This is the right one!

Wouldn't that severely limit your past history data to a very small sample size? — Tim Romero

Yes.
Which would you prefer... LESS and RIGHT or MORE but WRONG?

LOL - A well-known quandary for us data guys.

Is anyone around who can share about the 90% Confidence Interval stuff?

I added the ability to use different tie break factors for FTS and PrevStarters.

Get v1.1191.

We'll be demoing this one tonight.

In our system, we have something called "The Ask."

Just imagine that only the G and B races should really count.

What is an ELO system?? — ponyplayer

A relative rating system - think Chess Federation ratings.
(i.e. who you beat or lose to matters)

I have been working on an ELO based system for horse racing and wondering if anyone has done something similar and what was your experience with the system. I get a decent return but not fully profitable based on betting only one horse per race. What are others seeing? — Tim Romero

What a great first post!

Welcome.

I think the biggest challenge you face is that racing is far different than (say) chess because (almost) every chess player in (almost) every game is TRULY TRYING.

There needs to be a mechanism for lightening (or removing completely) races where a given horse was unsuited for some/any reason - including being off form.

This makes it difficult to maintain.

Not that many L's since most of the low odds horses qualify as d's and it's a narrow window. So the pool size for L's is low as expected.
But, as you can see, 44% winners with an AI Line of <=1.99.
56% winners with an AI Line of >=2.00, but there's a lot of horses in these pools.

BALO's: AI Assist set to Value Method: 0 and Hit Rate Importance: 100.
The d and/or L horses with NO Rank 1's. Found 5 of these. The d and/or L horses with ONLY one (1) Rank 1. Found 3 of these. None of these 8 horses won. Have seen a few more of these since testing and none of them won. Also, in short 5-6 horse fields, as expected, these d's and L's finish 2nd quite often. — Jim Pommier

This deserves a BIG WOW.

Guys, I am ill.
Doubt that I will be up to a workshop in the morning.

The pre-release of DET v2-The AI Assistant is in your DropBox now.

I've reached out to Brian and I assume he'll schedule a workshop for the a.m. tomorrow.

A little more...

You can begin with whatever you are doing now.

THEN, before you pull the trigger you can (essentially) ASK who of YOUR horses the AI ASSISTANT likes.

And THE ANSWER can be based upon WHATEVER YOU WANT...
*HIT RATE
*VALUE
or (literally)
*MIX of hit rate and value.

You can set the mix!

You can even do FTS separately.
Or look at low odds horses and leave out high odds horses.
Or concentrate on Pace-E vs Pace-LP and sort out the stretch run.

Brian's right:
It IS mind blowing.

Nice post, Brian.

Update on Mother’s Day morning.

Brian and Jimmy got their upgrade yesterday, and will get another one sometime today.

I expect there will be one more upgrade after that (likely by Monday evening).

At that point, they will be testing for small bugs as we ready for release.

I can't wait to see it and try it out!! — Jim Pommier

You being the other driving force around here, you'll have it just as soon as I get past a cosmetic sorting bug.

I expect that everyone will have it within about 10 days or so.

Yes, it's looking pretty amazing.

Not ready to be shown yet, and not what I demo'd a couple of weeks ago.

The idea actually originated from a conversation I had with @ponyplayer when we were discussing the idea of having a GREAT HANDICAPPER at your side at the racetrack.

I described the scenario as: "Imagine that you're at the track and Tom Brohamer is sitting next to you and he'll give you his picks in every race."

I asked the question, "Which would you prefer?"
1) Tom gives you his top 3 BEFORE you handicap?
2) Tom gives you his top 3 AFTER/i] you handicap?

If you choose #1 he gives you your contenders and you have to separate them for your final selection.
While if you choose #2, you pick your contenders and Tome provides the separation.

Of course, there is a 3rd option - just bet his best horse or two.

Different people might choose differently, but I like option #2.

THIS IS THE PURPOSE OF THE AI ASSISTANTS.

After a couple of decades involved in gaming in Nevada, I can tell you stories.

From my days of sitting box, I can also tell you the Las Vegas standard procedures did not allow for either shutting the dice or stacking the dice.

In my time in Las Vegas I met three very serious crap players who actually had full size tables in their homes to practice, throwing dice. I had one of them show me how – – his words – he set the dice without really breaking the rules.

What he did was, he took one die and turned it in a given direction while holding the second die and manipulating it in his hand so that when he picked up the other guy from the table, it would be in the position that he wanted it.

He desperately wanted someone to believe that his approach worked. After three hours or so of me dealing to him while he shot the dice I can tell you it was an abject failure.

It’s not my place to say something works or it doesn’t. However, in my experience between those three men and the many that I saw in the casinos, I can only say I never saw one that won, but they all seemed to believe that they won.

Here's a filter you can just copy & paste. Make sure you get everything between the lines.
___________________________________________

,FilterCopy:::,KY Derby , |SPCL| 0.0099.000 0.00 0.00 099990500050000 0999999 0999999 010010010001400 , ,0,0,0, ,0,0,0,0,999999,1,0,84009,9999,1, ,0,0,0,0,0,1,0, ,0,0,0,0,1,0,0,0,0,0,0,0, ,0,9 ,0, , ,CD ,0,99, ,1,0,0,0,0,1,1,0,1,1, ,10,10,B,99,99, ,1,1,1, ,0,0,0,0,1,0,0,0,0,0, ,1,1, ,0,1,0,0,1,1, ,0,9999999,9999999,9999999, ,2000000,9999999,9999999,9999999, ,1,1,1,1,1,1, ,0,1, ,0,99,99,99, ,0,99,99,99, ,0,99,99,99,0,99,99,99, ,0,32,0,8,0,8,0,8,0,8, ,4,16,1,4,1,4,1,4,1,4, ,0,200,200,200, ,0,200,200,200,0,0,200,200,200,0,0,200,200,200,0,8,0,8,0,8,0,8,0,8,0,8,0,200,0,200,1,1,1,1,99,200,99,200,0,200,0,999.99,0,99999,0,99999,0,99999,0,99999,0,99999,3,3,11,0,99,0,99,0,99,0,99,0,99,99,99,99,99,99,0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,

___________________________________________

Filter on:
Circ=9
CD
10 fur
May
Sat only
GR=1
Purse>2,000,000 (?? Guessing what it was a decade ago)

It does if you look only at Derby races.
Weren't you doing that the other day?

I don’t know if Derby winners not training on is actually happening more than in the past. But if it’s true, your belief that it’s due to being them lightly-raced makes sense to me. — PickMaster

With data, it's easy to test. You can use the CAREER STARTS field.

How do we overcome this???


Mayhemily
@EmilyOptixEQ
21m
the number of horses (16) in this #kyderby field (21) with gate issues in one of their past 3 runlines is unique and compared to previous years, which is extremely high. — ponyplayer

How do we overcome what?

Oops.
Faulty memory.