Random or Ordered Selection

Go to the store and you stand in line to checkout. Go to the bank or post office and you take a number and wait for your number to be called. Apply for a passport or to adopt a child and you wait your turn. All of these are first-come first-served systems, and dare I say, what we all would consider fair. If one of these were to use a random selection process where names were just randomly pulled out of a hat (or an e-hat) it probably wouldn’t go over very well. You go to the store and don’t know when or if your name would get called to checkout. You might not make it home in time for dinner. There would likely be many complaints and cries of “this isn’t fair, I got here before she did.”

When it comes to games of chance such as Powerball, selection of jury members, or conscription, a random selection process is implemented. These lottery solutions don’t care how long you have been trying to win, each ticket or name has equal chance, same as every other ticket or name. You could have played and lost the lottery every week of your adult life and it won’t change your chances in the next one. We generally accept that a random solution is fair in these domains, regardless of how many times somebody has won or lost previously.

When it comes to registering for ultras it is interesting what we perceive as fair. As a race becomes more and more popular and fills up in weeks, then days, then hours, and finally minutes, people start asking “when are you going to a lottery?” Somewhere along the way we have come to believe that a lottery, or random process for selection is more “fair,” more egalitarian. But because it really isn’t a fair way to choose applicants, especially as the number of applicants greatly exceeds the available spots, races have to go to great lengths to try to make it fair, to make it less random. We end up giving additional tickets to multi-year losers, people who have volunteered at the race, runners with previous finishes, or whatever type of applicant “deserves” to have their chances increased. Seems pretty obvious to me that because we have to go through these machinations to make the selection process less random, a random lottery process is not inheritantly fair.

I don’t know how we got to this point with ultra race selections. Or why there is this perception that lotteries are more fair than first-come first-serve solutions. Why not have a big multi-year queue? Runners can register at anytime of the year. When an RD wants to choose the field for the next race they simply go down the big wait list and let people in as room allows (and assuming they are still qualified when their turn comes). When a runner is chosen and runs the race they can choose to be added to the back of the queue and wait their turn for another try. Obviously, there might be alternate routes of entry for elite runners or whoever the race deems should be let into the race, but this would be for the general population.

So help me out. Do you think that lotteries are more fair than a first-come first-serve solution to ultra entrant selection, and why?

78 Comments

  1. Craig,
    I prefer no lottery when the races don’t fill up quickly. Everyone has time to sign up. But when it gets to the point where a race fills up in minutes, I prefer lottery. Your non lottery examples above aren’t the most applicable- imagine, for example, if there were only 200 passports available a year or 100 gallons of milk at the store. It would be chaos. I’ve been shut out of races that filled up in 3 minutes because of website malfunction (the website was freezing for some people, not others) or because I had to be in a work meeting in that 3 minute period. I prefer lottery to a few minutes of web page refreshing and stress.

    I’m just glad most of the races I do are not lottery races, and I continue to seek new, non lottery races as others fill up. Though, I am a 3 time WS loser…

    For races that are lottery, though, hopefully the entrants list can grow, either through new permits or reroutes or acts of Congress. We don’t want 10,000 runners or a Leadville 100 atmosphere, but allowing more runners can help, too.
    My two cents.

    • Hi Jon, doesn’t this occur every time Apple sells a new product? Or when Pliney the Elder comes to your grocery store? Black Friday at Walmart? First-come, first-served. Some get to buy the product others miss out. Some get rain-checks, or not.

      If you could let go of the issues with current online registration companies and imagine someone could implement a system that doesn’t crash, is it more fair to have an ordered or random selection to get into races? If 500 people try to get 400 spots, the first 400 get to run the race, the next 100 are first in the queue for next year. As a three time WS lottery loser are you ok with a first timer randomly getting a spot? What if you got to 10 years and you never got selected? Would you think it fair that a first year applicant gets a spot while you continue to get left out and have to search for non-lottery races?

      As for increasing number of starters in races, that is not easy. Assuming an RD can get a permit increase, as you pointed out there are all kinds of problems that are then created. Not good for the sport, IMHO.

  2. More starters is definitely not the answer. I would love to run WS. I’ve entered and not been picked twice, but I did get to go and pace my good friend in 2012.

    I don’t really think that there is a right or wrong answer here. The idea of a first come first served, with a queues for the next year is appealing. I keep getting older and worry that the year I finally get picked I’ll be some old geezer.

    Bottom line, there are a lot of great races out there (heck, I LOVE Leadville). Keep an open mind, try new things. Travel. See new places. Don’t live or die by one race’s lottery results.

    Keep up the good work Craig!

  3. You left out one factor in the formula for a race where its quota is defined by another agency. In the case of WS100 the U.S. Forest Service dictates the number of entrants and therefore constraining the spaces available to entrants. It’s just as important to include a wide variety of talent in a race. In my opinion, the experience should reached beyond the finishing time but the journey itself and how it impacts the human’s spirit regardless of their ability. Therefore, I say leave it to chance in a lottery.

      • The Dipsea race has those entrants that come in the previous year in a certain amount of time/placement and then there are those that pay their way into the race and those that get in by who they know. The snail mail process doesn’t work unless you are in Marin at the proper Post Office. The Dipsea process is a horrible example of what is fair. Maybe you could do an entry process to the highest bidders…make some real money on the race and let the rest be by lottery.

  4. Craig, totally agree with your argument when a race fills over the course of a short time period where an orderly process can result in an actual queue.

    The problem with Waldo yesterday was that it didn’t fill in 15 minutes, it actually filled in probably less than a minute. This is obvious to all 200+ of us that hit the register button at 7:00:01. The glitchy checkout may have taken 15 minutes, but the race was full immediately.

    I think if we are looking for a solution, rather than focusing on perceived fairness, we should focus on whether or not there is an orderly process that makes sense. Given the dramatic imbalance of supply and demand in this case, a first come first served free-for-all is never going to work. It is equivalent to 200 people showing up to the post office parking lot within a 10 second window.

    I personally think that a CIM-like queuing process would be a good model to contemplate for improvement in high demand ultras.

    Finally, can’t help but point out that the biggest driver of increased demand for Waldo yesterday is likely the changes to the Western States LOTTERY process. Kind of ironic.

