08/04/2011 11:33AM

How speed figures evolved into a tool for insiders

Barbara D. Livingston
Upperline, trained by Mike Stidham, won the Bayou Handicap in March with a 102 Beyer, 13 points higher than her previous best. Stidham said the career-high figure was a sign that Upperline would need time to recover. The filly has earned figures of 87, 90, and 89 in her three races since.

Several years ago a former Arlington Park racing secretary groused about a trainer’s decision not to enter a nominated horse in a stakes race. To the racing secretary, the horse in question at least had an outside chance to win. Why not take a shot? But the trainer had matched up his horse’s speed figures with the highest numbers earned by potential starters in the stakes and determined it would be folly to run.

“These guys and their numbers,” grumbled the racing secretary, who also dealt with East Coast trainers while in Florida. “Bill Mott, he’s in the Hall of Fame. Bill Mott doesn’t need to look at the figures to tell him when to enter a horse.”

That was then. These days, even Bill Mott, who hails from a decidedly old-school branch of the Thoroughbred training tree, is not beyond seeing what the figures might tell him about his stock.

“I’m probably not as dependent on them as some people,” Mott said. “But I’m always interested to see them, see the patterns, to see what a horse is doing.”

Even 30 years ago, speed figures were the arcane province of a few gamblers. Most bettors wagered using the vague notion of class: A stakes horse beats an allowance horse, an allowance horse a claimer, a mid-priced claimer a low-level claimer. The wagering public followed the lead of the sport’s cognoscenti, which disdained numbers as basic as final time. “Time only matters in jail,” some trainers intoned even into the 21st century. Insiders prized subjective analysis of a Thoroughbred over the mathematics that produce figures. The experienced horseman − flinty-eyed and savvy − could determine a horse’s quality.

DICK JERARDI: Final times can mislead - just look at Haskell and Jim Dandy »

But things have changed in racing, just as they have changed in other sports. The sabermetrics movement, defined by objective analysis, especially statistics, shunned the insider perspective in baseball and has altered the way the sport is run. In professional basketball, “stat geeks” use mathematical formulas in an attempt to assess a player’s contribution in more meaningful ways than a scoring average or a scout’s eye-test. Many teams have bought in, and most have advanced stats experts on payroll.

Over the years, a similar shift has occurred in racing. Class still counts, and horsemen still train horses, but speed figures have wormed their way into every nook of the racing world.

“I am shocked, delighted, and gratified at how they’ve expanded,” said Andy Beyer, whose Beyer Speed Figures, as much as anything, triggered racing’s figure revolution. “I think younger racing fans would scarcely believe how the importance of time – forget even the concept of speed figures − is missed for such a long time in racing and by such a large segment of the racing population. When I started out in racing, if you look at any literature from the time, the orthodoxy in racing was that time did not matter.”

And in a way, that was true. Racing conditions being so different from day to day at a track and even more divergent from venue to venue, comparisons of final times seemed worthless.

Speed figures attempt to address this problem, using raw time as a base, but translating that time through equations to account for differences in venue, in track surface or configuration. First, par (or average) times are set for each class level at a given racetrack. Each racing day, actual times are compared to these established pars to produce a track variant, which relates the speed of a surface on a given day with the average speed of the surface. Those findings then are applied race by race and plugged into equations that generate speed figures for individual horses on a card.

The idea is to estimate how fast a horse will run in a neutral environment. Baseball statisticians attempt a similar translation through “park factors,” which estimate the impact of a stadium on a player’s performance.

Subjective judgments enter the figure-making process for racing. A surface can change during a program, and a figure-maker must decide how to adjust the variant accordingly. A race may produce a figure that seems impossibly fast or slow. A figure-maker might move the number up or down to accord more fluidly with a horse’s established patterns. Sometimes a variant might need to be recalculated after a provisional figure is assigned. Recently, the filly Inglorious initially received a Beyer figure of 90 while setting a track record winning the Queen’s Plate at Woodbine, but her number from the race later became a 96.

Tweaks and revisions are part of the process − craft as well as math − but even roughed-out numbers are meant to tell a truer tale than the abstract class system. Beyer started making his figures in the early 1970s, learning from a classmate at Harvard, and he said no one else in his early racing circles used numbers to bet races. In New York, Len Ragozin was making another sort of speed figure, which came to be called The Sheets. Starting in the 1970s, horseplayers could buy The Sheets, but doing so required an interview and a promise to pay a percentage of handle to acquire the figures.

Few bettors were playing the races based on figures, and during these glory years, an astute figure-player could make a killing. The numbers might point out a horse stepping up from $10,000 claimers to a $25,000 race as sufficiently fast. The class-based approach would dismiss the horse as inadequate, producing immense value for the figure bettors.

