Monday, January 11, 2016

The Shape of Things to Come: Hakone, Where Things Are and Where They're Going

by Brett Larner
special thanks to Dr. Helmut Winter for assistance with graphics

The Hakone Ekiden, Japan's most watched, most loved, most prestigious road race, grabbed my attention when I first moved to Japan in the late 90's.  I started JRN in the summer of 2007 at the time of the Osaka World Championships, and from the beginning one of its main focuses has been trying to make this race, one that I've come to believe stands next to the Boston Marathon and Comrades ultramarathon as one of the world's three great races, visible and understandable to an outside world that for the most part had never heard of it before.  It's starting to take hold; in the lead-up to and after Aoyama Gakuin University's successful title defense at last week's 2016 Hakone Ekiden, print and web publications in Germany, Poland, Italy, the U.S., the U.K. and elsewhere picked up on JRN's work with coverage of this year's race, one Norwegian journalist even travelling to Japan to see Hakone live for a feature article.  I couldn't be happier.

Hakone is the most exciting race there is, presented that way by Nihon TV with the best live broadcast in the sport, but another part of what I'm interested in about it is the numbers.  It seems like just about every year that I've written about Hakone I've said that the teams are the best ever, that things just keep getting better and better.  That's been my impression, but what do the numbers really show?  For several years I've been researching the last 20 years of Hakone, every team roster, every athlete who has run it during that time, to find answers to this and a lot of the common questions and claims I hear about Japanese distance running.  With this year's race on the books here are the numbers, what they say about Japanese marathoning, and what they suggest is about to happen.

The Numbers

I've broken down the typical ability range of Hakone-level university men over half marathon, 20 km, 10000 m and 5000 m into 12 levels, 20 km getting into the counting thanks to October's Yosenkai 20 km, the Hakone Ekiden qualifier for second-tier schools.  With the exception of the spectacularly talented Mekubo Mogusu, a Kenyan who ran three sub-60 minute half marathons his third year at Yamanashi Gakuin University, the 12 levels take in everyone from the very best Hakone runners in the last 20 years at level 11, running under 1:01:00 in the half marathon, 27:30 for 10000 m or 13:20 for 5000 m, to alternates on the weakest teams at level 1 or even slower.  Media and fans love to talk about ace runners, the best athletes on most teams.  In the graphs below ace runner refers to anyone at level 7 or above.

Team levels were calculated based on the number of runners at each level in each school's top ten.  The times within each level are not really equivalent to each other in terms of quality, but they do give an indication of how runners will affect their team's performance at Hakone.  With an average stage length of 21.7 km, someone who can break 1:03:00 for the half marathon is more likely to have an impact at Hakone than someone who can break 29:00 for 10000 m but doesn't have half marathon credentials.

This graph shows the number of runners in the Hakone Ekiden field each year who had PBs at level 7 or better. Here and below, numbers have been normalized to account for small changes in the numbers of teams in the Hakone field from year to year.  After some growth from 1997 to the early part of the 2000s there was little overall change in the decade or so until 2012, apart from a small peak around 2006-2008 when Mogusu and many of today's best marathoners and corporate distance runners including Masato Imai, Yuki Kawauchi, Satoru Sasaki, Yuki Sato, Kensuke Takezawa, Yuichiro Ueno, Tsuyoshi Ugachi and others were running Hakone.  In the 2012-2013 school year the number of athletes at this level exploded and has continued to grow at an accelerating rate.

Hakone teams are made up of ten runners plus up to six alternates.  A similar pattern is seen in the average level of the tenth-best runner on each team, with some growth from 1997 to the early part of the millennium, limited change in the decade or so from 2002-03 to 2012 apart from a small peak from 2006-08, then explosive growth from 2013 on.  This shows that the growth is not limited to the best of the best but runs deeper.

