I’m not disputing that it wasn’t short for everyone that ran.
That said, if you are comparing performances from one course to another, that would be a limitation. It’s why XC times shouldn’t matter and why people shouldn’t pay so much attention to them.
I’m going to miss things that are “asked and answered.” Some of us can’t monitor these forums 24/7, like you do.
well there is this thread - see post 17 from one day ago.
Jaygrainger9 wrote: It has some good data and a good idea to be an objective ranking. But the formula does have some flaws, like it does not account for racing at altitude, and if a course is long or short but TFRRS says it isn't (like Roy Griak is actually 8,150m but is labeled as an 8k). So the idea is great but like anything in life it is far from perfect.
Nope. It accounts for ALL factors. Course length, altitude, weather, etc. It accounts for them by normalizing all races to the same scale. Those factors affect all the runners in the race so they get washed out in the normalization process which processes individual results across all races. A simplified example of this would be for a race at altitude where the average runner is 30 seconds slower than average then their rating for the race is adjusted by those 30 seconds. It does not matter whether that 30 seconds was due to altitude, course length, weather, etc.
I support this message. In a short course everyone runs faster. Short courses/altitude/hills courses are literally the reason I made the website.
In a LACCTIC simulation, Parker Valby has a 97% chance of winning. Doris Lemngole has a 1% chance, Katelyn Tuohy a 1% chance, and Hilda Olemomoi a 1% chance. 😆
Remind me of the projection from last year.
Lacctic is a representation of what we’ve seen so far. I’d say 97% sounds about right in terms of the performances this season. Of course, Tuohy could her best race this weekend, but that’s just speculation and not really a job for a computer.
That is the exact job of a computer — computers use tons of data and probabilities to make predictions all the time, weather being the most obvious example.
A prediction should account for the probability that things don’t line up with historical results. If Lacctic can’t do that, that’s fine, but then it should just provide the historical converted results and not give percentages.
Like last year? Eerily similar scenario. The algorithm implodes under certain circumstances. Call it the Tuohy uncertainty principle.
The algorithm seems to have "forgot" that Tuohy ran 15:03 6 months ago and is assessing her on current season results only, and since she, at least at ACC did not appear to run all out one might think her rating potential might be faster than what she has done. Valby seemed to run pretty much all out twice and that is what her rating is based on. Doris whose fastest rating is much slower than Tuohy's "forgotten" 15:03 is assessed twice (ie 2% chance) the chance of winning as Tuohy is. Not criticizing, one can only have the program do so much,but that seems to be what was included.
1. Results prior to this season. A good historical analysis could give you a reasonable estimate of how often excellence in prior years outweighs performance in pre-championship meets in the current season (hint: it's more than 2%).
2. Random error due to off days, great days, injuries, acts of God, food poisoning, boyfriend troubles, bed bugs, bad sleep, etc.
Like I always say, it is a tool. And Valby has a head to head win at Nuttycombe. But last year Lacctic also assessed that Chelangat had a better chance than Tuohy even after Tuohy had defeated her decisively at the Joe Piane based on the clearly short SEC course.
This post was edited 33 seconds after it was posted.
95% confidence intervals are at least 36 seconds long - that seems about right to me for runners who have backed up a rating twice (if there is more variation in their two ratings then they have a larger confidence interval). The reality is that Valbys performances are just THAT much better than Tuohy’s so far.
Do you believe it would be right to include Tuohy’s track performance for XC right now? I get it for earlier, but for now, after a whole season? I’m genuinely asking this because I’ve gone back and forth. Remember, this algorithm isn’t designed to just rank the Tuohy/Valby rivalry - it’s supposed to also do well for the 88k other runners it rates…
You need to figure out how to weight them empirically, but yes, of course you have to consider history, including track, xc last year, and as much before that as possible. A few pre-championship xc races are never going to give a great estimate in a sport where the best runners can often coast through most of the season right up until now.
I haven't seen the output -- or the algorithm, but I don't understand how you ended up with 36-second confidence bands and yet a nearly 100% certainty of victory over a woman and just beat by a few seconds.
H'm. Last year Tuohy blew it all up despite similar predictions. For example the best performance last year pre-Nationals was Tuohy's ACC performance, yet the algorithm did not show it. Likewise I think Tuohy's regional performance was better than Valby's this year. Would be quite amusing to see a repeat. Nuttycombe makes it really interesting.
Ben and Garrett are back on the mics with a mega episode! This week, the guys go through last week's regional results on both the men's and women's including...
95% confidence intervals are at least 36 seconds long - that seems about right to me for runners who have backed up a rating twice (if there is more variation in their two ratings then they have a larger confidence interval). The reality is that Valbys performances are just THAT much better than Tuohy’s so far.