    • Mike, thanks for your thoughtful comment. Not sure if we’ll ever know how many people logged in at 7:00:01 ready to sign-up for Waldo yesterday morning, but was it the technology that failed us or the first-come first-serve model? We used to register for Way Too Cool via active.com and the race would fill in single digit minutes. It was always an anxiety-ridden morning but so are public lottery drawings.

      I’ll have to look at the CIM model.

      As for WS driving demand for Waldo, not sure we’ll ever be able to put a number on how much the WS qualifying standards contributed to the increase in demand this year, but the correlation has not escaped me. However, there are races on the WS qualifying list that are sitting there with plenty of spots available including the other qualifier in Oregon, Pine to Palm.

  5. Yes, the only problem is those races that fill in minutes and then there is no way to get in an orderly line like at the Post Office. Whatever system there is, there should be a way for qualified entrants to have an opportunity to get in eventually if they keep re-applying (or at least have their odds increase each time).

    And they shouldn’t lose all their credit for past attempts if they miss one year- sometimes there is a family reunion or wedding, etc and it isn’t fair to tell someone with 5 years of attempts to choose between that and the race.

    On the other hand, we should be proud that we have this problem. In a normal free market system, the race directors would solve the problem by simply increasing the entry fee until supply and demand intersect, but that would price most people out!

    And yes, as the previous comment pointed out, the new WS system is unfortunately making it near impossible for many of us to get back into our favorite 100K races. I would like to see WS adopt a qualifying system where you simply have to finish in the top 80% of any 50M or longer race with at least 100 starters that isn’t in its first year.

    And thanks for asking. We understand how hard this is on the RD’s also!

  6. It is refreshing that you are addressing this issue. I don’t like lotteries – weighting helps but still doesn’t solve the problem. I like looking at models such as JFK, CIM and Dipsea. I like first come first served carrying over from year to year. Maybe entry should be open continuously, but only after one has qualified, ie, upon meeting a qualifying requirement, one could enter the queue.

  7. what would the qualification standards look like in the queue model? qualify evert calendar year to stay on the list? or qualify in the 18 months leading up to your scheduled year? i think first timers would probably have a slightly better finishing rate with a 3 year training plan and a very clear light at the end of the tunnel.
    also, I wonder what the tipping point would deb following Mark’s (tongue in cheek) suggestion about price increases. $1,000? $2,000? wouldn’t be very good for the race but all that race revenue could be put to use for the good of the sport.

  8. I am a bigger proponent of mandatory qualifiers than I am of lotteries. Are lotteries fair? Sure. Are they more fair than first-come first-served situations? I don’t think so. It’s hard to say what the solution is when the demand becomes so high, especially when there are variables in addition to the quality of the race that increase the demand (in this instance, the dearth of 100ks and the changes in qualifying standards for other races).

  9. I like the idea of having people “wait on line” online. If registration opens at say 7:00 a.m. on March 1st, then people are allowed to log in to the registration site at any point prior to signup and the first person to log in is assigned the number one spot in line. To make it effectively similar to standing in line, they must maintain an active status on the website in some capacity. I think this method helps those who want it most, get in, and those who are indifferent do not. It’s just like the people who camp out for the latest video game console.

    This method may have its limits too. You wouldnt want the signup process for Western States to be more grueling than the race itself.

    Just a thought. Thanks!

  10. There are now plenty of Ultra-running opportunities for everyone, just like there are plenty of marathons. Everyone who wants to run and Ultra or a Marathon, can certainly find one to compete in.

    But the fact is, there are some of both of these races that are more popular than others (WS 100; Boston Marathon; etc), and where ‘demand’ exceeds ‘supply’. If these were simply another consumer product, the tried and true ‘fair’ way to distribute the limited supply of entry’s would be via ‘price’. Entry fees could be raised or even auctioned off, and eventually supply and demand would achieve equilibrium.

    But these aren’t just another consumer product. Regardless of any individual entrant’s motivation, these are at the end of the day ‘races’. We all line up together at the start. A gun goes off. A clock is started. And at the end, the ‘results’ are listed in order of who finished 1st, 2nd, etc. These results are often supplemented with a distinction of how entrants finished relative to others of their gender and age.

    I suspect that everyone would agree that it would be absurd to report the results randomly. It would be equally absurd to order the race results based on who got their entry in first.

    With some exception, the highest demand races also happen to be the most competitive.

    It seems to me that the ‘fairest’ way to allocate a limited supply of ‘race entries’ is via race performance. The Western State’s 100 is not accessible to everyone. The race has long recognized the need for a qualifying standard. Why not take this a step further and use qualifying standards as a way to bring supply and demand into equilibrium?

    The model for this may very well be the Boston Marathon, with it’s age graded qualifying standards and entry system. The Boston marathon also supplements this system:

    1) Top Performers in Age division. The directors of the Boston Marathon will allow ‘elite’ and top age division entrants well after the entry deadline. In my view, this is a perfectly fair system for a ‘race’ in order to help ensure that the top runners have a chance to ‘race’ each other.

    2) Charity Entries. Boston (and many other races) also have a mechanism that allow a certain number of entrants to ‘buy’ their way into the race. This option isn’t available to everyone, but it has the benefit of providing additional resources to the race, and thereby offsetting the cost of entries for others.

    From a ‘fairness’ perspective, this system means that some people may never have a chance to run some races. In my view, that’s OK. I never had a chance to run in the High School State Meet. And I will probably never have a chance to run in the Olympic Marathon trials (let alone the Olympics).

  11. Craig I really like your continuous list idea. For those of us that like to plan it would be ideal beyong measure! Even if you end up talking about a 10+ year out plan. As for what is more fair? I have no idea. I do feel like the lottery is part of the process. I will never forget the moment when my Craig and I got into WS, UTMB and W2C. Those moments were so exciting and to me they were as much a part of the race as race day!

  12. Craig, you and all folks commenting, are pitching lottery versus first come as the two obvious options with plenty of every-day-life examples that are fully accepted as being reasonable and fair. I would like to point out another every-day-life example that our community has accepted as reasonable and fair. It’s called College Application. A review board looks at applicants’ resumes and decides based on test scores, extracurricular activities, etc. The more popular or renowned colleges get way more applicants than they can accept (Stanford accepts roughly 7%; sounds familiar ?). All high schoolers know this so most don’t even try, while those who do, have realistic expectations about their chances. In the end nobody complains, we all seem to accept that the better colleges get the better students………

  13. Craig – have you ever considered incorporating some sort of hunger games element into a race? I know that doesn’t directly correlate to the race selection procedure but it would be interesting to see how that type of race dynamic would impact that field. certainly speed still would be a compelling factor in the outcome, but who knows, incorporating some sort of battle elements might create a unique scenario? it would really get the mid to back packers to have more intensity when competing that is for certain. thanks for any input

    • Viper, the first year of Waldo 100K we inadvertently scheduled the race on the same day as opening day of deer hunting season. I wanted to give a prize to the first person to finish the race wearing antlers but my wife convinced me that was a bad idea.

