Pikes Peak Ascent and Marathon
Runners Forum => Race Info => Topic started by: John Garner on July 20, 2015, 02:07:49 pm

.... but were afraid to ask.
If you are interested in how we assigned bib numbers (and thus seeded folks into starting waves) and how this impacts your assigned starting time, read on. Otherwise, just keep training...
Part 1: Bib Number Assignments (in 5 easy steps)
Step 1: Analyze last year's results
We took all of last years results and broke it into 4 major categories based on the type of qualifier: 1/2 marathon, full marathon, Garden of the Gods, and PPA/PPM Vet. For the first three, we created simple qualifier vs ascent time (or ascent portion of the PPM) tables and used Excel to determine 3 best fit regression equations. They worked out to be as follows:
1/2 Marathon: 31.485*X^2+7.1018*X0.1666
Full Marathon: 6.9456*x^2 + 3.1414*X  0.136
Garden of the Gods: 10.728*X^2+4.7255*X 0.0666
For the veterans, had some extra data that we were able to use, naimly their historical results. We looked at their 2011, 2013, and 2014 Ascent Times (as available) vs 2014 Ascent time. With this we were able to generate 3 additional equations:
3 years back: 2.7082*X^2 + 1.8544*X  0.0585
2 years back: 1.2532*X^2 + 1.3813*X  0.0225
1 year back: 1.1367*X^2 + 1.3286*X  0.0198
In all cases, we determined an estimated ascent time for a given qualifier based on last year's results.
Step 2: Assigning "estimated times" for all of this years runners.
For folks who qualified with a 1/2, Full, or Garden time, this is easy. Take the qualifying time and run it through the equation derived in step 1. For vets, we looked at their 2012, 2013, and 2014 times and ran them through the appropriate 1, 2, or 3 year back equation derived in step 1. We then used their _fastest_ result to determine their 2015 estimated time.
Step 3: Assign all of the award winners, former winners, competitive entries, etc.
Being in this category has some perks. In this case, so long as these runners had an estimated time of 4 hours or less, they are assigned a bib number as outlined on the bib order lists. Those with an estimated time of over 4 hours were put back into the general entry.
Step 4: Assign the top 50 men and top 50 women to the first wave.
Assign bib numbers 101 > 149 to the top 50 females based on their estimated ascent time. Do the same for bib #150 > 197 for the top 50 males by estimated ascent time. The first starting wave is a bit larger than the rest, consisting of bib # 1 > 199. Bib 1 > 99 are the comp entries as outlined in step 3. bib 100 > 199 are for the top 50 men and women who entered via the general entry process. We did not want the first person (or even the 10th) crossing the finish line to not be the fastest person on the day,
Step 5: Assign the remaining runners based on their estimated ascent time.
We simply sorted the remaining folks and assigned them bib numbers starting at 201 until we had assigned everybody a bib number. We skipped the 00's to avoid confusion and the 98's and 99's to leave some room for oopses (like lost bibs).
Part 2: How this translates to starting times:
To determine your starting time, simply look at your assigned bib number. If it is between 1 and 199, you will start at 7:00AM. For everybody else, divide your bib number by 100. Take the quotient and add that to 7am. For example, if you are bib number 495, you will start at 7:04. If you are bib number 1659, you will start at 7:16.
The big question: What if you want to run with somebody else who is in a different wave?
Runner's may start in any wave _after_ their assigned starting wave. Thus, if you want to run alongside somebody else who was assigned a later starting wave, you are permitted to start with them. Your time will being when you cross the start line. If you are starting late, it is best to do so at the back of the pack, and slowly cross the starting line to make sure that the system sees your chip.
Finally: Is this system perfect?
Absolutely not, but we hope it will be better than the old system were 1000 runners were set loose at the same time.

Is this for both races or for the Ascent only?

I assume X is in fractions of a day since seconds, minutes, and hours run thru the formula all come out negative and days come out close to the values of X.

Is this for both races or for the Ascent only?
This is for both races.
I assume X is in fractions of a day since seconds, minutes, and hours run thru the formula all come out negative and days come out close to the values of X.
Yes. I used excel which represents 24 hours as 1. So 1 hour is 0.041666.
john

Thanks John!
This is helpful.
And it will be fascinating to see how it actually works out with regard to congestion. Getting excited!!

How does the staggered start time effect the cutoffs? Does the runner who starts at 7:06 have till 5:07 to complete the race?

I would think that the race time cutoffs should be staggered accordingly. The one cutoff that I thought might affect racers is a weather (ie lightning) cutoff. If you are starting in a wave 15 to 20 minutes past 7:00am it could be a factor. With all the crazy rain storms we've had this summer let's hope this does not play in. Overall, I do like the new system.

