# Final Fractions Theory: How the Derby contenders finished in their final prep races

LOUISVILLE, KY (WAVE) - My Kentucky Derby handicapping starts with the finish.

The Final Fractions Theory is based on horses’ last eighth-mile and three-eighths of a mile in their final 1 1/8-mile Derby prep race, the idea being to look for horses who came home the final furlong in 13.0 seconds or faster and the last three furlongs in 38.0 seconds or faster.

Part of trying to predict who will wear the roses on May 4 is eliminating horses, and that’s the main value of the Final Fractions Theory. The premise is that the final fractions tell us who can or won’t handle the 1 1/4-mile Derby’s additional eighth-mile at this time in their lives. It also is a good indication if a horse is in good form or tailing off and won’t be at his best the first Saturday in May.

Twenty-six out of the 29 Kentucky Derby winners since 1990 met at least one of the standards and 22 met both, including Triple Crown winner Justify last year (12.70, 37.11 in the Santa Anita Derby). The three winners who failed to qualify on either threshold were Animal Kingdom in 2011; 50-1 winner Mine That Bird in 2009 and Silver Charm (barely over in 1997 after pressing a wickedly fast pace in the Santa Anita Derby).

Another two Kentucky Derby winners “qualified” on the final eighth-mile but not the final three-eighths (Charismatic in 1999 and Grindstone in 1996), and two others qualified only on the final three-eighths (Sea Hero in 1993 and Unbridled in 1990).

The final fractions are calculated for horses taking into account how far off the lead they were at the three-eighths pole, the eighth pole and the finish of their final prep. For years, I hand-calculated the times based on the Equibase charts or relied on the now-defunct Twitter feed @DerbyContenders. In the never-ending pursuit of fine-tuning a theory, this year I’m going with the computations produced by turf journalist, thoughtful handicapper and early FFT supporter Ron Flatter of Vegas Stats & Information Network (VSiN). Ron’s fractions are based on the digital tracking and timing system Trakus, where available, which was for every prep but the Arkansas, Louisiana and Sunland Derbys. While I previously have simply gone without the 1 3/16-mile UAE Derby, he used Trakus to compute an “about” 9 1/2-furlong time. He also used the chart and video review to come up with a time for Master Fencer’s win in Japan.

The FFT doesn’t care about a horse’s final time. So Cutting Humor’s 1 1/8-mile final time of 1:46.94 in the Sunland Derby is the same as Roadster’s 1:51.28 over Santa Anita’s deep surface, Tacitus’ 1:51.23 in the Wood and Vekoma’s 1:50.93 in Keeneland’s Toyota Blue Grass. What is important is that Cutting Humor’s final eighth-mile in 12.18 seconds is the fastest in the field and his three-eighths in 37.01 is topped only by the top three finishers in the Florida Derby, headed by front-running Maximum Security’s 35.96. Contrast that with Vekoma’s 13.45 and 39.39.

This year, nine horses possessing enough points to get into the Derby’s 20-horse field are FFT double-qualifiers. Among my FFT takeaways:

Arkansas Derby winner Omaha Beach is the goods. If betting today, my picks would be Omaha Beach (12.38, 37.45) and Louisiana Derby winner By My Standards (12.49, 37.79).

War of Will is a wonderful horse who was most impressive in taking the Fair Grounds’ Risen Star but then struggled home ninth in a dreadfully slow time (14.19, 40.09) after apparently straining a muscle when the ground seemed to break from under him shortly out of the start. He might win the Preakness, but history says horses who run that poorly in their final prep do not win the Kentucky Derby.

Throw out the Wood Memorial horses from the win slot, though the beautifully-bred victor Tacitus (13.30, 38.47) has a lot going for him. He’s exactly the kind of horse that sometimes makes it hard to keep the FFT faith. But why have a theory if you’re going to go against it?

Throw out the Blue Grass horses. Having said that, another interesting prospect for multi-horse wagers is Blue Grass runner-up Win Win Win, who had a terrible trip, getting stopped rounding out of the far turn that no doubt contributed to his final three-eighths in 38.93 seconds but who finished the last eighth-mile in 12.69 once he got free.

I’m skeptical of the Florida Derby horses, because the top three came home so fast after victorious Maximum Security got away with a dawdling early pace.

Santa Anita Derby winner Roadster and his runner-up stablemate Game Winner, last year’s juvenile champion, both meet the FFT standard for an eighth-mile but not for three-eighths of a mile. The final Santa Anita Derby time was extremely slow for that track, whose surface was made much deeper after a high-profile rash of catastrophic injuries.

No one alive is better at winning the Kentucky Derby than Bob Baffert. But slow track aside, my sense is that Game Winner didn’t progress from age 2 to 3 as much as some of his contemporaries. I think Roadster will take a lot of action, in which case I’d rather go for a horse like Omaha Beach, who meets both FFT criteria, as does Baffert’s Arkansas Derby runner-up Improbable.

Cutting Humor, who set a track record in the Sunland Derby, is going to have to show me he fits with these horses before I get suckered in again. I made the mistake of trying to get too cute chasing a price and picked 2017 Sunland Derby winner Hence after a similar impressive performance, only to see him finish 11th at Churchill Downs.

The Final Fractions Theory originated many years ago when veteran trainer Phil Thomas had a theory that Derby winners came home the final quarter-mile of their final prep in 25 seconds or faster. I couldn’t calculate quarter-mile times from the Equibase charts, but I could the final eighth and three-eighths going back to 1990 and saw there was indeed a pattern.

Keep in mind that FFT is just one tool for handicapping the Derby. Factors such as pace scenario, class, pedigree, the ability to handle different racing surfaces, a switch in racing surface, tractable running style, post position all come into play.

When Trakus isn’t used, FFT requires a leap of faith that the human chart-caller is accurate in the margins between horses at the three-eighths and eighth poles. Because we’re back to computing to the nearest hundredth of a second from the nearest-tenth used by @DerbyContenders, remaining unsettled is what to do with a figure such as Master Fencer’s final eighth in 13:02 in Japan. Does FFT mean you round down so it matches the 13.0 threshold? Or is it really 13.00 and he’s over the limit?