  14. Two points:

    1.) All “fairness” to the “first-come” scenario goes out the window when the technology fails. The issue with Ultrasignup for many yesterday was that, once they got to the front of the line and accessed the “ticket window”, they still had to walk through the 5+(?) pages it takes from login to payment to confirmation. Along the way, several were bumped…as if they were at the ticket window and a mystery trap-door opened, shooting them out in a sewer. Presumably, they were bumped because, as they were trying to process their way through the website, the others “pushing shoving in the queue” (with Nintendo-like refresh clicking) were overloading the site, effectively pushing them out of the ticket window and out of the system.

    If runners are paying services like Ultrasignup 5-10% of the race entry for the convenience of online registration, such “trap doors” should not occur. Nor should mistakes such as inadvertantly selling 50 extra entries. Bad business, indeed.

    As hated as it is, Ticketmaster has the technology to handle such demands: once you’re inside, your ticket is protected for a period of at least five minutes — somehow shielded from the overwhelming page loads and refreshes when tickets go on sale.

    RDs – as well as runners – should demand this level of service for what runners are paying (e.g. >$35 for a race like Western) sites such as Ultrasignup.

    Idea: under the current Ultrasignup technology, I’m wondering if one could simply enter their email address into the race page, then be emailed a protected hyperlink for them to register. Presumably, this would “preserve” their race entry for a finite period of time (30 minutes? 60 minutes?) and allow for any untimely system crashes (on either end) without losing their spot. I believe ultrasignup already has this capability at hand (as I’ve received several dedicated entries over the years).

    Kind of like buying your ticket for the movie, then walking 20 feet to the ticket-ripper, who actually lets you into the movie.

  15. 2.) the lotto versus first-come…

    In the early 2000s, I was obsessed with U2. In their 2001 “Elevation Tour”, they had an exclusive “inner circle” to their floor tickets. The first 250 folks that arrived got entry to this area, which was encircled by a walkway where the band members would frequently walk onto.

    For three shows, I *slept overnight* outside the venue in order to be as close as possible to the stage. It was phenomenal. And, being in my mid-20s and without any obligation, this was only a minor sacrifice.

    The “line”, however, was always stressful: it had to be heavily policed on concert day: a “notebook” was kept where people had to sign in, then get “their number” written in Sharpie on their hands, to insure against those who would cut in line. It didn’t always work, but it was mostly effective. Nevertheless, it was stressful until the moment the doors open and you dead-sprinted from the ticket turnstile to the stage.

    It was also time- and energy-intensive: it was a day-long commitment, and you had to fiercely protect “your spot”! Often, by the time the concert started, you were exhausted from the damn line!

    Then, for their “Vertigo Tour” in 2005, they changed the process: the inner circle would be “lottery”. Still being young and exuberant, this was frustrating: I still got in line early — believing that I still could get a front spot, IF I “won the lottery”.

    3 of the 4 shows I was a “loser” – no lotto win, no inner circle. I stood there, seething, as random folks who showed up 3 minutes before showtime scanned their tickets and found out they gained access. Some of them had no idea what “The Circle” meant! The nerve!

    I thought, “What a waste! They don’t even care, they just showed up at the last minute and got in!”

    *****

    Nine years later, facing the same conundrum with races, I feel the same uneasiness and trouble: what is best?

    – Clearly a lottery is less energy-intensive and less stressful than a first-come/first-serve, but where is the reward for the effort, dedication and commitment to “the line”?

    BUT…

    – Should the events only be letting in those with the time and energy for such commitment? What about the family man/woman who cannot devote such energy to the entry process? What is the event missing by not including these folks?

    *****

    That said, perhaps a hybrid “Early Lottery, Late First-Come” system might be effective:

    – Open a %age of spots to an early lottery. Give everyone an opportunity to enter the lottery and not have to sweat or stress an ultrasignup bum-rush.

    Allow no waiting list, and non-refundable entries, charged immediately.

    Then:

    – 1-2 months prior to race-day, put all DNS’d entries, plus the remaining %age, into a first-come/first-serve entry. Reward those runners who A.) are capable of queueing and B.) truly are willing/able to run a chance to race.

    Lottery rejects would have a second change (do you STILL want to race, now with 4-8 weeks prior?) and theh dedicated, tech-savvy race junkies can get their bum-rush on.

    Such a hybrid system would allow a better variety of runner, as well as ensure a fraction of runners who are truly prepared to race, by virtue of their “last-minute” entry system.

  16. Isn’t the new exponentially weighted lottery kind of a nice hybrid of the pure lottery and the multi-year queue? I know there are no guarantees for multi-year lottery losers, but surely this is better than a 10+ year waiting list. Personally I like know there’s at least a chance to get in each year. Knowing I’m 5.. 6.. 10 years out… man, that’s a tough pill to swallow. ‘course, maybe that way you only get people that are in the sport for the long haul. And I think there’s value in that.

  17. Craig, I agree that a multi-year wait-list would be much more fair than a lottery, as long as there’s a way to qualify too (e.g. Montrail Ultracup, prior year top 10, etc). That way someone wanting to run WS has a very good idea when they get in rather than entering several lotteries simultaneously and dealing with the consequences. For example, getting into WS and Hardrock in both lotteries is unlikely for people with 1 ticket but it could spoil their experience because they couldn’t really turn down either, especially as they have to pay the entry fee either way on Ultrasignup.

    A list would mean someone would be guaranteed to get in at a given date and everyone takes their turn. Better than some people just being very unlucky and never getting in (or never winning the Powerball!).

  18. Real queues involve standing in line for whatever time it takes. How about having the line start at Foresthill on Cal St. – maybe some red velvet ropes. Check-out is at Rucky Chucky on Dec 6, 2014. On the other hand, systems that evolve to meet the needs of many aren’t bad, just because they’re a little complicated.