Looking the same way at the average level of each team's best runner, apart from the 2006-2008 peak, distorted here by Mogusu's 59-minute half marathons in the 2007-08 academic year, the progression has been closer to linear.  The best have continued to get better at a steady rate, showing that the improvement has not only been a case of more people approaching a ceiling.  On average, the tenth-best runners on Hakone teams now are almost as good as the best runners were 20 years ago.

Like the level of the best athletes, the average level of the best teams has followed a different pattern while continuing to improve.  Yamanashi Gakuin's 1999 Hakone team and Tokai University's 2006 lineup both broke new ground in quality, and in exactly the same interval of 7 years in 2013 Komazawa University produced a team far better than any ever seen before.  But where the level dropped back toward more normal levels after 1999 and 2006, this time it has continued to trend upward.  Komazawa in 2014 and 2015 and Aoyama Gakuin in 2016 were all better than any pre-2013 teams, and even #2 and #3-ranked teams like Toyo University and Waseda University's 2014 lineups, Waseda's 2015 team, and Waseda and Komazawa this year were better than Tokai's 2006 team.  Not just single star athletes but entire star teams are better now than ever before.

Putting everything together, over the last 20 years the average team level for the entire Hakone field follows the same pattern seen twice above: moderate growth from 1997 to the early 2000s, little change over the next decade except for a peak from 2006-2008, and explosive growth since 2013.  A revolution happened in the 2012-13 school year and it was televised, live and nationwide, with spectacular production values and professionalism.

The Revolution of 2013

After a decade or more of relative stasis, in the four years that this year's graduating class have been in action this revolutionary change has pushed Hakone and Japanese university men's distance running where it has never been, and while there are some signs that it is slowing down by most measures it is still on the way up.  What happened?  I think there were three main catalysts.
  1. Toyo University's 2012 Hakone Ekiden course record win.  Masato Imai achieved national stardom thanks to his uphill Fifth Stage heroics for Juntendo University from 2005 to 2007, sparking greater mass popularity for Hakone.  Under young head coach Toshiyuki Sakai, Toyo's Ryuji Kashiwabara achieved superstardom, celebrity, even, by breaking Imai's Fifth Stage record as a first-year in 2009 and going on to win Hakone's greatest stage the next three years.  In 2012, his last year, Toyo's team banded together to try to deliver the win to Kashiwabara, winning on the strengths of every team member without relying just on him.  In doing this Toyo became the first team to break 3:00/km for the entire 217 km-plus Hakone course.  Sakai's explanation was simple.  "Everybody talks about 3-minute pace," he said.  "I told them to think of a number starting with 2."  A young coach focusing not just on training but on psychology, positivity, on looking past the accepted standards at something higher, all in the glow of national stardom.  A number starting with 2.  Everyone responded, not just the Toyo runners.  Komazawa, coached by conservative old guard Hiroaki Oyagi, came back the next season with the best lineup ever seen up to that point, and it has only gone up from there.  The Hakone Ekiden Museum claims that the modern era of Hakone started with Aoyama Gakuin's 2015 win under young head coach Susumu Hara, but that was an evolutionary development.  The real era-changing revolution came from Toyo and Sakai in 2012.
  2. The NYC Half Marathon.  After Toyo's 2012 win the New York Road Runners invited two of its best runners, Yuta Shitara and Kento Otsu, who had been the top two Japanese collegiate finishers at November's super-deep Ageo City Half Marathon, to the NYC Half Marathon along with coach Sakai.  Japanese corporate runners rarely race seriously when they're outside Japan, but with little to no international experience Shitara and Otsu ran fearlessly.  Just 20 years old, Shitara outkicked World Half Marathon bronze medalist Dathan Ritzenhein (U.S.A.) to finish in 1:01:48, still the best time ever by a Japanese man of any age on U.S. soil.  The reaction back home was huge.  Knowing now that the NYC Half invite was up for grabs, at the next Ageo in November, 2012 Komazawa's coach Oyagi put two of his best runners, Kenta Murayama and Ikuto Yufu, into Ageo for the first time.  Average times at Ageo were almost one minute faster as all the Hakone-bound runners chased NYC invites, a trend that has been true every year since then.  Where only 36 runners had broken 1:03 in Ageo in its first 25 years, 18 more did it in one race.  More high level half marathoners meant more high level teams at Hakone, and the effects spread to the other university ekidens, to the National University Half Marathon, and beyond.  Komazawa publicly credited Murayama's experience at the 2013 NYC Half with inspiring the team to become what it did in 2014 and 2015.
  3. The 2020 Tokyo Olympics.  The September, 2013 announcement of Tokyo winning the 2020 Olympics bid came well after the big change was underway but had a massive impact as Hakone runners realized that they would be at their peak for a home soil Olympics.  In almost every interview since then, virtually every student athlete has said the same thing: "My life goal is to run the marathon at the 2020 Tokyo Olympics."  Course records and world records for depth suddenly became the norm in almost every major university race and even down to the high school level as the competition for the 2020 team got started early. What is it going to take to make that team, not just in terms of times but mentally, emotionally, spiritually?
The Shape of Things to Come

