JK is a genius Part III: Here is his mile by mile analysis of how much the hills will impact the 2020 US Olympic Marathon Trials

rojo wrote:

With over 700 Trials participants, I thought about making the following article the first pay per view article in LRC history, but decided to yet again give it away for free. When I ran the uphill/downhill, Las Vegas Marathon in 2000 trying to make the Trials, his hill analysis was spot on.

https://www.letsrun.com/news/2020/02/mile-by-mile-analysis-of-how-the-hills-will-impact-the-2020-us-olympic-marathon-trials/

John Kellogg wrote:

Please understand that the mile-per-mile elevation changes alone, not counting the cumulative effects of hills in 2+ hours of running (including all the quad busting from 1,382 feet of downhill running) shapes up like this at 2:10 to 2:12 pace on a flat course, ceteris paribus:

Mile 1  + 6.1 secs.

Mile 2  - 5.8 secs.

Mile 3  - 8.8 secs.

Mile 4  + 9.0 secs.

Mile 5  + 12.4 secs.

Mile 6  - 8.5 secs.

Mile 7  + 7.1 secs.

Mile 8  + 15.9 secs.

Mile 9  + 6.1 secs.

Mile 10  - 5.8 secs.

Mile 11  - 8.8 secs.

Mile 12  + 9.0 secs.

Mile 13  + 12.4 secs.

Mile 14  - 8.5 secs.

Mile 15  + 7.1 secs.

Mile 16  + 15.9 secs.

Mile 17  + 6.1 secs.

Mile 18  - 5.8 secs.

Mile 19  - 8.8 secs.

Mile 20  + 9.0 secs.

Mile 21  + 12.4 secs.

Mile 22  - 8.5 secs.

Mile 23  + 7.1 secs.

Mile 24  + 11.2 secs.

Mile 25  + 5.3 secs.

Mile 26  + 6.3 secs.

To Finish  - 4.0 secs.

The total slowdown over the course using these figures alone would be 1:24.6 for the top men. For the elite females running @ 11% slower, the slowdown would be 1:33.1. Remember that this doesn't account for a cumulative slowdown from the constant ups and downs. This is nearly impossible to predict accurately, partly because lighter runners may do better on hilly courses, particularly if there is a lot of downhill. But it's probable that all that quad busting could slow the times down by an additional minute (give or take) more than those 1:24.6 / 1:33.1 figures. Boston is normally slower for most people than flat courses are (unless there's a big tailwind at Boston), and it has a significant net downhill.

There also appear to be 35 (if I counted right - whatever, it's a lot) significant turns. This will disrupt the rhythm and create enough additional cumulative fatigue to slow things down even more. As a ballpark figure, I'm guessing this course could be 3-ish minutes slower for the top runners having good days than an ideal course with fewer than 5 hard turns and less than +/- 10 feet per mile elevation change (this course has +/- 53 feet per mile elevation difference). Assuming good weather and people willing to set an honest pace, anything under 2:12 for the dudes and under 2:25 for the chicks would be impressive on this puppy given the PRs of the entrants.

rojo wrote:

With over 700 Trials participants, I thought about making the following article the first pay per view article in LRC history, but decided to yet again give it away for free. When I ran the uphill/downhill, Las Vegas Marathon in 2000 trying to make the Trials, his hill analysis was spot on.

https://www.letsrun.com/news/2020/02/mile-by-mile-analysis-of-how-the-hills-will-impact-the-2020-us-olympic-marathon-trials/

John Kellogg wrote:

Please understand that the mile-per-mile elevation changes alone, not counting the cumulative effects of hills in 2+ hours of running (including all the quad busting from 1,382 feet of downhill running) shapes up like this at 2:10 to 2:12 pace on a flat course, ceteris paribus:

Mile 1  + 6.1 secs.

Mile 2  - 5.8 secs.

Mile 3  - 8.8 secs.

Mile 4  + 9.0 secs.

Mile 5  + 12.4 secs.

Mile 6  - 8.5 secs.

Mile 7  + 7.1 secs.

Mile 8  + 15.9 secs.

Mile 9  + 6.1 secs.

Mile 10  - 5.8 secs.

Mile 11  - 8.8 secs.