Not sure if it is an issue, but I entered my predicted finish time incorrectly. I put my flatland half marathon time. As we all know, I will not be even close to that time! My marathon time is 4:18 so I added 30 minutes to get my predicted finish time and then some. So my predicted time should be 5:00.
Thanks for all your help and all the great information! Getting excited!

Not sure if it is an issue, but I entered my predicted finish time incorrectly. I put my flatland half marathon time. As we all know, I will not be even close to that time! My marathon time is 4:18 so I added 30 minutes to get my predicted finish time and then some. So my predicted time should be 5:00.
The predicted time is just for fun and does not impact the wave seeding. If you want to update it, email me at registrar@pikespeakmarathon.com (include your real name), and I'll get it updated. Historically, a 4:18 averages out to about a 4:53 PPA finishing time. This is close to the general rule of "full + 30 minutes."
How does the staggered start time effect the cutoffs? Does the runner who starts at 7:06 have till 5:07 to complete the race?
Cutoff times will be staggered as well.

I understand that the cutoff times will be staggered. How will that affect someone who runs in a later group as far as the cutoffs along the way  i.e. how will a worker at let's say Bar Camp know that someone started in a later group than their bib number shows? Will all runners be allowed to continue past those cutoffs until the latest cutoff time or will there be some way of identifying exactly what time a person crossed the starting line at every cutoff location (assuming that the ultimate cutoff at the finish line is met)?
Thanks for your help!

John,
I would appreciate a detailing of the new cutoff plan, please.
Is it oneforone minute based on your miniwave start time?
Thanks,
Tim

I think John explained it above. Cutoff times will be staggered accordingly.
Fortunately, I don't fall in this circumstance, but, I bet the crews would still be out there supporting runners 5, 10, maybe 15 minutes late; but after that you can expect that they have broken down their setup and are returning to town.
They have to set limits somewhere...

Is it oneforone minute based on your miniwave start time?
If that were possible, yes. But the reality of the situation is that it going to be handled similar to the past: The folks on the trail doing the enforcement will provide a few minutes wiggle room in case somebody started a bit late and let them through. To make it a bit more interesting, we cannot force anybody to stop. All they can do is strongly suggest that they turn around, but the SAR folks are usually very good at persuading folks. :)
But at the finish line the cutoff is based on your chip time. Period. If you get to the top and your final time is over 6:30, you will be OCO and not receive a finishing time.

Is it oneforone minute based on your miniwave start time?
If that were possible, yes. But the reality of the situation is that it going to be handled similar to the past: The folks on the trail doing the enforcement will provide a few minutes wiggle room in case somebody started a bit late and let them through. To make it a bit more interesting, we cannot force anybody to stop. All they can do is strongly suggest that they turn around, but the SAR folks are usually very good at persuading folks. :)
But at the finish line the cutoff is based on your chip time. Period. If you get to the top and your final time is over 6:30, you will be OCO and not receive a finishing time.
Thanks, John. That's good that finish time is based on chip time. Still, I expect this will make for some interesting math at the finish line for those handing out medals after 6:30 has gone off the main clock because there will be a lot of runners come in with compliant chip times after that  based on the new setup.
I'm a little confused as to how you say it was handled in the past. I've seen many get pulled from the course as I rested for a few minutes just beyond AFrame . Don't know what they told them but I've never seen anyone continue after that conversation even though they were adamantly protesting. I can assure you that, as is, the vast majority of the last 2 miniwaves will be over 3 hours (clock not chip) at Barr and over 4:15 at AFrame simply by the structure of their historic pace and starting them over 20 min later than normal (vs the clock). I hope those pulling at Barr and AFrame are well aware of this because there will potentially be hundreds of people with compliant chip times easily capable of finishing under a 6:30 chip time who will could be 15 min over the raw clock cutoffs.

I hope those pulling at Barr and AFrame are well aware of this because there will potentially be hundreds of people with compliant chip times easily capable of finishing under a 6:30 chip time who will could be 15 min over the raw clock cutoffs.
The "clock" for those runners in the last wave will be starting at 7:22, not 7:00. We are not that cruel. :)

I hope those pulling at Barr and AFrame are well aware of this because there will potentially be hundreds of people with compliant chip times easily capable of finishing under a 6:30 chip time who will could be 15 min over the raw clock cutoffs.
The "clock" for those runners in the last wave will be starting at 7:22, not 7:00. We are not that cruel. :)
Great. Thanks, John.

AFrame folks will know your start time based on your bib number. It's pretty simple math....

If you look at last years results, there were a few runners that did not make the 4:15 cutoff at AF, but managed to finish below the 6:30 cutoff at the top.

