I Named Every Winner on the Card and Cashed One Bet...

Del Mar, Friday July 31 — the recap

There is a specific kind of bad day at the races that doesn't feel bad while it's happening.

You're not tearing up tickets in disgust. You're not watching horses you never considered come flying down the lane at 30-1. You're watching the race unfold, seeing a name you recognize, thinking I had him — and then remembering you had him fourth.

That was Friday. Eight races. Eight winners. Every single one of them was in my sheet by name. I finished the day down $244.

Let me show you exactly how that happens, because I think it's the most useful thing I can give you.

The scoreboard

One for eight on the top line. Eight for eight on the page.

The average finishing position of my headline selection was fourth.

The thing that actually killed me

I had two horses I loved. Not liked — loved. Speedy Wilson in the fourth and Kokosan in the sixth.

Kokosan was dropping from Grade 1 and open-company allowance races into a $25,000 claimer. His last four Beyer figures were 92, 85, 79, 76. Nobody else in the race had cracked 82 all year. He worked a bullet half at Del Mar, best of 101 horses that morning. Sadler wins 25% with claimers. Rosario rode.

Speedy Wilson had run six times at Del Mar and won three of them, with a quarter-million dollars earned over this exact strip, cutting back from a mile into a six-furlong race with three horses that all had to have the lead.

I singled both of them in every multi-race ticket I built. That's what you do with your best two opinions — you use them as the foundation and spend the savings on coverage in the races you can't figure out.

Kokosan ran third. Speedy Wilson ran fourth.

When you single a horse, there is no version of the ticket where you survive him losing. I had a Pick 5 covering the last five races that went four horses deep in three separate legs — genuinely good, wide coverage in exactly the races where I admitted I was guessing. It died in leg one, before any of that coverage ever mattered.

Every horizontal ticket I fired went to zero. All of them. Seven wagers, $190, nothing back.

The mistake I made twice in the same afternoon

Here's the part I actually want you to take away, because it cost me both the fourth and the seventh and it's a mistake I'd bet most of you make too.

Both races looked like pace meltdowns. In the fourth, three horses showed monster early speed figures — 126, 121, 120 — at six furlongs, in a seven-horse field. In the seventh, five horses showed 96 or better early in a seven-furlong stakes.

Classic setup. Too much speed, they'll all cook each other, the closer picks up the pieces. I took the best closer in both races.

And I was right about the meltdown. Watch what happened:

Race 4. Simple Song (126 early) — "vied inside, gave way," sixth. Hejazi (121) — "ins, tipped out, tired," seventh. For All Mankind — "vied between, folded," fifth. Three horses hooked up early, three horses cooked.

Race 7. Courtside Action, the fastest early figure in the race and undefeated — eleventh, beaten forty lengths. Lit, the second fastest — seventh.

The speed died. Exactly as predicted. And in both races, the winner was a speed horse anyway.

Captivator was one of the three I fanged in race four. He was the third fastest of the three. Chart comment: "3 deep, clear, held well." He wasn't in the duel. He sat just outside it, three paths off the rail, and when the two inside of him folded he was already clear.

He's A Knockout was the third fastest of five in race seven. Same story. And Party Animal, who was the slowest early figure of the five speed horses, ran second at 25-1.

Here's what I now believe, and I'm writing it down so you can hold me to it: when three or more horses show top early speed, the two fastest of them do collapse — and the third or fourth fastest gets a clean stalking trip and wins. The closer doesn't get there. The race is too honest early for the closer to matter, but the horse sitting a length off the duel is in perfect position.

Every deep closer I leaned on Friday ran up the track. Speedy Wilson fourth, Penny Worth ninth, Start The Ride fourth, Impressionistic fifth, Bad To The Bones eighth of nine.

Now read the chart comments of the winners:

• Race 1: "bid 2w, held"
• Race 2: "clear to 1/16, held on"
• Race 4: "3 deep, clear, held well"
• Race 5: "3 deep, clear, held late"
• Race 6: "stalk, bid 2w, held gamely"

Six of eight winners were pressers or stalkers who got clear. Zero deep closers won a race on that card.

The rail read: I was wrong to look for a bias at all

The temporary rail was at 24 feet — the same setting as the previous card, where the inside completely inverted and all eight winners came off the rail. I went in with a tripwire: watch the first two turf races, and if the winners come three or more paths wide, the rail is dead.

Race three looked like a screaming wide bias. The winner shifted out, the runner-up rallied three-to-four wide, the third-place filly came outside, and the two horses who saved ground finished eighth and ninth, beaten seventeen and twenty lengths.

So I'd have upgraded the wide closers in the turf routes later. And I'd have been wrong.

Race six went to a horse who stalked and bid two wide, with the runner-up two-wide-then-inside and the third-place horse two-wide bidding three deep. Tight, mid-path, nothing extreme. My designated wide-bias horse ran eighth of nine.

The turf wasn't biased Friday. It was neutral. My tripwire only has two settings — inside speed or wide closing — and there was no setting for "the track is fine, just handicap the race." Forcing a binary onto a neutral surface is how you talk yourself out of the right horse.

The wager nobody hit

Del Mar cards a $1 Place Pick All — you need a horse in the top two in every race on the card. Friday nobody hit all eight. The seven-of-eight consolation paid $2,342.60.

Mine died in leg four. Speedy Wilson, singled again.

What stings is that in three of the four legs I missed, the horse I needed was already on my page with an explicit instruction next to it. March of Time, who ran second in the fourth — I wrote "on every ticket." Tribal, who won the sixth — I wrote "on every ticket." Janz Joy, who ran second in the eighth — I'd flagged her as a conditional add.

I left all three off the Pick All to hold the ticket cost down. That's not a handicapping mistake. That's a discipline mistake, and it's cheaper to admit than to repeat.

What changes

No singles in the Place Pick All. Two per leg minimum, and the second horse is whoever I've already called the danger. It makes the ticket cost more. It should.

Before I single a horse in a Pick 4 or Pick 5, I write down what happens to the ticket if he runs third. If the honest answer is "it's dead in leg one," he isn't a single. I'd literally written in my own notes that Kokosan's race was the leg most likely to cost me the Pick 5 — and then I singled him anyway.

Pace meltdowns get the third and fourth fastest early horses on the ticket, not a closer.

And the honest number

Eight winners named. One cashed. $277 out, $32.60 back.

The handicapping wasn't the problem Friday. Everything I needed was on the page. What failed was the ranking and the ticket construction — the part where opinions turn into money — and that's the part I can actually fix.

See you Saturday.

The Golden Rail Report. Del Mar, July 31, 2026. Nothing here is betting advice. Wager responsibly, and if it stops being fun, stop.