Mile 12  + 9.0 secs.

Mile 13  + 12.4 secs.

Mile 14  - 8.5 secs.

Mile 15  + 7.1 secs.

Mile 16  + 15.9 secs.

Mile 17  + 6.1 secs.

Mile 18  - 5.8 secs.

Mile 19  - 8.8 secs.

Mile 20  + 9.0 secs.

Mile 21  + 12.4 secs.

Mile 22  - 8.5 secs.

Mile 23  + 7.1 secs.

Mile 24  + 11.2 secs.

Mile 25  + 5.3 secs.

Mile 26  + 6.3 secs.

To Finish  - 4.0 secs.

The total slowdown over the course using these figures alone would be 1:24.6 for the top men. For the elite females running @ 11% slower, the slowdown would be 1:33.1. Remember that this doesn't account for a cumulative slowdown from the constant ups and downs. This is nearly impossible to predict accurately, partly because lighter runners may do better on hilly courses, particularly if there is a lot of downhill. But it's probable that all that quad busting could slow the times down by an additional minute (give or take) more than those 1:24.6 / 1:33.1 figures. Boston is normally slower for most people than flat courses are (unless there's a big tailwind at Boston), and it has a significant net downhill.

There also appear to be 35 (if I counted right - whatever, it's a lot) significant turns. This will disrupt the rhythm and create enough additional cumulative fatigue to slow things down even more. As a ballpark figure, I'm guessing this course could be 3-ish minutes slower for the top runners having good days than an ideal course with fewer than 5 hard turns and less than +/- 10 feet per mile elevation change (this course has +/- 53 feet per mile elevation difference). Assuming good weather and people willing to set an honest pace, anything under 2:12 for the dudes and under 2:25 for the chicks would be impressive on this puppy given the PRs of the entrants.

not even a hobby jogger wrote:

I love this analysis...I plotted this out for my wife using her training paces before Boston last year (another course with lots of rolling hills...albeit it more downhill than up). She is a super-responder to downhill running (most runners are not). I ended up with 2.9 seconds lost (not far from the 2.4 used in this article) per 10 feet of elevation gain but 3.1 seconds gained per 10 feet of elevation loss. I used these numbers to create a band for the Boston course and she pretty much nailed it without even looking at the band or watch. I have witnessed MANY runners....both average and good...that simply can not run down hills to their advantage. So the analysis here of 2.4 lost and only 1.8 gained does not surprise me at all. It will be interesting to see if there are runners in the field that can take advantage of the downhills to milk small amounts of time off their competitors.

Sarcasm font wrote:

not even a hobby jogger wrote:

I love this analysis...I plotted this out for my wife using her training paces before Boston last year (another course with lots of rolling hills...albeit it more downhill than up). She is a super-responder to downhill running (most runners are not). I ended up with 2.9 seconds lost (not far from the 2.4 used in this article) per 10 feet of elevation gain but 3.1 seconds gained per 10 feet of elevation loss. I used these numbers to create a band for the Boston course and she pretty much nailed it without even looking at the band or watch. I have witnessed MANY runners....both average and good...that simply can not run down hills to their advantage. So the analysis here of 2.4 lost and only 1.8 gained does not surprise me at all. It will be interesting to see if there are runners in the field that can take advantage of the downhills to milk small amounts of time off their competitors.

If your wife loses 2.9 seconds per 10 feet of uphill, but gains 3.1 seconds per 10 feet of downhill, she should just find an absolutely crazy hilly course and set massive PRs.

She should run a full 26 minutes faster on the Pike's Peak marathon than she would on a flat course, with its 7800 feet each of uphill and downhill.

If we can find a faster marathon, she could do even better! Your wife is probably the only person who gets faster as the course gets tougher! If we find a hilly enough course, she could set the world record!

To drop an hour off of her flat course marathon using your calculations, she would need a course with 18,000 feet of uphill and downhill. That is an average slop of 25%. That sounds really high, but there are some backroads with slopes that high, it isnt just limited to paths. So just find a nice mile long stretch of road with an average of 25% grade and create a course that goes up and down it 13 times. If your wife's PR is under 3 hours, she could be the first woman to run under 2 hours! Suck it Kipchoge!