-
RESEARCH from HSH: aML 38% and aboveUNQUALIFIED BETS
(A new BALO!)
Race Prg Horse Win 2/08/24 GP 02 6 SGT. PEPPER 3.40 _________________________________________________ 2/08/24 GP 09 5 TATANKA _________________________________________________ 2/08/24 FG 06 6 LAUGHINALLTHEWAY _________________________________________________ 2/08/24 GP 04 1 PEP RALLY _________________________________________________ 2/08/24 MVR06 4 LIVING THE DREAM _________________________________________________ 2/08/24 MVR08 1 MOBIL MEMORY 2/08/24 MVR08 4 KAT TRAX _________________________________________________ 2/08/24 FG 09 6 DEER CROSSING 2/08/24 FG 09 7 UNIFIED REPORT _________________________________________________ 2/08/24 TP 02 9 RIBALDRY (IRE) _________________________________________________ 2/08/24 PEN02 2 CHARM ALARM 2/08/24 PEN02 4 GIRVINIZED _________________________________________________ 2/08/24 DED02 5 FROSTY BREW 6.00 _________________________________________________ 2/08/24 MVR04 1 ESTILO CLASICO 2/08/24 MVR04 3 SCRATCHOFF _________________________________________________ 2/08/24 PEN03 4 GIUSEPPE _________________________________________________ 2/08/24 DED03 2 MAGNOLIA SPEED _________________________________________________ 2/08/24 TP 04 8 TIGER PAW (IRE) _________________________________________________ 2/08/24 DED04 3 RAZZ 2/08/24 DED04 9 DEAL WITH HIM 3.80 _________________________________________________ 2/08/24 DED05 2 SERENA'S SERENADE 3.60 _________________________________________________ 2/08/24 TP 06 1 I'M SO NAUGHTY 2/08/24 TP 06 7 SHE'S SPLENDID _________________________________________________ 2/08/24 PEN06 1 CONTINENTAL DIVIDE _________________________________________________ 2/08/24 CT 05 6 STRATEGIC _________________________________________________ 2/08/24 DED07 5 FLOWER CAT 4.00 _________________________________________________ 2/08/24 CT 06 4 GRAN ANDREWS 4.80 2/08/24 CT 06 5 HANALEI'S HOUDINI _________________________________________________ 2/08/24 PEN08 1 LUCCA'S LEGACY 2/08/24 PEN08 4 BRIGHT PATH 4.80 _________________________________________________ 2/08/24 DED08 3 RUN FOR JOY _________________________________________________ 2/08/24 TP 09 7 NORBERT _________________________________________________ 7-34 30.40 $0.89
7 wins from 34 bets and a $net of $0.89 per $2 wager. -
RESEARCH from HSH: aML 38% and aboveYesterday, @ponyplayer and I tested what I called the KILLER object.
I tested and he watched.
Basically, we divided ALL AI Line horses under 2/1 into two groups:
Qualified & Not Qualified.
