The first to know

[nobeyerspls]

Handicapping is an endeavor that greatly rewards those who become the first to recognize the winning ability of a horse who is otherwise overlooked in the betting. I read the thread on Bayes Models and view the theories discussed as being unable to assist with that task. The same is true of speed figures as they assign a nice number (as much as a fifty point improvement) after the fact.
When a trainer welcomes an unraced horse to his barn he has a puzzle that needs solving. We see the attempt to solve it in the past performances. Class, distance, surface, equipment and several other issues unfold over time and after twenty or so races a horse is fairly well defined with some moving through their conditions and finding their place in the class/distance/surface spectrum. It is well before this happens that an opportunity for a huge score appears.
For those who want to bet a little to win a lot, laboring in the vineyards of horses with fewer than five starts is the right place to be. You'll find angles and strategies, some of them counter intuitive, that pay off nicely because you'll be the first to know. The best question anyone will ever ask you is "how the hell did you find that horse?" Whatever your answer is it is unlikely to include statistical analysis or the use of speed figures.

[rokitman]

Astute analysis, No. As the often-bashed 46Z takes every opportunity to point out, the same thinking can be expanded to horses with with more than five races as long as what they are going to do now is "off-the-beaten-path" in some form, making what is in their running lines less predictive. Horses showing up for the first time in the oddball 7.5F turf distance, for example, as Z has already pontificated on.

[Overlay]

The best question anyone will ever ask you is "how the hell did you find that horse?" Whatever your answer is it is unlikely to include statistical analysis or the use of speed figures.

It depends on what statistical analysis is being used for. If it's being used to narrow a field down to one selection through a process of elimination, I would agree with you, as that process tends to discard the type of horses that would prompt such a question. By contrast, if it's being used to maintain visibility of the winning probability of every horse in a field, the answer to the question would be that, while the horse may not have been the single most likely race winner, it was the one with odds that compared most favorably to its actual chances.

[nobeyerspls]

Astute analysis, No. As the often-bashed 46Z takes every opportunity to point out, the same thinking can be expanded to horses with with more than five races as long as what they are going to do now is "off-the-beaten-path" in some form, making what is in their running lines less predictive. Horses showing up for the first time in the oddball 7.5F turf distance, for example, as Z has already pontificated on.

That's true. Horses trying a new surface or under a new trainer are in this category as are the late bloomers. I used the less-than-five starts because most of my mega-tickets came from those.

[1st time lasix]

By contrast, if it's being used to maintain visibility of the winning probability of every horse in a field, ..., while the horse may not have been the single most likely race winner, it was the one with odds that compared most favorably to its actual chances.[/QUOTE] Overlay is on to something here. IMO: Here is the essence of playing in a paramutual pool. Finding entries that will outrun their odds and at the same time detecting vulnerability in the entries that will most likely dissapoint their lofty staus as top contenders based on the proposition {odds}at hand. No matter how you recognize it. For those that want to bet small and win big.....they then have to structure exotic tickets either "horizontally" or "vertically" to capture this discrepancy. Nothing like "having" a ticket few others have in a paramutual pool.