This was the sort of system inefficiency the earliest purveyors of sabermetrics in baseball exploited when making personnel decisions. It wasn’t that the things most other teams and bettors esteemed were meaningless, but often, something the masses weren’t seeing had greater importance.

“Those are ancient times, when you could get monster prices on horses with outstanding figures,” Beyer said.

MORE: Team Tizway built on patience and trust | Jockeys Smith, Sutherland discuss Battle of the Exes

DRF WEEKEND: Barbara Livingston's Saratoga Scenes | Q&A with Jerry HollendorferHandicapping roundups

In 1992, Beyer licensed his figure methodology exclusively to Daily Racing Form, which began publishing speed figures in the running lines of every horse, and thus an enterprise existing on the margins of racing began to go mainstream.

“Even people who were totally oblivious to speed figures couldn’t help seeing that big bold-faced number,” Beyer said.

Within a few years the average gambler had factored speed figures into the handicapping process. The Sheets grew ever more popular, the Beyers were bold-printed in the Form, and “figure” horses began massively moving money.

“When you’d look up at the odds boards, they’d have a tremendous opening effect on the betting,” said trainer Todd Pletcher, who first became familiar with figures while still working as an assistant for trainer Wayne Lukas in the early 1990s.

“In the beginning, you could get amazing-priced horses, and those guys were making a killing back then,” Mike Stidham said. “That’s why, for me, the numbers are more valuable now for placing horses.”

Stidham was the trainer the former Arlington racing secretary disdained for not entering his stakes-nominated horse because of what the figures told him. Stidham trained a string in California when he fell under the influence of the Ragozin sheets nearly 30 years ago.

“They really came in handy for me with horses that you would nominate to three or four different stakes and figure out where you fit best,” Stidham said. “So many times in these stakes, you have horses coming from New York, the Midwest, the West Coast. In the old days you’d see horses running at Santa Anita running six furlongs in 1:08 flat, and then they’re coming from the East Coast running 1:11. That 1:11 might actually have been the faster horse.”

More people than ever believe in the numbers. When Drosselmeyer won last year’s Belmont Stakes, one of the most prestigious races in the country, he earned a pedestrian Beyer of 94. An earlier generation might have pegged Drosselmeyer as a potential star when he made it back to the races this year following a long layoff. These days? Drosselmeyer was a tepid 2-1 favorite in his comeback, a $60,000 stakes race at Tampa Bay Downs in which he finished fourth.

Stidham has a stakes filly named Upperline who earned a career-best 102 Beyer winning the Bayou Handicap this spring at Fair Grounds. The number was 13 points higher than Upperline’s previous best, and Stidham figured much time would be required for Upperline to recover from the performance. Indeed, Upperline finished ninth and fourth in her next two starts, and even when she won her most recent race, she ran nowhere near as fast as she did in New Orleans. But because Upperline achieved that big number, her connections believe she is capable of delivering a performance that could win a major race. Upperline is under consideration for the Grade 1 Beverly D. Stakes, which probably would not be the case if her trainer didn’t believe the numbers.

“Without a doubt, you can use numbers to guide training,” Stidham said. “Any time a horse makes a big move forward in a short period of time, you’ll almost always see the horse take forever to get back to that kind of number. As a trainer, you can find that hard to believe. You think, ‘My horse is healthy. She’s doing good, she’s training great. Why won’t she run that number back?’ But you see it all the time. Most of the time you’re just wasting your time if you try to hurry back.”

And there lies the reason it has taken longer for speed figures, an advanced form of statistics, to penetrate racing’s insider ranks after gaining acceptance with the public. Insiders, like trainers, can be blinded by their apparent expertise. They know what makes horses tick, and they know their horses and the endless intricacies of the game better than anyone. All this may be true, but that does not mean a different perspective, one formed by people who might not know a fetlock from a foretop, can’t help guide racehorse management and training.

“I think it’s a great tool to gauge your horse’s performance,” said Pletcher, who has been doing just that for the last 10 years.

The flow of speed figures, from racing’s outsiders to insiders, is exactly the course the advanced-stats movement took in baseball. Bill James, a pioneer in the sabermetric movement, started writing about novel ways of viewing the sport in the early 1980s, but it would take 20 years for his ideas to fully lodge in brains that were making personnel decisions.

“Within the industry, in order to be in the industry either you were a player yourself, or you worked your way up, but you were indoctrinated, and when everyone around you is focused on scouting or whatever, it takes a unique kind of stubbornness to say there’s another way of looking at the game,” said Rany Jazayerli, who in 1996 co-founded Baseball Prospectus, an influential sabermetrics-based publication. 

From the beginning, sabermetrics questioned traditional ways of measuring skill. A pitcher winning 20 games was revered, but how much of his performance might have been due to unusually good luck, or the fact he pitched from an abnormally high mound in a stadium with cavernous dimensions? If such circumstances could be factored into stats, the way speed figures accounted for the way a raw time was achieved, they should be more meaningful than the old measures. But baseball’s power structure, grizzled scouts and former players, was more likely to say, “What do these people know? They didn’t play the game!”