  19. Great post, Craig. Call me old school but I’ve always been more of a first-come, first-served person. While it was always stressful, I kind of liked those heady days all those years ago (was it ’06?) when the Way Too Cool signup frenzy happened. I don’t know, I guess I just like the purity of lining up for entry.

    And, there is also the supply/demand argument. I imagine there are more than a few MBA types who are observing the growth in ultra running, particularly with the marquee events, and are dumbfounded that prices have not gone up accordingly. Just look at tuitions at highly selective colleges! Of course, I am not at all suggesting that WS or Waldo or anyone else rise their prices because of demand but it is something that has happened in other places and causes one to wonder, what if.

    Finally, I certainly don’t envy you, Meghan, and other RDs in high demand events as I imagine it can be extremely difficult to keep customers happy when many of them are chronically disappointed.

    • AJW, I loved those WTC entry mornings, too. Perhaps I liked it because I always got in even when it filled in single digit minutes.

      In terms of supply and demand, my hunch is WS could charge $1500 to $2000 and fill to capacity each year. Waldo is probably around $200-250. Fortunately for the community, many of these popular races are run by non-profits with volunteer boards and we aren’t trying to get rich.

      Finally, there isn’t any place else I’d want to be except in this seat. Not sure if Meghan agrees with her seat – yet :-)

  20. Pingback: Daily News, Mon, Mar 3

  21. AJW, you are a Dad!

    Anyhow, I have always been intrigued by the process used by VHTR in VA. Their process is very fair, open, transparent to all and tied to how the stock market closes on a pre-determined day. I think they are using elements from both FCFS and lottery.

    • Michael, the MMT lottery is indeed a very transparent lottery process with an ordered wait list. I cannot argue against the brilliance of the randomness of the lottery. But, it is a lottery. The ordered wait list places volunteers, past winners, five year finishers, all ahead of the previous year lottery losers. It gives no credit for being a lottery loser more than one year. And, similar to the old WS two-time loser rule (ironically, this is what MMT called it also) it quickly becomes overwhelmed with losers from the previous year. It doesn’t scale for ordered selection as demand increases.

      https://www.vhtrc.org/mmt/entry.htm

      • Craig, You obvioulsy studied the VHTR process more that I did. I should have known that!
        The question, “Randon vs Ordered Selection” does not have a straight forward answer. Maybe the answer is somewhere in the middle and I do agree that the lottery may need to change at some point. For example, look at NYC. They had a 4-time-lottery-loser gets gurateed entry on year-4 rule and they abandoned it in 2012 making 2014 the last year one may enter as a 3-time loser (lucky me I am in this group I am running it in 2014) – They also keep raising the price making it the most expensive Marathon in the world and they still receive more than 100K applicants each year and there is fee to get in the lottery.

        This debate is very good and some great ideas have been introduced. Maybe you all can come with a better way to do it which will gurantee entry at some future point with a multi-year progressive waiting list. Just don’t introduce a first come first served rule relying on technology and resulting in crashed websites. Maybe allow time windows when people can enter this list.

        Anyhow, we really do aprreciate what you do and the new qualifying standards were definitley needed.

  22. WS100 is “The” 100 to run, like Boston or New York is for road marathons. Treating it as such I think would help the entry process. To that end make the qualifier a 100. Want to run the 100 that started it all, run a 100. I think this requirement would thin the entry herd a bit.

    • Anthony, this post wasn’t specifically about the WS entry process. There are many races out there trying to deal with the increasing demand. What do those races do when demand exceeds supply? How do they fairly choose who gets to run?

  23. Thanks for putting this touchy topic on the table, Craig. Like the analogies you are making with regard to way to manage queues. I’d add the airline frequent flyer model one which I believe WSER is already leveraging anyway through special considerations, sponsorship and volunteering spots.
    I don’t believe a 10 or 20-year long list is very practical in our sport. We already have to register to some events 10 months ahead of time and many things can happen while training for an ultra. Not just physically (injury) but also with the rest of our lives, personal or work-related, with such an advanced and unpredictable planning.
    I feel that the extra tickets for returning losers bring more fairness to the lottery approach and kind of bridge with the longer-term queue your are suggesting. But, looking at how Mark Tanaka keeps losing year after year (I think 7 times now), that’s not enough. That’s where the automatic entrance for 2-times losers was good in the past, in my opinion.
    Last but not least, you (the WSER Board) could even go further and use as much subjectivity as you want to pick the field you want, with the mix of runners you want (elite, international representation, long-time applicants) through applications (and essays like the Barkley Marathons), the way Colleges do in the US. A possible blend of objective performance (e.g. race results, like universities use ACT or SAT scores), motivation and other criteria at your discretion. And you could have several categories, with preset contingent, on which you’d apply different criteria.
    Oops, did I open a new can of worms for our sport? Hope not. Like someone said in the thread, there has never been so many opportunities to run ultras and the number of 100-milers is just exploding.
    Run Happy out there!

  24. I think there isn’t going to be a one size fits all model for fairness in race entries.

    I agree with others that say that a first-come, first-serve model seems fair when “serving” takes place over the course of days or weeks, but is much less satisfying when “serving” occurs within seconds. I’m uncomfortable with discriminating against people in caregiving and lifesaving professions, and that is effectively what happens when a race sells out in less than 13 hours.

    A multi-year queue sounds fair on paper, but kicks the can down the road. In upcoming years, how long will that queue become? With Western States, would you not be five years deep? If I apply in 2017, how many years will I have to wait before my name comes up? Do I have to keep qualifying? What would that do to the tone and feel of the race if the queue got up to 10 years?

    If a lottery seems like an acceptable alternative to a community, then why fight against it?

  25. As Jean points out, I think managing a multi-year queue would be challenging with people dropping out with injuries or other commitments. However, if the process was automated and could estimate at any given time when one would be able to compete, planning ahead should not be any more difficult than it already is. For example, if the top 200 people on the list get in and you are in the top 200, you know you can run the race at least a year in advance, which is actually more time to prepare than the current process. The key is that the list is transparent and you know where you stand at all times.

    For people with unexpected commitments that come up a couple years after entering the queue, maybe have an option to move down the list by 200 spots for next year, after all you have priority over anyone behind you anyway.

    If this process could be automated, I think it could be really nice, alleviating stress on both the runners and race director.

    Of course, you could still have guaranteed entry for more competitive runners whether it is qualification standards like Boston or the MUC.

