Luckily for Honolulu, the third largest marathon in the United States behind New York and Chicago, a vendor which sells customized keepsake DVD's for runners had set up four digital cameras at various angles at the finish line to collect video of nearly every finisher. Computer programmers were able to write special software which allowed the visual images of runners finishing to be compared with results from the timing system. When a missing runner was found, his or her name was added to the finisher database by observing the elapsed time on the race clock over the finish line. Barahal said that he had high confidence that the display clock was correct because it was synchronized with the stopwatches of race officials at the start line.
The timing problem was initially discovered when runners came to Kapiolani Park the day after the race to pick-up their finisher certificates, a longstanding Honolulu Marathon tradition. As more and more runners received the news that they did not have a finish time, Barahal consulted his deputy in charge of handing out finisher shirts. About 23,000 finisher shirts were distributed, but the official results showed fewer than 21,000 finishers. The media quickly picked up the story, which made the front pages of Hawaii's newspapers.
Barahal immediately confronted the problem and sought to restore the race's reputation as the world's largest and best-managed tropical marathon. He feels with the verification of the missing finishers, he and his team have accomplished this.
"On the other hand, it's not fun to have your event dragged through the mud in the newspapers and have your event ridiculed," he said. "As a runner, you kind of learn to take your lumps. We knew we had the video."
The exhaustive analysis, which began with verifying the 27,827 runners who had registered for the event and working through the intermediate timing marks all the way to the finish line, found several interesting pieces of data:
=> 3,070 runners who were recorded by the timing system as finishing were not recognized as starters
=> 55% of finishers had inaccurate finish times (SAI time was more than two seconds off observed time)
=> Only 42% of finishers received all seven intermediate splits available on the course
=> 10,715 finishers had their times detected BEFORE they hit the finish line
=> 7,708 finishers had their times detected AFTER they hit the finish line
Barahal thinks that SAI's use of ultra high frequency (UHF) transponder tags was one of the reasons for the inaccurate and missed readings. "The characteristics of UHF are such that it is easily reflected," said Barahal. "Moisture affects UHF."
The race was contested in rainy conditions, which made timing the race more challenging than usual. However, last year's Boston Marathon was held in the midst of a severe rain storm and had no unusual timing difficulties. Boston used ChampionChip, the best known transponder timing system in active sports, the system which Honolulu had used previously. Barahal said he would move back to ChampionChip timing in 2008.
Despite the difficulties in Honolulu, SAI Timing & Tracking will be used to time tomorrow's City of Los Angeles Marathon, the company's first large timing job since Honolulu.
Know someone else who's interested in running and triathlon?
