The most infamous incident at Delta was the "fog jockey" incident. — RanchWest

I remember that!

Another set of stats that might be ripe is workouts. Is there a difference between when the horses win and lose? Does the trainer work the horses long or short? Any difference for distances or surfaces? Fast or slow? Frequency? Does the typical layoff after a race disclose anything? Does the trainer have access to a private training facility? — RanchWest

I love stats that nobody else has.

Personally, I like to concentrate of performance stats.
That is, things like: "How does this trainer perform with horses who..."
Rank 1st for speed rating in their last race?
(or 2nd,3rd,FrontHalf,RearHalf)

• Best of Last 2 Races?
• Best of Last 3 Races?
• Best 2 of Last 3 Races?

That worked.

You did not share what you think you shared.
Download it yourself and see.

Theoretically, yes.
The standard approach for a whale group is to lose 3.5% and make it up on rebates.
I'm not suggesting that his approach provides enough handle for whaledom, but it should work providing it is profitable and remains such.

Because his selections (logically) are low-odds...

• He can bet a lot of money if he chooses to ramp up.
• It will take only 1,000 bets or so to verify that the system is valid.
• The next statistical challenge would be some sort of seasonality.
  (i.e. the sample is from RECENT history. Will it work in August?)

If I had such a system in my arsenal, I would use a mild chase-the-money strategy to level out the fluctuations and simply grind out a profit.
This assumes I actually desired to be a 4-day-a-week player - i.e. have a job.

Note that I suggested beginning with a premise and then testing that premise. That avoids, in part, the backfitting issue. Of course, as you suggested, you still have to forward test because it can still regress. — RanchWest

Of course. But you guys aren't normal players. LOL

@RanchWest & @Steven but you are both smart guys, who knew all of that above.

IMHO, collaboration between guys like you two really makes a difference. Not only can you spread the work, but you can also share ideas. And the more diverse the collaborators, the more powerful the potential outcome.

@Steven & @RanchWest,

It's all in what one can prove - to one degree or another.

Meaning:
Spot plays are generally built on a backfit.
Most guys look at a database and build a set of rules that worked in the past.
These rules worked because they were contrived based upon history.

Math Guy says: "That's when the testing needs to begin."
Handicapper says: "Huh? But I already proved that it works."
Math Guy says: "No, you proved it WORKED IN THE PAST. Now you must test against a new sample going forward."
Handicapper says: "But I don't have another sample."
Math Guy says: "Then you CAN'T PROVE anything."
Handicapper says: "Oh."

Later...
Handicapper says: "Why didn't it work?"

I think this is anecdotal but I've always heard that on any given day, 95% of the people at the track lose — Steven

I think that more people have winning days - maybe as high as 15% of REGULAR PLAYERS. It very much depends upon individual strategies.

But this is all anecdotal. I could be far off on all of it.

And yes, back to the subject of this thread, they are two different concepts. I think this is anecdotal but I've always heard that on any given day, 95% of the people at the track lose and of that remaining 5%, about 3% (or maybe even 4%) are only winners for that day, the remaining 2% (or 1%) are the only long term winners. So to beat the game, you have to be in the top 1% - 2% of the people there at the track. — Steven

My estimate is 4 out of every thousand.

I agree Dave, totally different concepts. But I think that whatever concept you choose it has to fit your personality. — Tony Kofalt

That is a huge piece of understanding.

I know that almost everybody says that they're "All about winning," but when questioned, they usually admit that they'd rather keep using their current approach because it's fun even if it means they lose long term.

And that is completely fine.
I know a lot of guys who really enjoy their personal approach to analysis, and the challenge that produces big pleasure on good days.

I'm just cut from a different cloth. For me, the challenge is winning forever. LOL

And it didn’t match the official result?

I would contend that that is because there are near 40% winning favorites. The problem is that if you are using a line as a selection, you won't get many longshots.[/b. — RanchWest

Exactly right.

And that is reality.

Meaning, the "best horse" in the race SHOULD BE THE FAVORITE most of the time.

An odds line is an odds predictor, not a performance forecast. — RanchWest

I would contend that they are directly related.
Years ago, two of our players - a data geek & a programmer - banded together to form a partnership.

The concept was that they would build a model entirely based upon HOW THE PUBLIC BETS.

The idea was that they'd build this model, tweak/skew the hit rates towards the low odds horses (which represents reality) and produce an accurate win probability for each horse.

At the time I did not see the value in it. (I was wrong.)

Those two became a PROFESSIONAL PARTNERSHIP, but it turned out to be short-lived as the data geek was a guy with a real gambling problem. The "programmer" mortgaged his house to take a year off to play and never looked back. (As God is my witness, this is a 100% true story.)

Sadly, the programmer died very young from throat cancer.

[Special note for very long-time HSH users who might recall the player: His initials were "TD."]

@No Moore

Did you play TUP yesterday?

However, it's not directly the distance that is the reason for variable length on speed figures, it's the variable velocity caused by the distance. — Dustin Korth

Really an excellent answer. :clap: :clap:

I avoided wading into this thread because I KNEW someone would say it.

Wow Dave- you achieved a $net of $.89 on these favs. To me that is amazing!! Great job — Tony Kofalt

That changed last year.

A lot changed in 2021.

BTW, the changes can be very abrupt.
They're driven by the fact that the whales retool every year and bring out their new stuff in January.

I would prefer you didn't publish results that cannot be validated (especially for a small sample). Otherwise, it's just like the babble on Fox or the after the fact analysts on CNBC. I won't even mention the other fake news network. Sorry! — Biniak

You can prefer what ever you want, but I doubt you will ever get your preferences fulfilled.

If all the conversation on a forum is only for VALIDATED RESULTS then the bar is just too high and there will be zero new ideas put forth.

If I publish anything, it would be based upon SOME STUDY, but it can never be enough to call it rigorous, otherwise, the cost will be so high as to be unaffordable.

@JRand,
You CAN message me, of course.

But you CAN'T get the answer in return. :grin:

I'd like to get on that list as well. — JRand

Sorry, but no.

Read this thread and you will understand.