The Exceptionally Fair USSA Points System

By Published On: August 18th, 2016Comments Off on The Exceptionally Fair USSA Points System

There were 747 scored USSA races contested in the 2015-16 season. That means ski racers around the country had 747 opportunities to lock down valuable points toward building their athlete profile. Of all those races, two stood out as events that raised a question about the fairness of the results.

Dave Waller, the chairman of the Alpine Classification Working Group, developed the process that USSA now uses to identify outlying races like these. His background is in computer science and statistics. Professionally, he used advanced mathematical techniques to solve complex business problems. He was introduced to USSA when his daughter got involved in ski racing in the Central Division, and he attended his first USSA Congress in the mid-to-late 90s where he found a way to utilize his skills in the sport his daughter loved.

At his first Congress he met Allen Church, former chairman of the Alpine Classification Working Group. Church encouraged Waller to lend a hand in the Fall adders and the various analytical aspects of the races. The big breakthrough regarding penalty adders came around 2010, when USSA decided it wanted to apply a minimum penalty to races.

“What they were trying to get at was to stop the spring points bonanza that was going on in certain parts of the country,” Waller notes.

“There is no huge advantage to low-penalty races, at least not to the extent that most people think.”

So, Waller and Paul Mahre started running an analysis of USSA races, and what they found was surprising. There is no huge advantage to low-penalty races, at least not to the extent that most people think. However, they did identify what they now refer to as exceptional races.

According to a recent USSA email to members, “The cause of an exceptional race could range widely from penalty manipulation to random situational circumstances. In the past, there have been examples of high-level athletes intentionally tripping the start wand or throwing their skis sideways before the finish to artificially ‘slow’ their times. … In other cases, high-level athletes have simply skied conservatively in challenging conditions, been too sick to perform at their usual level, or fallen in one run and still been able to finish inside the top 10 of a weak competitive field.”

Adders, which have ranged between 20-30 points, bring the results of the exceptional races in line with an average race and maintain the integrity of the overall USSA points system.

“The purpose of the (points) system is a little bit like a golf handicap,” Waller explains. “If you are an eight handicap, you shoot around 80 every time out. In skiing you say, ‘I’m a 20-point racer.’ Your points allow you to compare yourself to any other 20-point racer in the country, and you should be competitive with them.”

If a 60-point skier scores 20 points, that athlete should consistently compete with other athletes in that points range. If an athlete scores far below their skiing ability, in a sense they are “devaluing or compromising the accuracy of the system,” according to Waller.

This USSA initiative is not about a few individuals having a standout race. In fact, it addresses races where the majority of athletes in the field have their best race of the season by a long shot. Jeff Weinman, the director of competition services at USSA, sees this initiative as an opportunity to keep the races fair.

“The ‘why’ is: We want to ensure a fair system for every athlete, and what we’re identifying isn’t necessarily a case of manipulation by an athlete or by an official, but finding competitions that are just so far outside the norm that really they give an athlete an unfair advantage – whether that be selection to teams, selections to other competitions and things of that nature,” he explains.

The two exceptional races from last season – which have not been publicly identified by USSA – were flagged because a substantial number of athletes competing in these races scored a significant point improvement. The three factors that are measured to calculate whether or not a race will be flagged are average improvement between seed points and points achieved in the race, average percent of point improvement, and percent of the field that scored a point improvement.

AverageRacesChart

For example, in one of the races identified from last season, “59 percent of the field scored a result 25 percent better than their seed points which equaled a 35-point improvement. 6.95- and 27.46-point racers finished 8 and 11,” according to the meeting minutes of the Alpine Classification Working Group from the 2016 USSA Congress meeting. Basically, there are circumstances where sub-20 point skiers score 100-point finishes, warping the results of that single race and also affecting the entire scoring system nationwide.

In some of these situations, athletes can be disqualified or sanctioned for race point manipulation. More importantly, these circumstances can create very unusual point results that could “distort the integrity of the USSA scoring system, allowing less-skilled athletes to be ranked, seeded, or selected in front of more qualified athletes,” according to the email sent to all USSA members.

Races that meet the criteria are not identified as exceptional until after the season is over, and the process to properly categorize these races is extensive — always keeping in mind that the goal of this process is not to penalize certain venues or athletes, but to maintain the integrity of the entire system.

“I get all the race data about the middle of April,” Waller explains of the system. “I calculate the statistics … for all the various disciplines and genders. There will be half a dozen races that are suspect that I’ll flag. Some of them, I conduct a quick review and see that it’s U14s probably their first scored race and you don’t see any sandbagging going on, so I let it go.”

But when Waller manually reviews the results and sees a race where there are low-point racers finishing in fifth through 10th or worse, that’s the first indication that the race might be exceptional. Additionally, when the Alpine Classification Working Group sees that athletes who finished in the top 20 have never skied within 20 points of their scored result, that’s another indication. Finally, they look at the pace skiers over the course of the year and measure their performance in the race in question against how they’ve skied during the rest of the competition year. That’s how the exceptional races go from a statistical algorithm identifying a half dozen to USSA ultimately flagging two. In the 2014-15 season, one race was flagged by this system as well.

Once they are flagged by the Classification Working Group using the designated control limits, race organizers and jury members are informed and offered an opportunity to explain any particular circumstances of their competition at the annual Classification Working Group meeting during USSA Congress, held annually in May.

ExceptionalRacesChart

At USSA Congress, the Alpine Sport Committee then decides whether or not the exceptional races will “receive a warning, be assessed a penalty adder to bring the race back within statistical norms, or be thrown out entirely,” according to the USSA email. In order to inform athletes and teams affected by these decisions, USSA members will now be notified electronically of any exceptional races within 30 days of USSA Congress.

On an international scale, this practice is yet untested. Weinman, who also chairs the FIS Alpine Skiing Classification Subcommittee, says that an initiative like USSA’s has never been discussed in his committee. That said, he does hope to eventually be able to run FIS data through the USSA system in order analyze the data of FIS races around the world for USSA’s own edification.

“I trust officials will take this seriously now,” Waller shares. “They’ve seen three races receive significant adders. They need to know races producing results far beyond reasonable expectations will be flagged and reviewed in detail.”

As Waller and the USSA team refines the system for identifying such races, more variables will be added to their considerations and, hopefully, the system will become a bit more automated. At the end of the day, according to USSA, “The overall goal of this effort is to ensure that athletes are competing to the best of their ability, and that the USSA ranking system is as accurate as possible within the statistical norms of the sport at large.”

Disclosure: USSA and SkiRacing.com are media partners.

Share This Article

About the Author: Gabbi Hall

A California native, Gabbi moved to Vermont to ski on the NCAA circuit for St. Michael’s College, where she served as team captain and studied journalism. Before joining Ski Racing, she worked as a broadcast TV producer and social media manager in higher education. She can be reached via email at gabbi@skiracing.com