  26. I’m not sure you can really devise a “fair” system. Everyone has their own idea of what fairness entails. Unlike your example of standing in line, the internet doesn’t really work that way. You have 400 people with their finger on a button ready to go. They are all “in line” per se at the right time, but one person’s internet is provided through Comcast and the other is through Qwest. The Comcast line goes down 2 minutes into the registration process so one of the people that was queued up and ready to go loses their connection and has to start over. Too bad for him. Race is sold out. His commitment was there. He followed all the rules, but his internet provider let him down. Another situation – You have two people who have access to one computer, and both want to sign up for the race. Which one do you try to register first? Can you possibly fill out the information fast enough so that they both get in?

    I think internet registration, while simplifying things, has also really complicated things. The people who are registering expect that they system they are using will work.The RD’s who are using the system expect that it will work. And then, all of a sudden, it doesn’t. A server goes down. Or too many people are let in even though a cap has been set in the system. People are left wondering why, if person x and y took the exact same steps, person x was the one lucky enough to gain entry, while person y didn’t. How was that determined? While on the backend it is technically a first come first serve system, it seems oddly random due to the way internet lines, connection speeds, proxy servers and all sorts of other things work behind the scenes.

    One idea that hasn’t been mentioned that I’m curious about, why not exclude people from a race who have already had an opportunity to run it? Or exclude them for a year. You write the rules in such a way that you don’t get to put in for the lottery or you don’t get to register for the race the year after you got to run it. With some of the races that are completely oversubscribed this could be a fair way to manage it. Let people who ran it the previous year still gain entry through competitive routes, i.e. winning different races, placing in the top XX number from the previous year, but for the lottery or general sign up there’s a year where a runner doesn’t get to put in for it.

    But the systems aren’t really set up for that. Instead the systems seem to be set up to encourage or reward multi-year runnings of the same race. Special buckles, pins, parking spots, etc … if you complete the race a certain number of times. Perhaps that is a place where a change could be made. To which people might cry “but this is my favorite race!” Yet if that is the case there is nothing to preclude you from getting involved. Run it one year, volunteer the next, and then try your hand at whatever registration process there is the next year to work to gain entry again.

    The essay idea that has been broached is an interesting one. But my heart tells me that unless you can strip identifying data from applications, it’s just going to become all about who you know. A bunch of people sitting around a table looking at applications where it is less about the essays and more about the personal experiences those people in the room have had with the applicants. Which on the one hand is fine. Frankly, if it’s your race it’s your rules. If you want to just let your friends in that is completely your prerogative. But on the other hand it becomes a club that you can’t break into unless you know the right people. And there are probably some really great people that would be missed in this process b/c a person hasn’t been able to break into the club yet.

    I still come back to the idea that there really isn’t anything that is going to be deemed as fair. There are less spots available than runners who want to run. End of story. People are always going to feel like there is a cult of exclusivity to which they cannot break through, whether it is because of an internet connection or lack of luck in a lottery. The upside is, in the end, this will probably lead to good things for this sport. There are more 100s springing up on the calendar. I’ve seen a few 100k’s crop up this year. RD’s are hard at work trying to fill the demand for races.

    Thanks for the opportunity for this discussion. I think this is a fascinating subject and one to which there is no easy answer.

    • Indigosage, you are right, historically our races have placed value on longevity, on multiple finishes. We celebrate those who can come back year after year. When you look at a guy like Twietmeyer and his 25 WS finishes including 15 straight top five finishes how do you not shake your head in amazement. I have argued in the past that races with those types of runners have benefitted from those runners. They become the legends of the race. They take their passion and excitement for the race back to their communities. I’ve argued that a race full of first-time runners would not be the same. That we’d be remiss if we ignored the contributions of those types of runners to the popularity and success of races. But perhaps it is time to rethink that as the sport is in a very different place today.

      • I couldn’t agree with you more and I think that’s why it is such a conundrum. It would be such a shame to give up those legends. I look upon this folks with awe and it’s exactly that type of thing that stokes the fire for those of us just getting into the sport. It’s the same reason why I don’t like the idea of dispensing with the 100k qualifier for WS as I think having some folks at that race who are trying their hand at the distance for the first time is a part of the race’s history. So the question becomes, how does one honor history and meet the needs of folks today at the same time?

      • I agree wholeheartedly with this one. I was drawn to WS in large part due to the proud legacy of the race and the runners who built that legacy. Gordy Ainsleigh, Tim Twietmeyer, Ann Trason, Doug Latimer, Bjorg Aulstreim-Smith, the Jims, it is on their shoulders that we continue to run today. I get chills every time I step on that trail in large part because of the history these legends built there. It’d be a shame for a selection process to crowd those folks out if they still wish to race.

        That said, I also have great respect for those who have run a fair share of WS’s or HRH’s and who voluntarily choose to step away.

  27. Even if the technology could be made more robust, say with an instant ticket to allow a 30 minute registration window, it doesn’t feel more fair to me to have a race for an eventual 0.1 second window of a mouse click to be first served. And agreed that long time-windowed advanced registrations don’t work for ultrarunning. When technology limitations are almost certainly going to irritate people as predictably as for Waldo yesterday, runners have now accepted lotteries. And yes, the lotteries should and could be made more difficult to get into, using qualifying standards, volunteer requirements (Craig’s favorite), and short time windows like a few hours.

    That said, it seems the real problem now is how to reward people for annual persistence and I know it’s not just about WS, but that is where the Hardrock system for first time applicants would be better. One additional ticket for each loss is not enough, and increasing two-fold for each year seems fairer to me. But really it’s all about each race and the RD or board for that race to decide. Differences between each race make it more interesting and bloggable!

    It’s never going to be fully fair in the sense of allowing normal people to run ultras. Sure you could be like the Olympics, UROC or Stanford and let in the most qualified applicants. And yes, some races should definitely do this, but if lots of races went this way the sport would lose so much of its beauty and history. How cool is it that we can run in same race as Timmy, Ian, Nick et al?

    Clem of course thinks this is all hilariously funny as he teases me that he’s looking forward to when I’m a 20 time loser and still trying to get in when I’m 80 years old. It is kind of fun to be at the top of the loser list. There’s lots of other great races, no waiting list for Sri Chinmoy 6 Day yet!

    The other humorous part is that people feel like pricing has not been used as a demand limiter. Someone should plot the entry fees of races like W2C and Miwok and others over a 20 year period… But I digress.

