Those of you who have seen our most recent State of the Program video saw the 55-race test I did.

Here were the conditions of the test.
Hypothesis
The AI Best Horse object might be good enough to win all by itself.

My Steps
1. I designed a simple strategy to use AI Best Horse.
2. That strategy had 4 simple rules, which are available to everyone on the video.
3. I scribbled the rules on an index card - which is shown in the video.


Pre-decision
A. The field is divided into Low & High odds, where high odds are natural odds and above (using the AI Line).
B. The AI Best Horse is pitted against the tote odds, producing a $net and a Grade.

The Rules:
1. All A+ and A horses are marked to be bet.
2. We must have an L (low odds) horse. If there isn't one, we must add one.
3. If there is MORE THAN ONE qualified "L" ...
or
4. If there is NO qualified "L" ...
... then we use The Neal approach to pick the L to bet.

Note: "The Neal" was the invention of @Neal Freedman - one of our Dream Team members. Simply put, it measures the difference between the book% of the TOTE and the AI LINE.

The Actual Test
1. I picked a random day using deTERMINATOR's random day function.
Aug. 22, 2024 was the day selected.
2. I started with the 1st race on the day and committed to do several hours of racing.
I rejected playing CNL and one other very small track - don't recall what it was.
3. I committed to playing EVERY RACE, provided it had at least 5 betting interests.
IOW, I was not permitted to skip any races.
4. All races would be $30 bets, using the built-in Dutching function.
Because of rounding, sometimes this resulted in betting as little as $29 and as much as $32.
5. All qualified horses would be bet.

Below are the results.


This is pretty amazing output.
____________________________
DUTCHING: TOTE ODDS represents the actual dutched betting for the 55 races.
Hit 54% of all races, with a $2.84 $net.
____________________________
FLAT BETTING (A, A+) represents betting $2 on each A or A+ horse. IOW, betting only the horses that computed as having a $2.20 or higher $Net.

IOW, these were the horses that were projected to be $2.20 $net or better.

Again, the results were pretty fantastic.
The hit rate was low at 10% but we were fishing mostly in deep water.
Notice the average pay of $25.74 and the $Net of $2.57.

Hitting back-to-back $50 horses didn't hurt, but the plays were quite obvious.
____________________________
FLAT BETTING (BAD BETS) represents all the horses bet that were not A or A+ HORSES (according to the rules provided).

IOW, these were the Projected LOW ODDS horses that were NOT supposed to be good bets.
____________________________