“We weren’t writing to teams,” Jazayerli said. “We were writing to fans we thought would be more receptive.”

Then, similar to the Beyer figures appearing in Daily Racing Form, a sabermetrics-focused writer named Rob Neyer began working for ESPN in 1996. Now the movement was reaching more eyes than ever before. Around the same time, the general manger of the Oakland Athletics, Billy Beane, began hunting for players who looked good on a computer, not necessarily in uniform. One of the Baseball Prospectus writers, Keith Law, was hired to work as a consultant for the Toronto Blue Jays in 2001. Now almost all major league teams employ an advanced-stats guy, and even the rare holdout, such as the Atlanta Braves, are at some level working off advanced-stats principles.

“They’ve permeated the sport so much, you don’t have to be a stats person to know them,” Jazayerli said.

That’s true for racing, too. Even horsemen one wouldn’t expect to be tuned into numbers, like Mott or Christophe Clement, have them in their head. Clement has enough faith in the figures that he won’t try a stakes race if three or four expected runners look better than his horse. And while someone like Mott isn’t using speed figures as a training guide the way, say, a true believer like Bobby Frankel would have done, he trusts the figures to color what he believes he knows about his horses.

“Sometimes I’ve been a little bit surprised at the contrast in what you see and what the numbers are,” Mott said. “There have been times like that.”

Mott also can speak to the current pervasiveness of speed figures within racing. “I know when I have a horse run fast, because the next day or two, I have someone calling and trying to buy the horse. I check the speed figure, and they got a really good number.”

Indeed, speed figures tend to set the price for the private purchase of horses off the racetrack. Stallion registries advertise the best speed figure a horse earned during his racing career. And people are even trying to project what kind of a number might be generated from the mating of a horse that has run a figure of X with another who has achieved Y.

“That’s interesting to me, how people are looking at [figures] from a potential purchasing standpoint in yearlings and stuff,” Pletcher said. “Now they’re looking at it generationally.”

Trainers’ faith in figures can hold field size down in stakes races, a fact that probably motivated the former Arlington racing secretary’s angst.

“They’re part of the reason the field sizes are smaller,” Mott said. “Guys will enter a horse in the race, and if they don’t look good enough from the figures, they’ll scratch.”

But all this is not to say that horsemen across the land spend more time dissecting figure patterns than making out training charts and watching horses jog across pavement to judge soundness. Perspectives have shifted in the speed-figure era, but the practice of racing horses has not fallen fully under the sway of numbers.

“I think you have to be a little bit careful that you don’t let the numbers control you,” Pletcher said. “I think some people look at the number and that’s it, but there are a lot of other factors involved.”

This, too, is the way the stats revolution has shaped sports. Bob Bellotti (www.bellottibasketball.com), a pro basketball stats consultant who started creating advanced NBA statistics in 1988, said the vocal skeptics who yell at stat-heads to “watch the game!” or in the case of racing, “watch the race!” aren’t misguided.

“Statistics are a tool that can be valuable in assessing players and teams, but I have found that you learn a lot more by analyzing the stats and watching the games,” Bellotti said in an e-mail that could easily pertain to racing. “The two go hand in hand. By watching the games closely, you then can develop questions about a particular team or player’s patterns that you otherwise might not have picked up on if you were only looking at the numbers. Using the stats, you can then more reliably answer those questions.”

And while the figures don’t lie, at least in the eye of the numbers faithful, the horse with the best figures doesn’t win every race. And now that figures are everywhere, the wise-guy ploy is to find the overlooked horse who doesn’t boast the obviously superior numbers. Unearthing such a creature can give special satisfaction even to someone, like Clement, who sometimes leans on the numbers.

“That’s very satisfying to me,” Clement said with more than a trace of mischief in his voice, “to beat the numbers. To use horsemanship and what you see to beat the math.”

Putting speed figs to work

Three examples of racing decisions that have been influenced by speed figures:

1.  Victory Gallop
Purchased on the advice of speed-figure expert Jerry Brown at the end of his 2-year-old season in 1997, Victory Gallop went on to win the Rebel, Arkansas Derby, and Belmont Stakes at 3 and the Foster and Whitney at 4.

2.  Drosselmeyer
After upsetting the 2010 Belmont Stakes with an unimpressive 94 Beyer Speed Figure, bettors made him just a 2-1 favorite in his comeback race, the Challenger Stakes at Tampa, in which he finished fourth.

3.  Upperline
Upperline won the Bayou Handicap in March with a 102 Beyer, 13 points higher than her previous best. Trainer Mike Stidham said the career-high figure was a sign Upperline would need time to recover; the filly has earned figures of 87, 90, and 89 in her three races since.