  28. Lottery is fair. Law of probabilities should give comfort. It is a law…

    While I sympathize with those who have been unlucky, sometimes very unlucky, this is the best bad solution. Increasing tickets for every year entered is a nice touch. A queue process is too problematic as the line grows, dreams dashed (if too far back) planning is destroyed (how do you plan when there is too much variability in list-as I would assume there to be when you take into account potential qualifying standards, injury, life events, etc).

    A group of friends and I showed up at the ranger station in Bishop to get over-night permits for Whitney. We got there as early as possible, fully expecting a queue. One man was there ahead of us and protected his “ground” like it was his special right to a chance to climb just because he got up earlier than us and was willing to “work” for his spot.

    It turns out that the station switched to a lottery process (hand a ticket to everyone waiting). You could tell the man was visibly shaken, his world turned upside-down. At the time I was pretty convinced that it was the right thing to do. Otherwise I can imagine Black Friday like circus where people forgo comfort, sleep and human decency to fight for a coveted “spot”. That seems like the obviously wrong thing to do.

    In the end, the right to climb Whitney, or the right to run a capped race (at below market value none-the-less) is a huge privilege, and not actually a right at all. As a privilege I think it is wrong to assume that you should be able to outwork everyone else to earn it (otherwise you might was well make it market value).

    • Jeremy, thanks for the perspective. While I agree it is a privilege not a right to run ultras or to climb a mountain, some people get real passionate about these things. And real disappointed and/or angry when they don’t get selected. I suspect this is only going to become more and more of an issue with the popular races in the future.

  29. Craig here is another Idea. After you take out the top 10 spots any other you hold back. Split whats left into thirds 1/3 first come first serve, 2nd 3nd Lottery but only first timers. Last third those that have completed before.

  30. This is a fascinating discussion.

    I received a confirmed entry to Waldo at 7:15 AM on Saturday, only to be told a few hours later (in a personal e-mail from the Queen herself! I almost fainted.) that Ultrasignup had allowed in 60 extra people (over the cap of 165) and that my confirmed entry had been cancelled.

    I was on my computer at 7:00 exactly, and I tried mightily to sign up. By the first-come, first-served rule, I should have gotten an entry. Instead, Ultrasignup delivered me “server unavailable” error messages, told me my shopping cart was empty, and left me in a couple of other states from which my only resource was to reload the Waldo page, start over, and try to register again. And again. And again.

    It is VERY clear to me that Ultrasignup did not have adequate resources to handle the situation that we saw on Saturday. One commenter above suggested that it was like the Apple Store on new-product day, but that’s not it at all. We have all seen the overnight campers, the long, orderly lines, and the customers willing to pay much more than the product will cost in a few months. That’s NOT what happened at Ultrasignup on March 1.

    The analogy of a Wal-mart on Black Thursday where there are 600 shoppers and only 35 Lord Business’s Lair Lego sets (or whatever the hot new limited edition toy is) is more apt. Except that this Wal-mart has six front doors, two of which don’t open, and two of which route you directly from the entrance to the checkout counter and the exit without allowing you to pick up any products from the aisles.

    Meghan (the Waldo RD) did the best she could under the circumstances. She contacted all of the rejected runners ASAP with the bad news. After the dust had settled a bit, she offered us all automatic entry into next year’s race. That’s above and beyond, given the limitations of the race’s permit. I will be at this year’s race, cheering and volunteering.

    If I were Meghan, I would have serious questions for the people at Ultrasignup. What is the race paying for if not a sign-up service (it’s right there in the site’s name) that works? It gave hundreds of people error messages, did not process people in the order in which they arrived, and gave faulty confirmation messages to 60 people.

    The one thing that does frustrate me that nobody has brought up is this: I finish races. So far, in a running career spanning hundreds of races and over thirty years, I have never DNF’d. Maybe I don’t push myself hard enough. Maybe I am a special flower. Who knows? But it will niggle just a bit when 30 of those lucky registrants don’t even make it to the starting line, and another 30 or so of those Wal-mart shoppers who were lucky enough to get through the right door and get that Lego set manage to go all of 32 miles on race day and drop out at Charlton Lake. Sour grapes, I know. I’ll get over it.

    If you were one of the lucky ones who managed to get a Waldo race entry, please train. Please take care of your body. Please study the course, talk to people who have run it, and run on the course prior to the race if you can. Plan your nutrition and hydration. Respect the people who wanted to run but could not by preparing for and running your best race. Respect the Waldo.

    Co Jones (three-time Waldo finisher)

    • I had the same experience at checkout, (there has to be a better way!) and I was thrilled to hear that we 60 or so rejected entrants would get first crack at registration next year. I also agree that finishing an ultra multiple times should count for something. Being really consistent should count for something. A race should be competitive for those who can be, but also available to the older, seasoned mountain goats. The diverse field is one of the fun things about ultras.

  31. Craig – thanks for soliciting input on this. However selection processes of the future are managed, it’s nice to know that the folks who will be directly shaping those processes are attuned to the concerns of the community. My two primary concerns – the traditions of some of the older races (particularly WS100) and the fact that the queuing analogy isn’t entirely accurate – have already been broached. At this point I’d like to just put in one more plug for a lottery process. In fact, I think the current (as of now) WS100 process is about perfect. It remains democratic and allows for a mix of first-time entrants and more dedicated entrants, but also gives extra weight to those who are more committed. It is far more transparent than an electronically-mediated queuing process. And faster runners can still race their way in. I haven’t actually built and run a simulation, but with the exponential increase in tickets, those folks who are currently 6 or 7 time losers should have better than a 90% chance at getting in next year. I agree with the previous commenters who observe that when demand exceeds supply some folks will be unhappy regardless of how spots are allocated. I’d also like to chime I regarding the comment that these races are ‘below market value’. While this is undoubtedly true for a race like WS100, raising prices simply to match supply with demand would ruin races that do this, regardless of whether a profit were realized. I would hate to see so many people priced out of the market and would be saddened to see ultra running becoming exclusively another rich man’s pastime. Doubtless this is something that is already clear to the WS board and others directing established, in-demand races.

    • Ethan, yeah, it seems that we should steer clear of technological solutions, whether it be for first-come first-serve or lotteries. It’s interesting to note that both WS and Hardrock, while both loaded with high tech prowess, do manual lotteries.