Looking at the average of the ten fastest marathon times in Japan and worldwide each year, up until about the time of the 1992 Barcelona Olympics Japanese marathoning developed at the same rate as it did worldwide.  After Barcelona, where Koichi Morishita won Japan's last Olympic men's marathon medal, the progression in the world standard continued at the same rate as Kenya and Ethiopia really came to the forefront, but Japanese marathoning abruptly flatlined.  Despite a string of 2:06 national records between 1999 and 2002, in the last 20-25 years the average top ten Japanese marathon times have stalled at around 2:10:00.

The possible reasons for this, a physiological ceiling, limitations of training approaches and the corporate team system, limitations of psychology, sociopolitical changes, would take up another article.  But it is notable that, as shown above, for much of the same period of time there was limited progression in quality at Hakone.  Zooming in on the last 20 years in Japanese marathoning and Hakone, the Y-axis inverted for Hakone data for ease of comparison, a possible pattern emerges.

It's not a perfect comparison as the Hakone numbers include some Kenyans, not all marathoners ran Hakone, and Olympic selection years like 2008 cause spikes in performance that aren't reflected at Hakone, but in general, over the last 20 years when there have been more university runners at the ace level, sub-1:03:00 half marathon and the like, 4 to 6 years later Japanese marathoning has been faster on average.  When Hakone fields have been weaker, 4-6 years later Japanese marathoning has been slower.  Researching the Hakone numbers further back past 1997 would take time and funding beyond the scope of this article and the correlation may not survive more rigorous statistical analysis, but it does look like there is a relationship there, and one that makes sense.

Going back to the full graph of the number of ace runners, then, suggests what we are going to see in the next few years.  There have never been anywhere near as many Japanese university runners anywhere near as good as right now.  The runners who were part of the 2013 revolution haven't started running marathons yet.  2016 is an Olympic selection year, so marathon times should be faster than usual.  Four years after the revolution is 2017, six years is 2019, the start of qualification for the 2020 Tokyo Olympics.  Four years from now is 2020, still in range of this year's biggest Hakone names.  If the trends of the last 20 years mean anything, in the next 4-6 years we are going to see a lot more Japanese marathoners running fast times, the first really significant overall change in Japanese men's marathoning since Barcelona.

How fast?  One criticism often given is that Hakone runners are simply hitting the ceiling, physiological or otherwise, at a younger age and that they're not going to go anywhere past that.  Given the apparent ceiling in men's marathon performances over the last 20+ years that's possible, but with the steady increase in quality of the best Hakone runners shown above in mind, a look at the all-time Japanese top ten lists for the four major distances suggests otherwise.

Athletes in green are people who graduated in 2013 or later, while athletes in yellow set all-time top ten marks in 2013-2016 but graduated earlier.  Things have been bubbling toward longstanding national records since the generation of Sato and Imai, part of the 2006-2008 spike in Hakone performances, took over, but nobody could break through the ceiling, not one of them could take one of the major national records.  This would give support to the idea that Japanese running has hit its limits, that Hakone is an illusion and that university athletes are just moving toward those limits earlier with nowhere to go.

