Saratoga & Del Mar · 2026 Meets · Model v0 · the anti-tout
Transparent model race-cards

Del Mar — Fri Jul 17, 2026

4 value plays today · 6 races we're passing
Today at Del MarMostly Sunny, 76°F, wind 10 to 15 mph · rain 3% — no off-track flagNWS forecast
How this card is made & how to verify it

Edge Card v0 is a transparent heuristic, not a trained model: Prime-Power softmax blended 50/50 with the de-vigged morning line, plus a published rules-based pace adjustment. Constants on the methodology page. Structured analysis, not proven edge; tracked publicly either way.

Model v0-heuristic-1.2 · cfg b00cb21f · code 958a2d2e · takeout (DMR) win 15.43% / exacta 22.68% · card SHA-256 910012f7fdcb… · Scratches as of 12:24 PM ET. Late scratches void at scoring; the card is never edited.

Powered by Brisnet data
Flags 4 of 10 races
New here? How to read a flag
Model
Our estimated win chance for the horse.
Fair
The odds that chance implies — the price where a bet breaks even.
Board
The morning line, the track's early guess at the odds.
Play at X or better
Only bet if you can get that price or longer. The edge is in the price.
Pass
Most races. Our number agrees with the board, so there's nothing to bet.

Odds note: 7/2 is the same as +350.

Race 2 · 1 1/16 miles · turf · OClm 80000n1xFlag
3VICTORIOUS DREAM
Model
Fair
Board
24%
3/1
15/2
our win %
break-even
the morning line
We flag it because the board (15/2) pays more than our fair price (3/1). Same horse, longer price.
The market
12%
Edge Card
24% +12 pts
▸ PLAY AT 7/2 OR BETTER
Ranks #1 in our power figure blend. Even pace
Data · Brisnet
The model’s read ▾

The number. Our model makes VICTORIOUS DREAM a 24% chance to win; the morning line, once you strip out the track’s take, implies about 12%. That 12-point gap is the flag — we price the horse at a fair 3/1 and only bet if the board is still paying 7/2 or better at post.

Why we rate it. It tops this race on our power-rating blend — the strongest runner on the figures — so a longer board price is the market underrating it, in our read.

The setup. No strong pace edge either way — an even race on the figures.

How the number is built. We turn Brisnet’s power ratings into a within-race win probability, blend it 50/50 with the de-vigged morning line, then apply the published pace rule; fair odds are 1 ÷ that probability. Every constant is on the methodology page.

How to play it. Only a bet if you can still get 7/2 or better at post. On the tote you’re paid at the closing price no matter when you bet, so if it’s shorter than that by the gate, the value is gone — pass.

This is a v0 heuristic built on figures and pace. It does not yet read trainer, jockey, trip, or track bias — structured analysis, not a proven edge, and tracked publicly either way.

Full field ▾
#HorseModelFairBoard
2RED CHERRY26%3/18/5
3VICTORIOUS DREAM 24%3/16/1
4STRUCK BY HER17%5/17/2
1PEPPERMINT DASH12%15/24/1
5SOUL SISTER11%8/16/1
6WAITING FOR YOU10%9/16/1
Race 3 · 5 1/2 furlongs · dirt · Moc 150000Flag
1CHARLIE'S CLOCK
Model
Fair
Board
32%
2/1
5/2
our win %
break-even
the morning line
We flag it because the board (5/2) pays more than our fair price (2/1). Same horse, longer price.
The market
27%
Edge Card
32% +6 pts
▸ PLAY AT 5/2 OR BETTER
Ranks #1 in our power figure blend. Even pace
Data · Brisnet
The model’s read ▾

The number. Our model makes CHARLIE'S CLOCK a 32% chance to win; the morning line, once you strip out the track’s take, implies about 27%. That 6-point gap is the flag — we price the horse at a fair 2/1 and only bet if the board is still paying 5/2 or better at post.

Why we rate it. It tops this race on our power-rating blend — the strongest runner on the figures — so a longer board price is the market underrating it, in our read.

The setup. No strong pace edge either way — an even race on the figures.

How the number is built. We turn Brisnet’s power ratings into a within-race win probability, blend it 50/50 with the de-vigged morning line, then apply the published pace rule; fair odds are 1 ÷ that probability. Every constant is on the methodology page.

How to play it. Only a bet if you can still get 5/2 or better at post. On the tote you’re paid at the closing price no matter when you bet, so if it’s shorter than that by the gate, the value is gone — pass.

This is a v0 heuristic built on figures and pace. It does not yet read trainer, jockey, trip, or track bias — structured analysis, not a proven edge, and tracked publicly either way.