QUALIFIED BETS
Race Prg Horse Win 2/08/24 GP 01 5 MYSTIC SUNRISE 2.80 2/08/24 MVR01 2 FUEGO CALIENTE 2/08/24 GP 03 2 ANDIE ONE 2/08/24 MVR02 1 PRANCIPANTS 7.60 2/08/24 MVR03 8 INDY PRINCESS KOKO ______________________________________________________________________ 2/08/24 FG 01 1 RUN DREAMING 5.00 2/08/24 GP 05 4 SHARAR 2/08/24 FG 02 1 SCOOTERANDDABIGMAN 2.80 2/08/24 MVR05 6 HARSH TRUTH 2/08/24 GP 06 5 JOE DI BAGGIO 4-10 18.20 ______________________________________________________________________ 2/08/24 FG 03 5 HIGH TIDE 2/08/24 FG 04 3 STRAWBERRY SUNDAE 2/08/24 MVR07 3 SWEET UNIQUE 7.60 2/08/24 GP 08 6 PRETTY LIAM 2/08/24 FG 08 5 CARPIS 5-15 25.80 ______________________________________________________________________ 2/08/24 TP 01 8 STUNNEM 8.04 2/08/24 PEN01 5 FASHIONISTA 11.40 2/08/24 DED01 10 SLEW SKIES 2/08/24 FG 10 6 NERVY 3.20 2/08/24 TP 03 6 SAVANNAMARIE 8-20 48.44 ______________________________________________________________________ 2/08/24 CT 01 6 FIRST HOMESTEAD 2/08/24 PEN04 6 UNCLE DAR 4.00 2/08/24 CT 02 4 CRUZ IN THE CASTLE 6.20 2/08/24 TP 05 3 GATITA SUAVE 2/08/24 CT 03 5 KEEP ON SPURRIN 10-25 58.64 ______________________________________________________________________ 2/08/24 PEN05 1 BACK FORTY 4.60 2/08/24 CT 04 1 LATE NITE MISTRESS 2/08/24 DED06 1 ASTERN BANKER 2/08/24 PEN07 6 DOWSER 2/08/24 TP 07 5 UP AND DOWN 11-30 63.24 ______________________________________________________________________ 2/08/24 CT 07 4 DUNCAN IDAHO 2.60 2/08/24 CT 08 6 DE'S CASTLE 12-32 65.84 $2.06 ______________________________________________________________________
-
RESEARCH from HSH: aML 38% and aboveRank of 4th or worse...
Field Rank Starts Pays Pay% $Net IV PIV ExW PExW 158 rcPwr 4-9 317 91 28.7 $1.46 1.78 0.90 51.0 101.0 83 rFTB2L4 4-9 339 97 28.6 $1.49 1.81 0.92 53.7 104.9 157 rcFT 4-9 151 42 27.8 $1.52 1.72 0.90 24.5 46.5 95 rPWB2L3 4-9 503 155 30.8 $1.53 1.96 0.95 78.9 162.3 75 rSCB1E 4-9 862 274 31.8 $1.54 2.05 0.97 133.7 282.6 1655 rgLPHi 4-9 757 226 29.9 $1.54 1.91 0.95 118.5 239.1
-
January is Difficult - in case you haven't noticed...While the number of races are small, if it is a trend, I imagine that it will go back to normal in the near future.
This builds a case for some trend reporting. -
Hey Dave!!! <About Weight Loss>
Good to hear from you.
I happen to know that you've done well with this.
Tell the people how much weight you've lost. -
It’s 5:30 am…
Knowing that there are people who care is important to me.
Thank you for the kind words.
While I appreciate this, I want to make it clear that this post is about STOPPING THE BLEEDING far more than looking at how much blood there is.
_____________________
Most of us are of a similar age.
This is a time when we are at the greatest time of EXPERIENCE & KNOWLEDGE in our lives.
Sure, we have lots of battles and wars to fight... and the biggest of all is The Battle to Live.
But this should be a time of victories for us.
Whether it be success at the window, weight loss, strengthening self, general health or building wealth.
I say that it is time to get stronger. -
Class - How do you determine it and then use it?
Look at the new 25,000 series of factors.
https://www.community.horsestreet.com/discussion/comment/7266 -
Positioning vs Beaten Lengths
That was an interesting book.
The 5+ points count was a real stroke of smarts and it6 is deeply embedded in DET (as well as HSH).
I did not find the rest of the contender ideas to work well. I believe there are so factors in the field list that represent those. -
New NVIDIA supercomputer $3,000That's easy compared to picking winners in 2yr, old maiden races. — Andrew Kramer
Chart me. LOL -
Ques003 UNDERSTAND BALOs - Bet Against Low Odds HorsesDave, how do you handle a race in which the Green 4 pace object is the same as the red algorithm object? There is no discrepancy with the whales in this case. Does that Green 4 object still serve a role in determining whether a horse is BALO or not? — Jeff Capper
Good question.
First
The RED is not AI.
It simply means that the whales do not agree with the AI's profile.
Second
Since the advent of The Neal, there are several better ways to find BALOs.
