Updated on 09/15/2011 1:36PM

Figure patterns are inexact


COLUMBIA, Md. - "Three-and-out."

In recent days I've seen this expression used in print by more than one handicapper. And I've heard it repeated by a few players around my habitat at Laurel Park. Where did it come from? What does it mean?

The source is the well-reasoned chapter titled, Probabilities and Patterns, in the book "Beyer on Speed," first published in 1993. Andy Beyer coined the expression "three-and-out" to describe one very specific pattern of speed figures: the steady improvement of a horse for three races (for example, 55-63-72), followed by a subsequent decline. A study of 4,518 horses with this pattern revealed that 4 percent ran the same top figure next time out; 25 percent ran an even higher figure; 71 percent ran a lower figure; and a statistically impressive 52 percent ran more than six points lower.

Beyer clearly defined his three-and-out as "a sequence of three improving figures." But I'm afraid that its recent usage among some in the press and public has deviated from that original concept. Instead of three consecutive improving figures, some commentators have taken it to mean three consecutive big figures. There is a huge difference.

The past performances of Xtra Heat have been used, erroneously, to support this inaccurate version of "three-and-out." Here are her recent Beyer Figures (most recent first): 106-118-120-117-113-100. After Xtra Heat put together big Beyers of 117-120-118, the wrong-headed commentators foresaw a bounce, and based their prediction on what they called Beyer's theory of three-and-out.

In fact, however, the authentic three-and-out would have warned of a probable bounce after the 117 or the 120 - neither of which happened. But then a minimal decline from 120 to 118 cannot really be considered a bounce.

Clearly, Xtra Heat belongs with the 25 percent who ran higher figures even after improving over three and even four consecutive races. That's part of her great claim to fame this year.

All this raises the very complicated issue of speed figure patterns. It's a fascinating subject and, in my opinion, a useful handicapping tool. When I evaluate the past performances of any horse, one of the first things I look for is a clear pattern of recent Beyer Speed Figures. Here are some very basic categories:

* Repeat performers. These horses show a remarkable consistency in their recent Beyers - a string of four or six or eight consecutive figures within a few points of each other (84-81-85-82-84-82). Of course, the string has to stop sometime, so you can't just assume that a horse will repeat that same figure on any given day. But you can generally assume that these horses have very limited upside potential.

* Good figure/bad figure. These horses have recent races alternating high Beyer/much lower Beyer: 83-61-79-58. Once again, you can't just assume the pattern will continue, but it does give you a lead to follow, a framework to continue your analysis.

* Potential bouncers. After a big-figure, draining effort - especially after a layoff or an unusually big jump in Beyers - you have to consider a possible bounce. After one big effort, I'm very skeptical about a horse's chances. After back-to-back big efforts, I throw him out.

If he turns in a third big effort, well, I just shake my head and take my losses. After all, nothing is 100 percent in this game.

* Cyclers. Here's the proper place for the three-and-out phenomenon (or four-and-out, or five-and-out). When these horses peak and then decline, you can try to catch them when their Beyers begin to improve again. A typical pattern: 55-63-72-51-65-?? Many of these horses continue to improve until they again reach their previous highs. Of course, many of them don't. But for the significant number that do follow the cycling pattern you often get tremendous odds because the horse's best races are well-buried four or five lines deep in their past performances. Even if they don't win, these types can give you a high-priced filler in the second, third, or fourth slots in exactas, trifectas, and superfectas.

All of which leaves the largest category of all: None of the above. The random runners. In this category we find the great majority of horses, runners who have no clear pattern in their recent speed figures. They represent the often inexplicable pattern-less disorder in the form of many runners.

The search for figure patterns comes with a litany of warnings.

Many factors can intervene to affect a horse's form: trainer changes, pace situations, tough trips, track biases, long layoffs, severe class drops, different surfaces, Lasix, distance changes, and so on. So you need to know as much as possible about how each figure was earned before you can make real sense of any potential patterns.

In addition, there's no real explanation for why some horses run in these patterns. Some horses will be repeat-figure performers for six consecutive races in June and July, and then in October and November they will go through a perfect three-and-out cycle.

Same horse, different time, different pattern. So it's not something inherent in the horse's physical make-up - biorhythms or some such thing.

And these patterns are always temporary, limited to a particular short span of time - usually only a matter of weeks or months.

As with all other approaches to handicapping, there are no rigid rules in the use of figure patterns. There are only probabilities and percentages, and judgments that always have to be made