      We ran Monte Carlo simulations on the last WS lottery applicants with the new 2^(n-1) model. I’ll include the old linear model probabilities in parens. 1st year = 5.3% (6.5%), 2nd year = 10.4 (12.6), 3rd year = 19.7 (18.3), 4th year = 35.5 (23.6), 5th year = 58.5 (28.5). The multi-year lottery losers didn’t start gaining any significant ground until the 4th year. Of course, we have to guess on how many applicants we’ll have in future years, but it will never reach 100%.

      Waldo and other races just facing the “problem” of demand exceeding supply will resolve this problem in a variety of ways. Some of the for-profit enterprises may indeed increase the price to maximize profits, but many of the most in-demand races are run by non-profits which is a good thing, IMHO.

  32. The thing that annoys me is when people sign up for big races, taking spots from others, and don’t take it seriously. The solution is very simple. Do what Boston does and only reward those that are at least semi-serious about the sport. Have qualifying times that are pretty strict by age group. Make it so that only, say, 15% of 100 milers can qualify for WS100. If you’re a male under 40, you have to run 100 miles in under something like 18 hours, under 20 if you’re in your 50s, etc. A little slower for women, and have a special category for people with physical disability. A small number of spots can be held for charity entries for people that, for whatever reason, are simply incapable of hitting the qualifier. Don’t like it? Get off your butt and train. Anyone is capable of getting into the top 15% of their age group if they will just train better than their competitors.

    • No doubt popularity of 100 milers on a track would increase in this scenario. A qualifying time for a trail race is like comparing cross country times, too variant. A qualifying race(s) at least evens the variable of terrain.

      As for you premier races like WS100… One piece of the Ironman qualifying that could be considered is the top x% of finishers in certain age/gender groups also qualify, along with the currently used top overall finishers qualifier. There is plenty about Ironman registration that I don’t like, but this part is intriguing to me (and I’m 30, I wouldn’t be getting these slots).

      One last side note that may make me unpopular… I was a first time lottery winner at WS last year and I actually didn’t want to be… I entered the lottery to increase my odds for 2014 or 2015 and got selected. I’m not saying I didn’t appreciate it (it was an honor and and experience I’ll never forget), I just was hoping for a later date – so the lottery can also do a slight, though less so, disservice to the other side of the pendulum as well.

  33. I like the idea of a queue. I was fortunate to be selected for WS this year on my 3rd try but would love to see those who have been waiting the longest get the spots. To be honest it would help me plan my race calendar if I knew I was say 400th on the list I could reasonably guess that it wouldn’t be this year but next as long as I have a qualifier. I guess the downfall of the queue would be that one could reasonably expect to rise to the top every 4 or 5 years? Maybe longer, so WS would be something you might only get a chance to do a few times. I will be doing my best to show up prepared to be a finisher. I feel like I owe it to every person who didn’t get drawn to give 100%. I know this is a special opportunity and it drives me nuts to see people not even show up or drop due to things they had control of ahead of time.

  34. I think the Hardrock process is most fair. Getting exponentially more tickets for every year you “lose” heavily favors the multi-year losers, whereas in WS lottery the increase of chances is linear. HRH is essentially a multi-year queue with some random chance in it. If you stay in the queue you’re very likely to get selected by year 5 (or whatever year the math works out it is) because of exponential growth of chances. Also, HRH has a forgiving qualifying policy by letting you keep the qualifiers for 2 years. This way if someone has a life event and don’t manage to qualify, they won’t lose their extra tickets. Finally, I really like the requirements to having to qualify every year (or every 2 years), because it rewards people committed to the sport. If a multi-year queue is implemented, continuous qualification should be a requirement. Just like finishing the actual race, getting in requires perseverance!!

    Other ideas are letting people defer admission by 1 year, but you have to let the RD know far in advance and qualify next year. The college application/Barkley/essay idea is interesting too, but seems like it would require a lot overhead from the race committee. Same for managing a wait list. I’ve been saying for a while that Ultrasignup should implement an automated wait list feature and make it available for race directors. That way popular races always start at capacity.

  35. With regard to supply, demand, and prices that could be charged, this is already happening. WS might not charge as much as it could, but that doesn’t mean it charges as little as possible either. $400 is definitely on the expensive end for 100 milers. Western States is a huge operation and thus costs more in order to participate in it (this isn’t even touching contributions from sponsors). On the other hand though, I’ve run Old Dominion twice and paid, I think, $135 for the entrance fee. Of course, Old Dominion is significantly smaller, offers zero swag, and is an entirely different experience. Each race has its place. My point isn’t to say that WS charges too much, to be clear; rather, I’m simply attempting to say that $400 is still $400 and that the race could probably be run at a much lower cost by sacrificing some things that make the race exactly what it is. Just my two cents.

  36. Craig,

    Thanks so much for soliciting feedback and encouraging a conversation on this. Beyond the sheer numbers issue of the lottery, the overall problem with the WS entry system is that it’s highly regional and greatly disadvantages east coast runners.

    In this latest iteration of the MUC, there are no races east of Wisconsin. Sure, you can say that east coast runners should get on a plane and race for a top three or four spot at Sean O’Brien, Rocky Raccoon, or Lake Sonoma. But the reality is–beyond the sheer expense of getting to one of these races because of travel costs and time away off from work–we can have brutal winters in the east coast. This winter, temperatures (not wind chills) have regularly been well below zero in the morning. Our roads are covered in ice. Our trails are under two feet of snow. And that’s not even considering the issue of heat acclimation. Despite these challenges, east coast runners like Jason Lantz and Larissa Dannis almost qualified for WS this year at Rocky Raccoon but fell just short. How would they have done if they lived in Texas or on the west coast? Almost certainly much better.

    There are many ultrarunners on the east coast and we take the sport seriously. But the system as currently constructed gives little opportunity for east coast runners to make it into WS beyond the abysmal odds of the lottery. WS is on top of the ultrarunning race circuit for the time being. But if you don’t provide more opportunity for east coast runners to gain access to the race, you risk becoming a regional race. In a few years, WS could easily become a west coast championship with some race becoming the equivalent on the east coast. Vermont 100, which sold out in under two hours this year, is already coming close to that status.

    How to solve this problem (assuming that you do view it as a problem)? I suggest providing additional automatics to the winners of a select number of large and challenging/historic 100 mile races. Something like the following: Vermont, Massanutten, Grindstone, Wasatch, Angeles Crest, and Pinhoti. This would also address the clear contradiction of providing automatic entry for a flat 50-mile race like Ice Age (and not for a hilly historic 100 mile race).