Third
There are also two types of BALOs:
1. In the Handicapping process (i.e. Race Window)
2. In the Final Step.
These could also be called:
1. Handicapping-Based
2. Value-Based
IOW, if there is a low-odds horse who you have determined is a BAD BET, he is, in effect, an AFTER HANDICAPPING BALO. Those are generally far better (in my experience). -
Time DecayIt is a difficult formula to find.
I had to design my own.
It's based upon a 120-day half-life. -
What the models were trained on?Since there is no database there cannot be any testing unless it is done manually.
-
What the models were trained on?
The challenge with the approach you have outlined has many challenges.
_____________________
1. Since all the handicapping is dynamic, there is no mechanism that would allow automated testing.
Thus, to test against (say) 5,000 races, SOMEONE would have to literally bang those 5,000 races in manually.
In addition, if you DID pound in (say) 5,000 races, you'd logically want to do it live. Those 55 races took me about 6 hours with the tracking. That would put a 5k test at around 550 hours.
_____________________
2. Because I cannot have the data being exported and, therefore, easily available to the whale world, there can never be an automated testing process.
_____________________
3. Because the whales are the great influencer of our time, and they retrain their models annually, this new testing will need to be done once a year.
_____________________
FWIW...
In my 3+ decades in the horse betting business, I have found that the typical handicapper has no problem using the concepts and factors that he THINKS are important, and will use those without hesitation, despite the fact that he has done no real vetting.
He will consider a single pretty good day as enough proof to start using his new idea.
I suggest that the only way forward with The deTERMINATOR is to adopt a similar belief system.
Remember, I never said that AI BEST HORSE is the answer that will make you a professional. You will probably need to add your own secret sauce to make it work the way you want it to. (Truth be told: I already have created my own, derived BEFORE I did the 55 races.)
I really don't want 200 guys doing the same thing.
In fact, I'd make that a requirement if I could.
(On that particular topic, I have a practical answer.) -
Help with making specific kinds of betswhat if I don't want to Dutch the horses, play only 3 even if 4 are possibilities, etc. — Andrew Kramer
Perhaps share your questions here?
And, did you see this thread?
About That 55-Consecutive-Race Test -
What the models were trained on?As a follow up to a question I asked Dave during last week's meeting, can we get a general sense about what data the AI models are trained on? Is it a fair assumption that they were fed all of the 3000+ factors from HSH during training? Similarly, were they then evaluated against out-of-sample test data? — Atakante
A GENERAL answer would be: It started with one year of everything, plus multiple years for some of the small sample types.
As for what factors were permitted, the answer is all but a handful that had some known issues.
I do not understand what "evaluated against out-of-sample test data" means. If you mean was the AI TESTED, the answer is no.
I ask b/c it feels right to look deeper into how AI predictions fare in real life with an eye towards their performance based on factors/variables the models were NOT trained on. Given tens of thousands of pace lines the models are using to make inference, it seems to me an uphill battle trying to prove them right/wrong with even couple hundred manually tracked races for those variables they already considered during training. — Atakante
You are correct.
There is zero proof.
If memory serves, the chaos race variable/factor was NOT an input to the models so that's one candidate factor worthwhile manually diving into. Is it a safe assumption that race-level factors were excluded from training data? If not, is there a short list of other candidates or a shorthand logic to find others to manually "study"? — Atakante
Chaos = correct, but there was a fitness function involved.
Race Level: Nothing was left out.
___________________
The questions you have asked are actually pushing the envelope of need-to-know.
The AI is a very strong engine.
After my recent small sample test that was a day chosen at random, I am actually surprised that it performed so well.
I simply threw together a logical "system" based upon what has been discussed and it crushed the game. After going back and correcting a handful of races I had tagged, and correcting a few mistakes, manually assembled these statistics.
This is pretty amazing output.
DUTCHING: TOTE ODDS represents the actual dutched betting for the 55 races.
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.
FLAT BETTING (BAD BETS) represents all the horses bet that were not A or A+ HORSES (according to the rules provided).
Dave Schwartz
Start FollowingSend a Message