Do you believe it would be right to include Tuohy’s track performance for XC right now? I get it for earlier, but for now, after a whole season? I’m genuinely asking this because I’ve gone back and forth. Remember, this algorithm isn’t designed to just rank the Tuohy/Valby rivalry - it’s supposed to also do well for the 88k other runners it rates…
95% confidence interval is 36 seconds long, Valby beat Touhy earlier this season by 12 seconds and the Alabama girls by 10 seconds, yet the algorithm puts Valby at 97%? It really doesn’t make any sense. It doesn’t account for variability potential, which you need in a sport like XC where who even knows what sort of effort the runners put in before champs and getting a cold could easily alter the results.
To do percentages, the algorithm would need to go back over a long time and analyze historical in-season performances vs championship performances and incorporate that variability into the percentage.
95% confidence intervals are at least 36 seconds long - that seems about right to me for runners who have backed up a rating twice (if there is more variation in their two ratings then they have a larger confidence interval). The reality is that Valbys performances are just THAT much better than Tuohy’s so far.
Do you believe it would be right to include Tuohy’s track performance for XC right now? I get it for earlier, but for now, after a whole season? I’m genuinely asking this because I’ve gone back and forth. Remember, this algorithm isn’t designed to just rank the Tuohy/Valby rivalry - it’s supposed to also do well for the 88k other runners it rates…
I think this is the challenge - how to assess the very top who do not run to their full potential often other than in championship. Sometimes maybe in a loss. I'll go back to Kelati vs Monson in 2019. But for the other probably 99% the algorithim is probably good.
95% confidence intervals are at least 36 seconds long - that seems about right to me for runners who have backed up a rating twice (if there is more variation in their two ratings then they have a larger confidence interval). The reality is that Valbys performances are just THAT much better than Tuohy’s so far.
Do you believe it would be right to include Tuohy’s track performance for XC right now? I get it for earlier, but for now, after a whole season? I’m genuinely asking this because I’ve gone back and forth. Remember, this algorithm isn’t designed to just rank the Tuohy/Valby rivalry - it’s supposed to also do well for the 88k other runners it rates…
I think this is the challenge - how to assess the very top who do not run to their full potential often other than in championship. Sometimes maybe in a loss. I'll go back to Kelati vs Monson in 2019. But for the other probably 99% the algorithim is probably good.
Amirys Tynissma (don't think I spelled that right) recently made an IG post that makes me think she is ready to throw down. If this is true, and Sam Bush rises to the occasion as she has done at nationals that past 2 years, things look really good for NC State. Especially with a rested Chmiel and Stevens. I think they are ready to bring it. So exciting. I think it is going to be a great race.
On a different note - does anyone else think it is weird that Parker Valby is the face of the new Nike NXN spikes. I had seen several advertisements with her holding them up and saying how much she loves them. I get that she has a deal with Nike, but it seems bizarre that she is the face of those spikes when Katelyn Tuohy is the queen of NXN. Did Parker Valby even ever qualify for NXN? Seems audacious. It makes me hope that Tuohy beats Valby on Saturday and then goes on to turn pro with someone other than Nike. It just struck me as such a weird choice to have Tuohy's rival be the face of NXN spikes. Just my opinion.
I think this is the challenge - how to assess the very top who do not run to their full potential often other than in championship. Sometimes maybe in a loss. I'll go back to Kelati vs Monson in 2019. But for the other probably 99% the algorithim is probably good.
Amirys Tynissma (don't think I spelled that right) recently made an IG post that makes me think she is ready to throw down. If this is true, and Sam Bush rises to the occasion as she has done at nationals that past 2 years, things look really good for NC State. Especially with a rested Chmiel and Stevens. I think they are ready to bring it. So exciting. I think it is going to be a great race.
On a different note - does anyone else think it is weird that Parker Valby is the face of the new Nike NXN spikes. I had seen several advertisements with her holding them up and saying how much she loves them. I get that she has a deal with Nike, but it seems bizarre that she is the face of those spikes when Katelyn Tuohy is the queen of NXN. Did Parker Valby even ever qualify for NXN? Seems audacious. It makes me hope that Tuohy beats Valby on Saturday and then goes on to turn pro with someone other than Nike. It just struck me as such a weird choice to have Tuohy's rival be the face of NXN spikes. Just my opinion.
Parker is being paid by Nike, Tuohy is not. Seems like the right choice made by Nike for their spikes. If Tuohy had NIL with Nike, of course even a better choice given her past success there.
On a different note - does anyone else think it is weird that Parker Valby is the face of the new Nike NXN spikes. I had seen several advertisements with her holding them up and saying how much she loves them. I get that she has a deal with Nike, but it seems bizarre that she is the face of those spikes when Katelyn Tuohy is the queen of NXN. Did Parker Valby even ever qualify for NXN? Seems audacious. It makes me hope that Tuohy beats Valby on Saturday and then goes on to turn pro with someone other than Nike. It just struck me as such a weird choice to have Tuohy's rival be the face of NXN spikes. Just my opinion.
Marketing, especially with females, is more about appeal than results. This has been demonstrated over and over again. It's surprising that you don't get this.
Help us build the best running shoe review site for a chance to win a LetsRun t-shirt.Help us build the best running shoe review site for a chance to win one of 10 LetsRun t-shirts.