When ADI engineers met Secretariat, a "proud horse of the generation"

When Secretariat won the 1973 Preakness, the official time was 1:55, about 2 seconds longer than the time recorded by a handheld timer (1:53.2).

I was a teenager in 1973 and was mowing a lady’s lawn when this race was on. She insisted I come in and watch the race. She grew up in Kentucky and her family was involved in horse racing, so she told me all the “behind the scenes” details and how if Secretariat won the Preakness it could be a Triple Crown. There was something different about the Preakness, I guess the track length and/or surface was different. I saw her cheering when Secretariat won, little did I know then that 39 years later I would correct that timing error.

My job was to review race tapes and determine what I thought were the correct times, and why. I had software that counted video fields. A frame is made up of two fields, so using fields gave me twice the time resolution, 1/59.94 Hz. My plan was to count the frames, then calculate the worst-case time error, multiply it together, and that was it. But it wasn't that simple. I also thought it would be easy to find equipment specs from 1973 (or earlier) and figure out the old NTSC standard. I wondered how accurate the 16.683 ms field rate was. I was able to find specs from that era, but not from 1973. However, after looking at the math for oscillator and thermal drift errors, I found that the error was extremely small. Counting frames was much harder than I expected. The cameras at the start and end were not perpendicular to the track, so I had to estimate. I also had to interpolate between fields. It turned out that these were the biggest sources of error. I came up with a time of 1:53.08, with a range from 1:53.00 to 1:53.15. I thought 1:55 was definitely incorrect.

During this process, I learned that horse racing does not start when the gates open, but rather when the race begins on the track. This short distance is called the "runout." It varies from track to track and can be up to 375 feet from the gates. At the Preakness, the runout is about 150 feet. I also learned that horse racing used to be timed to 1/5 of a second, or a resolution of 0.2 seconds (1:53 0/5 seconds). Today's tracks are timed to 1/100 of a second, or 10 milliseconds.

In my written testimony, I speculated on what might have caused the error. It's possible that the start lights triggered the timer too early. A bright flash (sunlight hitting the mirror or a camera flash) could have saturated the receiver, causing it to trigger falsely. After giving my testimony, a witness to the 1973 race told me that a piece of trash fell on the track just as the gates opened, and a man ran onto the track to pick it up. He left the track near the start lights. This is my speculation about what happened, but we will never know for sure.

This was the third hearing to try to correct the Secretariat’s times and track records. Penny Chenery (Secretary’s owner) had been working for years and was determined to get this corrected before she died. I never met Penny during this time, but after the times and records were corrected, she wrote me a very kind letter thanking me for my help. I would have liked to have met her and she seemed like a very interesting and unique person. I learned of her thoughts and concerns through Leonard Lusky, who was working meticulously to gather all the evidence she could. She did an excellent job getting the Maryland Racing Board to understand and change the Secretariat’s times. Penny died on September 16, 2017 at the age of 95. I was glad to be a part of the effort by all involved to get the truth cleared up before she died and give her peace of mind.