But the new generation who have come up as part of the post-2013 explosion look different, more confident, ambitious, already raising the ceiling.  Suguru Osako, Tetsuya Yoroizaka and Kota Murayama all broke national records on the track in 2015, Masato Kikuchi just missing joining them in the half marathon after tying the 20 km NR en route.  It's possible that the number of high-level performances at the shorter distances may be partially due to previous underdevelopment as a consequence of focus on the marathon and the times are still a long way from being globally competitive, but Osako thinks he can go sub-13 and sub-27 on the track and a sub-hour half marathon doesn't seem far away.  None of the major runners in this generation has done a marathon yet.  What will happen when they do?  Kenta Murayama plans to run his first at Tokyo next month along with three of the biggest performers at Hakone this year, Toyo's Yuma Hattori and Aoyama Gakuin's Tadashi Isshiki and Yuta Shimoda.  Maybe they will all burn out and have short careers, but given the numbers above it's reasonable to think that once that ball gets rolling we should see an impact on the all-time marathon lists, and when that happens you are talking real times.  There's nothing to suggest Japanese men are going to start running 2:03 or 2:04 marathons, but given the numbers involved 2:07 and 2:08 should become normal, with 2:06 in range of the top men the way 2:07 is now.  Given how much Kota Murayama and Yoroizaka took off 2:06:16 man Toshinari Takaoka's 10000 m national record last fall, maybe 2:05 is possible after all.

They face the major problem of the slow-moving and conservative corporate system adapting its attitudes and practices to cope with the massive wave of talent coming in, and in some respects it's not encouraging.  Despite both Yuta Shitara and Kenta Murayama having proved at the NYC Half that they had no fear about racing internationally against the best in the world as university students, in the early days of their careers in the corporate leagues they ran with unbelievable timidity at the Beijing World Championships, taking the last two places in the 10000 m after kicking ass domestically.  There's an obvious problem with the psychology the older corporate coaches are instilling relative to what's happening with younger coaches like Toyo's Sakai and Aoyama Gakuin's Hara, but no matter how high and how strong you build the wall, if the wave is big enough, powerful enough, it's coming over.

On the all-time lists above, every single athlete who has set an all-time top ten mark since 2013 was a Hakone star, an ace runner.  Nobody in these last two generations who didn't excel at Hakone has risen to the top level of current Japanese men's long distance.  It's hard to overstate how much it resonated with the public when Imai, the first modern Hakone star, ran 2:07:39 in Tokyo last year.  Some people say that Hakone, created to cultivate Olympic marathoners, has become too big, that too many runners are doing too much too young, that it is burning them out mentally and physically and distracting them from aspiring to the Olympics.  Who knows, that may end up being the case with this new generation, but the numbers simply don't back those claims up.  With the massive popularity of the Hakone Ekiden and the chance it gives for young, innovative coaches to do their thing and for more than just the three elite who someday make the Olympic marathon to have their own day in the sun there have never been so many runners so good in Japan, and it looks like the best is just around the corner.  Will the revolution bring a new golden era or a wasteland?  Let's talk again in another 4 to 6 years and see how things have played out.

© 2016 Brett Larner
all rights reserved


TokyoRacer said...

Very comprehensive analysis! And very interesting.

Scott Brown said...

Yes great to be able to see the data like this. The dynamics are fascinating and now, with the numbers, even more so. Appreciate your efforts putting it together.

Jotham Burnett said...

Awesome stuff Brett but can't help but wonder about doping

Brett Larner said...

Anything's possible, I guess, but it would depend how plausible you think it is that all the universities got together in 2013 and decided to bastardize 90 years of history and suddenly start doping all their male students without touching their women's teams or the corporate leagues doing the same. Personally I would tend toward "implausible."