If you look at last years results, there were a few runners that did not make the 4:15 cutoff at AF, but managed to finish below the 6:30 cutoff at the top.
One thing to keep in mind is that the cutoff folks are at the aid station, the timing split is two (or so) switchbacks above at the AFrame Sign. A number of folks will get past the aid station and then take a break before continuing upward.
This is why I waited until _after_ the AFrame aid station to stop and put on my rain pants back in '08. I figured that with the weather as bad as it was they could call the race at any time and wanted to get above the last cutoff. The rough math indicates that I made it past the aid station by a few minutes. :)

Last year as I was getting my water bottle filled at AFrame, a runner next to me commented, "I'm really glad to make the cutoff at AFrame." Another volunteer said, this is not the cutoff, the cutoff is at the timing mats, you still have a ways to go. Then another volunteer said quite loudly so everyone could hear, this aid station is not the cutoff, the cutoff is at the timing mats.

That was my experience as well. In fact in 2013 the aid station folks at the lower level at Aframe passed me thru but I didn't make it to the mats in time. I didn't know they weren't at the same place. Also I am a little confused now about the wave starts and cutoff times. The info packet states "hard" clock times not x hours from start based on chip time. Basically based on that I will have 20+ minutes less time to make the cutoffs than those starting well in advance of me. Can you clarify?

The timing mats were first introduced at AFrame in 2012. Before that the Aid station at AFrame was the cutoff. This is may be the cause of some confusion.

Here is the thing: the folks enforcing the cutoff have no way of knowing exactly when you started. You could have bib # 201 and start at 7:22 with the last wave to run with a friend. With a graduated system, one would get stopped and the other allowed to continue on.
So for this year, to keep it as simple as possible, all of the the oncourse cutoffs are based on the old 7:30 start times. Oddly enough, this gives everybody an 8+ minute advantage from a cutoff perspective when compared to runners last year. We basically relaxed the cutoff times.
Ultimately, if you are near the cutoffs, you are in serious trouble of missing the one that is always enforced evenly for everybody: The finish line. At the end of the day, the timing crew knows when you started. If your final chip time is over 6:30:00, you will get a DNFOCO no matter what.

Absolutely the best solution. Keep it simple for participants and volunteers.

Hi Johnthe mathematician in me is curious how you came up with the prediction equations. I'm considering incorporating them into my course on math modeling (a nonmajors 1styear college course titled "Real World Problems & Math Modeling Solutions."). Absolutely no rush on a replyyou've got plenty on your plate over the next few weeks.
Thanks for keeping this board running smoothly.

I don't want to muddy things more  but the "cutoff" times at Barr and AFrame are NOT, as I understand it, based on a runner's start time, but a raw clock time that starts at 7:30. Since the last wave (which I'm in) starts around 7:217:22, there's still almost an 8 minute buffer for those cutoffs. I'm a big fan of the staggered start. Being at the back of the pack, and more of a hiker than a runner, in prior years I spent a significant amount of time behind folks who started out fast and slowed down and even took breaks on the "w's".
Only the finish would be figured on a runners start time using chip time. Last year, a few runners made it to finish with a chip time over 6:30, got a medal, but did get a DQ once their chip time showed they didn't make the 6:30 cutoff.

Hi Johnthe mathematician in me is curious how you came up with the prediction equations. I'm considering incorporating them into my course on math modeling (a nonmajors 1styear college course titled "Real World Problems & Math Modeling Solutions."). Absolutely no rush on a replyyou've got plenty on your plate over the next few weeks.
Thanks for keeping this board running smoothly.
The process was relatively easy and simple in theory but ended up being a bit time consuming. I had the registration database with all of the data going back several years. I also had the results database.
With that in hand, I was able to take the 2014 results and build up a <qualification time> > actual table. The folks who used a full marathon or half marathon as their qualifier were rather straightforward. The recent veterans were a bit harder, as I had to lookup their times from 3, 2, and 1 year prior (and toss the slower of the two in any given year for doublers) and produce 3 tables. For the garden of the gods qualifier, I used the TCR series result to get a decent number of datapoints vs just using the folks who used the Garden as their qualifier.
Once I had the qual > actual times for each qualifier, I just pulled it all into excel and plotted a simple x/y chart for each qualifier. Then I had it compute a regression line (2nd degree polynomial usually looked best). It is not a perfect solution, as the raw data often looked like a shotgun blast. As a result, the R^2 values in the 0.6 (half marathon times) to 0.7 (peak vets) range.
Vola, runner qualification time ranking. (Just don't let a real statistician find out what I did, they would probably cringe. :) )