Gunksrunner Ultra Rankings

2

The Gunksrunner Ultra Rankings arose, as most anything important does, out of a conversation on a training run several years ago.  (By which I do not mean to imply that these rankings are in any way important.). We were discussing the USA Triathlon ranking system, by which any USAT member can receive a national ranking simply by finishing three races over the course of a year.  And so the idea for the GUR was born: a comprehensive ranking of every finisher of every ultramarathon in the country.  

Without a central governing body or a complete database of athletes, an all-finishers ranking proved untenable.  But the spirit of the idea remains.  The GUR includes results from every US ultramarathon (and some international races as well), identifying the top finishers and ranking over 6000 men and women.  It’s meant to be as inclusive as possible.

Though the idea germinated from the world of triathlons, the GUR formula is based largely on the World Golf Rankings, which calculate a player’s ranking based on the results of head-to-head competition in tournaments of various difficulty and prestige, against fields of variable strength. Thus, the highest rankings are achieved by the athletes who perform their best on the biggest stages, against the best competition.  The rankings are based on cumulative point totals over the course of the calendar year.  Point values are determined by a formula that accounts for field size, race prestige, and field strength.

Race ratings

Each race in the GUR database is ranked on a five-point scale based on field size and relative importance. Small races with primarily local interest receive a rating of 1.  Larger races (usually with more than 75-150 runners) that attract some regional interest and competition rate a 2.  Level 3 races have larger fields, often 200 runners or more, and draw from at least a regional pool of competition, possibly with some national interest.  (US Skyrunning races and WS Golden Ticket races have a minimum rating of 3.)  Level 4 races are national-class events that attract top flight competition and generally have a strong tradition behind them.  (USATF national championships are given a minimum rating of 4.)  Level 5 is reserved for the true “majors” of the sport, such as Western States, UTMB, and the 100K world championships.  In 2015, only twelve out of nearly 1300 races were give a rating of 5.

The race ratings are then used to determine how many finishers will receive ranking points for that race, and how many points they will receive (please note that some of these values have been adjusted for 2016):

Race Rating # of scoring places Points for 1st Points for 5th Points for 10th
1 3 5 n/a n/a
2 5 10 1 n/a
3 10 15 4 1
4 15 25 6.25 2.5
5 25 40 10 6

Field Strength

In golf, regardless of the size or importance of the tournament, more points are available if more top players are competing.  This is also the case with the GUR.  Every runner in the top 50 of the rankings is assigned a certain “rating value”.  The cumulative ratings value for a race gives an approximation of the field strength, independent of size.  That rating value is then translated into a multiplier.  All points in that race for each particular finisher are then multiplied by that factor:

Runner Rank Rating Value
1 25
2 20
3 17.5
4 15
5 13.25
6 11.5
7 11
8 10.5
9 10
10 9
11-20 8
21-27 6
28-33 5
34-38 4
39-42 3
43-46 2
47-50 1

How It Works

Here’s a concrete example from a recent race, the Bandera 100K.  As the USATF 100K trail championship, Bandera receives a race rating of 4.  (Even without the USATF designation, Bandera would likely be rated as a 4 based on its size, the history behind it, and its relative importance in the current ultra scene; at the very least, as a Golden Ticket race it would receive a 3.)  As a level 4 race, we know that the top 15 men and women will receive ranking points, with the winner scoring at least 25 points.

Next, we calculate the field strength multiplier.  The men’s race saw four runners in the top 50 toe the line: Paul Terranova (#4), Jim Walmsley (#18), Mario Mendoza (#25), and Christopher Dennucci (#26).  These ranks correspond with ratings values of 15, 8, 6, and 6, respectively.  Adding these values gives a total ratings value for the race of 35, which corresponds to a multiplier of 1.8, as we see below:

Total RV Multiplier
> 150 5
120.25-150 4
100.25-120 3
80.25-100 2.5
66-80 2.2
50.25-65.75 2
40.25-50 1.9
30.25-40 1.8
25-30 1.7
20-24.75 1.6
16-19.75 1.5
11-15.75 1.4
7.5-10.75 1.3
4-7.25 1.2
1-3.75 1.1

Therefore, each scoring finisher will have their points multiplied by 1.8.  So Jim Walmsley’s win, which would normally be worth 25 points, actually nets him 45.  

The women’s race had two top-50 runners (Cassie Scallon at #11 and Julie Koepke, #44) for a total ratings value of 10 (8 for Cassie, two for Julie) and a corresponding multiplier of 1.3.  Thus, sixth-place finisher Katie Graff, for example, would receive 5 x 1.3 = 6.5 points.

The current spreadsheet and rankings can be viewed at:

https://docs.google.com/spreadsheets/d/1rwvAO57ZCcZDCgmObGlY_c-eBKStB3eUYDIHDJbVp3c/edit?usp=sharing

Last year’s results are available here:

https://docs.google.com/spreadsheets/d/1ZU3YcJCEFRsDhdnafI6SApYYR-rJSJtyESrbRIj54qs/edit?usp=sharing

Further discussion of methodology, including a (made-up) FAQ, is at:
http://gunksrunner.blogspot.com/2015/06/ultrarunning-national-rankings.html

Share.

About Author

Jason Friedman is a runner and an emergency medicine physician in upstate New York. He as worked as a coach at the high school and collegiate levels, has directed running camps, and has owned a running specialty store, and has provided medical care at the Western States 100, Manitou's Revenge ultra marathon, and the Survival of the Shawangunks triathlon. He has been racing ultras since 2006. He lives in New Paltz, NY with his wife, Jodi, and their daughters, Alexa and Dylan. You can follow Jason at his blog

2 Comments

  1. Ian Maddieson on

    I’m fascinated by rankings and all sorts of other data on ultrarunning, but I see a conflict here between the desire to be ” as inclusive as possible” and the idea that only the top finishers in an event get any points in this ranking.

    • Jason Friedman on

      Ian, you’re definitely right…”as inclusive as possible” reflects the logistical limitations of doing this by hand, pulling results from various disparate sources (Ultrarunning, Ultrasignup, Facebook, various race websites, etc). Without the ability to fully automate the compilation of results, I’m limited by the practicality of it. I’ve made the decision to attempt to include every domestic ultra (and several international ones). The tradeoff is that I just can’t include every finisher in those races.