    Regardless, I’d love to hear your thoughts on the issue. Do you have any plans for adding east coast races to the list of automatics? Is it even important to you or the WS Board to address the current bias against east coast runners in the entry system?

    Thanks again for providing this forum for discussion!

    • Hi Jonathan, thanks for your comment. Sounds like your concerns are primarily about the Montrail Ultra Cup. A couple of thoughts:

      1. Montrail wanted all the races in the MUC this year to be post-WS lottery. Not a whole lot of choices on the east coast between December and June, with the criteria we used for selection (can’t have another shoe sponsor, competitive field, quality, etc).

      2. AJW created a new 100K race in Virginia with the goal of being in the MUC next year, to explicitly address this “east coast bias”. Only 39 people have signed up for the race. I’m coming out to help, maybe see you there and we can talk in person about why so few east coasters have signed up.

      3. The MUC will continue to evolve. With the change of qualifying standards moving up to 100K, you’ll likely see the MUC follow suit.

      4. WS becoming a regional race? There are as many critics from the local constituency that we have disadvantaged them with the new qualifying standards and are becoming less “regional.” We also get feedback from overseas runners that we are biased against them, while at the same time get feedback from North Americans that we favor the overseas runners more. We also hear it from older runners who think we’re biased against them. We hear it from the Rocky Mtn runners that we’re biased against them. I can assure you we are very aware of all of these biases whether they are real or not. Check out this article out from my local newspaper.

      http://www.auburnjournal.com/article/western-states-100-decision-raise-qualifying-standards-draws-concern

      • Hi Craig,

        Thanks so much for your detailed response. This is great information.

        I agree that my primary concern is the current structure of the MUC and the implications for runners on the east coast. In particular:

        1. Requiring that the qualifying races take place between Dec and June (and without a shoe sponsor) leaves few options on the east for obvious reasons. The only current option that springs to mind is Bull Run Run but they’re already pretty full.

        2. I LOVE the idea of AJW’s new race but few people on the east are ready for a competitive ultra in mid-March this year. I can’t speak for Virginia, but it’s been absolutely brutal in New England. I still have over a foot and a half of snow on my yard and ice on my roads. I run every day but it’s very difficult to get in very long runs because of temps/wind chill and impossible to run on trails these days. That’s why I’m hoping for some kind of automatic entry option on the east coast during a time of year in which east coast runners are able to train and race effectively (mid-May through October). Any chance AJW would consider holding his race sometime in early to mid-May? If it were part of the MUC during that time of year, I bet it would sell out quickly (and have a nice competitive field of east coast runners).

        3. It would be great to see more 100k or 100 mile options in the MUC (though, again, hot trail runs in February significantly disadvantage east coasters).

        4. I definitely do not envy the position that you’re in. You’ll never make everyone happy with so few slots and more and more people wanting to enter the race. And that’s why I very much admire the way you’re reaching out for feedback. WS is lucky to have you.

        Thanks again. I wish I could come out for AJW’s race but it’s too far from New England this time of year for a road trip (with yet another snow storm expected later this week!).

  37. Jonathan and Craig, Sorry to reply late but I have been traveling. Craig said it right, I created the TJ100K in an attempt to fill an east coast void on the Montrail Ultra Cup and to possibly gain consideration as a Western States qualifier down the road. While I have no direct connection to the WSER I am a member of the WS “family” due to my relationships and history with the event. I can assure you that everyone involved with WS understands the perception of bias and they are working to address that perception. Then, at the same time, the event is also evolving and changing in ways we could not have dreamt of five years ago.

    As for the Thomas Jefferson 100K which takes place next Saturday, I am committed to provided east coast runners with a premium event. If, over time, we become a WS qualifier, a MUC race or even a UTWT event we’d be thrilled. But, in the meantime, we seek to provide a solid, high caliber race.

    My one disappointment with this year’s event is, quite frankly, that more people haven’t signed up. After hearing all the gripes from the east coasters I assumed that this would be an attractive new race. Instead, I had many people asking me if there was a relay division or a 50K option, which there isn’t and won’t be. In the end, I guess, it’s just a matter of putting your money where your mouth is. If you want a MUC race in the east you need a big draw, if you want a big draw you need runners. It’s a classic chicken and egg scenario.

    And hey, it’s not to late to register, we’re open until Thursday:)

    AJW

  38. Hi AJW (and Craig),

    Many thanks for your comments. After putting all the work, effort, time, and energy into the race, it must be very frustrating to have low registrations. The course looks fantastic, the website is great, and you’ll no doubt put together a first rate race experience for the lucky ones there this year. I’m sure the numbers will be much larger next time.

    It’s hard to believe that Montrail has abandoned the entire east coast in the Ultra Cup. I can’t speak for their sales/marketing strategy, but there are a lot of ultrarunners on the east who take the sport seriously (and buy a lot of shoes). Regardless of the number of runners for this year’s TJ100k, Montrail should seriously consider adding it to the MUC next year, hopefully in late April or early May (because of the obvious weather considerations). AJW, you’ve already more than proven yourself as an ambassador for the sport and you’ll be an amazing RD. We all know that if the race were added to the MUC, it would sell out quickly and attract a very competitive field. Sean O’Brien didn’t have to “prove” itself before being added to the Cup.

    Again, I think the problem this year is timing and weather. At least in New England and New York, few people are ready for a competitive 100k in mid-March this year. That may be why you’re getting requests for a 50k or relay. And the terrible weather just keeps coming–we’re expecting over a foot of snow later this week, making travel to Virginia this weekend nearly impossible (I’ve seen a horrible accidents on the highway this year and have had two separate flights cancelled because of weather). The problem certainly isn’t the enthusiasm of runners on the east coast as races in decent weather sell out very quickly (e.g. Vermont, Massanutten, BRR, Stone Cat, etc). If Montrail wants a MUC race in the east, they’ve got to do it sometime between late April and early-May.

    I really wish I could make it down to the race this weekend. It’s going to be fantastic. Hopefully next year!

  39. The solution is to auction off the race entries. The highest bidders get entry. The excess funds can be given to charity.

    Of course some entries can be set aside for elite runners, and some other entries can be assigned by lottery.

    Why not do that?

Leave a Comment