she ran all out at club cross but it was a pretty close race and she had the lead close to the end so I'd have to think it was at least a 170