Full field ▾
#HorseModelFairBoard
1CHARLIE'S CLOCK 32%2/12/1
6BILLY GOAT20%4/15/2
4LOST ANGELES16%5/14/1
3BOSS MAN BOLT14%13/24/1
5FREE SOLO12%15/26/1
2JACK EM UP6%15/112/1
Race 4 · 5 furlongs · turf · Clm 40000Flag
1CODE DUELLO
Model
Fair
Board
24%
3/1
9/2
our win %
break-even
the morning line
We flag it because the board (9/2) pays more than our fair price (3/1). Same horse, longer price.
The market
18%
Edge Card
24% +6 pts
▸ PLAY AT 7/2 OR BETTER
Ranks #1 in our power figure blend. Contested pace likely (4 early-speed types)
Data · Brisnet
The model’s read ▾

The number. Our model makes CODE DUELLO a 24% chance to win; the morning line, once you strip out the track’s take, implies about 18%. That 6-point gap is the flag — we price the horse at a fair 3/1 and only bet if the board is still paying 7/2 or better at post.

Why we rate it. It tops this race on our power-rating blend — the strongest runner on the figures — so a longer board price is the market underrating it, in our read.

The setup. The race projects a contested early pace (4 speed types), which tends to set up for horses that come from off it. We adjusted its number down for that (×0.9).

How the number is built. We turn Brisnet’s power ratings into a within-race win probability, blend it 50/50 with the de-vigged morning line, then apply the published pace rule; fair odds are 1 ÷ that probability. Every constant is on the methodology page.

How to play it. Only a bet if you can still get 7/2 or better at post. On the tote you’re paid at the closing price no matter when you bet, so if it’s shorter than that by the gate, the value is gone — pass.

This is a v0 heuristic built on figures and pace. It does not yet read trainer, jockey, trip, or track bias — structured analysis, not a proven edge, and tracked publicly either way.

Full field ▾
#HorseModelFairBoard
1CODE DUELLO 24%3/17/2
4DONCIC14%6/13/1
8SMART CODE13%13/26/1
5NASTY HABIT12%7/14/1
3ALPHABETICAL ORDER12%15/26/1
7GREGORY'S PRIDE11%15/26/1
2MUBTADAA10%9/112/1
6TOM SEAVER3%30/115/1
Race 8 · 1 mile · turf · OceansidHL150kFlag
2SECURED FREEDOM
Model
Fair
Board
11%
8/1
15/1
our win %
break-even
the morning line
We flag it because the board (15/1) pays more than our fair price (8/1). Same horse, longer price.
The market
6%
Edge Card
11% +5 pts
▸ PLAY AT 9/1 OR BETTER
Ranks #1 in our power figure blend. Even pace
Data · Brisnet
The model’s read ▾

The number. Our model makes SECURED FREEDOM a 11% chance to win; the morning line, once you strip out the track’s take, implies about 6%. That 5-point gap is the flag — we price the horse at a fair 8/1 and only bet if the board is still paying 9/1 or better at post.

Why we rate it. It tops this race on our power-rating blend — the strongest runner on the figures — so a longer board price is the market underrating it, in our read.

The setup. No strong pace edge either way — an even race on the figures.

How the number is built. We turn Brisnet’s power ratings into a within-race win probability, blend it 50/50 with the de-vigged morning line, then apply the published pace rule; fair odds are 1 ÷ that probability. Every constant is on the methodology page.

How to play it. Only a bet if you can still get 9/1 or better at post. On the tote you’re paid at the closing price no matter when you bet, so if it’s shorter than that by the gate, the value is gone — pass.

This is a v0 heuristic built on figures and pace. It does not yet read trainer, jockey, trip, or track bias — structured analysis, not a proven edge, and tracked publicly either way.

Full field ▾
#HorseModelFairBoard
6MR. A. P.14%6/15/1
12LATER THAN PLANNED12%7/16/1
5BUST OUT12%15/24/1
2SECURED FREEDOM 11%8/112/1
3UNRIVALED TIME8%12/112/1
4PROLETARIAT8%12/18/1
8CHARLIE'S CURLIN8%12/16/1
1IRISEACH7%15/18/1
11JORDI BEAR6%15/112/1
7TIZ ALL THAT5%15/112/1
10ONE MORE FREUD5%20/115/1
9MO KOKO3%30/130/1
Passes no bet — our number agrees with the market
R17HARD TO FIGURE24%8/5
R54LOOKS LIKE N ANGEL18%7/2
R613FIRST PEACE19%3/1
R75REDIAL15%4/1
R99FRANK BULLITT17%4/1
R109ULYSSES ROSE22%7/2