Aerodynamic cycling gear is designed to perform best in average conditions—the wind a cyclist is most likely to face. Every engineer in the business and many techie triathletes know this to be true, but Ironman Hawaii is far from average.
The Ironman World Championship course lives up to its reputation for heat and wind. Gusts rip across the road, swirling from every direction at some point during the 112-mile trek. At the start, the wind is usually working with the athletes. Then it typically switches to one side for a stretch of road before coming from the opposite direction during the biggest climb and descent on the course. The same wind that helped during the first hour or so smacks the riders head-on during the last stretch. Some days are calm, but others are so turbulent that even the strongest riders are nearly pushed off the road.
Applying national weather averages or meteorological stats taken in Kona produces misleading results because the wind direction and speed is so dependent on location on the course and the day’s conditions. Because of the extreme wind conditions, gear choice for Ironman Hawaii is a unique puzzle, and measuring wind everywhere on the course is the best way to determine what is effective for this specific race.
Mavic research engineer Brieuc Cretoux developed a tool to record the wind angles experienced by a cyclist. This device positions a carbon sail that moves with the passing wind about a foot in front of the rider. As the rider pedals, the system records the position of the sail every time the bike wheel makes a complete revolution. We sent a cyclist out on the Kona bike course in 2013 with this system attached to the front of his tri bike to measure the wind conditions at the Ironman World Championship three days before the race.
The device has a few shortcomings that may influence the data. First, it cannot measure wind angles wider than 30 degrees. Second, the sail that measures wind can be physically jostled by bumps in the road. Third, wind speed is not taken into account, only angle. A tailwind can create wide wind yaw angles, but has a much smaller impact on performance than a headwind that creates the same effective wind angle. Fourth, Specialized aerodynamicist Chris Yu speculated on Twitter that the rider’s body may redirect air crossing the wind vane, influencing the angle of the wind passing over the vane without reflecting the conditions impacting the rider.
Lars Finanger, a former pro triathlete and current Category 1 cyclist, volunteered to ride the Ironman Hawaii bike course with the wind vane. Cretoux equipped his bike the night before the test. Finanger started rolling from the Kona Pier close to 7:30 a.m., one hour after the start time of the professional men’s race and 30 minutes after the age group start, so he would ride the course at the same time of day that the race takes place. He covered the entire route, with five stops along the way. The first few miles twist through town, so without a closed course, he was forced to stop at a few red lights.
Conditions vary dramatically from day to day. Normann Stadler rode to two world titles during the previous decade, one on a calm day and another by punching through the stiffest winds in years. On the day with tranquil wind in 2006, Stadler set the bike course record with a 4:18:23 split. Two years earlier, he also earned the decisive lead during the ride, but split 4:37:58. The race played out the same way in both years, but the difference in conditions was so significant that even the strongest rider in the field was slowed by nearly 20 minutes. The day Finanger rode the course with the wind vane was violently windy, more similar to 2004 than 2006. Former Kona podium finisher Dirk Bockel was out training that morning and tweeted, “That was brutal out there today—wow” in response to Finanger’s description of the conditions during the test. The data supports Bockel’s perceptions: Hawaii’s winds were extreme. Cretoux said he has never recorded conditions even remotely similar.
During another test in a different location with slower than average winds, Cretoux found the rider faced primarily shallow wind angles. This ride, conducted around Mavic’s headquarters in Annecy, France at an average speed of 21 mph, showed that 75 percent of the wind experienced by the rider was within 4 degrees to either side of head-on. And 97 percent of the distance covered offered winds within 10 degrees to either side (20 degrees total). Kona couldn’t be more different. In Hawaii, the rider faced shallow wind within plus-or-minus four degrees for just 31 percent of the distance covered; expand that range to 10 degrees to either side and still only 66 percent of the ride is accounted for. The rider in Kona was fighting through crosswinds ranging from substantial to extreme for the other 34 percent.
The results are depicted in an image in the gallery.
Kona’s infamous crosswinds were on full display during the test, and the result was wind conditions (yaw angles) that are dramatically wider than average. The test conducted in Annecy with still winds doesn’t represent an average ride because many days are likely windier than that one, however. Cycling engineers typically select between 5 and 15 degrees of yaw as the most important range of wind angles, depending on the weather figures and rider speed each company decides to use when designing a product. Aiming for the average is effective if an athlete is going to use one item—wheel, bar, frame, helmet or any other component—for every ride, but this test shows that Kona requires an entirely different strategy.
The effective wind conditions also swung wildly within this one ride, and not every portion of the course is equally important, especially for the professionals. A poor swimmer who needs to chase up to the leaders early in the ride may be especially interested in the conditions during the first portion of the ride, while a breakaway specialist will be most affected by wind later in the ride. Wind angle during the 12K loop around town was very shallow.
Wind started to kick up during the next stretch of the road between town and Kona International Airport. Wind here was strong and sideways, but fairly consistent. Most of the crosswind came from the rider’s left. This type of wind greatly impacts aerodynamic performance without ripping control away from the rider since he can lean steadily against the oncoming breeze.
After crossing the airport and heading toward Kawaihae, wind became less predictable. Wind in the 4- to 10-degree range hit the rider for roughly the same amount of distance, but instead of coming from one side, it swirled from both. While the difference in direction doesn’t impact much in terms of wind tunnel performance, it makes a huge difference for a cyclist on a bike. The 180-plus-pound experienced test rider had to fight to keep the bike on the road and subjectively found this portion of the ride to be very challenging. He came out of the aerobars on a few occasions to brace against the gusting wind. The cost of these swirling conditions is difficult to quantify but has a real impact on actual performance. Gear that helps keep the bike under control can lead to more time spent in the aerobars on this section of the course.
Strong winds coming off the land pushed the yaw angle much wider between Kawaihae and Hawi compared to previous sections. Finanger had to fight to keep the bike upright, but the relative consistency of the wind angle, centered between 10 and 20 degrees, made the task a little easier.
Despite riding faster over the downhill portion of road from the Hawi turnaround to Kawaihae, which reduces effective wind angle, the test rider experienced the widest crosswinds of the entire ride during this section as well as some wind from the opposite side. The wind vane was jarred back and forth while descending from Hawi, and Cretoux speculated that the data could be skewed by disruption of the vane from sources other than the wind. The widest crosswind the vane can measure is 30 degrees of yaw, and it repeatedly topped out during this span. If the device could measure broader angles, this chart likely would have stretched even wider.
Over the final 35 miles from Kawaihae to the finish, the test rider started to crack, and his pace dropped substantially. Early in the ride the wind during this section of the course came off the land but eventually switched, and during the final miles he faced a crosswind from the right (off the ocean) that was working against him.
Conditions during the test were extreme, even for Hawaii. The yaw angles experienced by the rider wouldn’t be as wide if the test were conducted on a more typical day—these data represent yaw angle for a fast rider facing fearsome conditions.