Building Breeders’ Cup Saturday Doubles and Pick 3s: Where Coverage Pays
Five Saturday cards show why the horses you add to a ticket matter as much as the winners you find: seven CHV angles helped the Daily Doubles in this sample, while a tighter A-grade build worked better for Pick 3s.
By Jeff Bessa · CHV Insights · Breeders’ Cup Saturday analysis
This is the second article in our Breeders’ Cup Saturday series. The first study examined win bets across 45 races and found that price mattered alongside modeled win probability. Here, the question is different: how do we assemble a two- or three-race ticket without paying for too much coverage?
The build order below comes from experience using the Charting Horse Value handicapping application to construct tickets. It is not derived from the findings of the win-bet article. We tested that practical construction against the historical results and then compared what happened when we added more horses.
How the tickets were built
For each race, we started with the Bottom Line Win Bet selection, then added the top modeled W% horse, the top Trainer Jockey Rating (TJR) horse, and any remaining A-grade horses. For the Daily Doubles, we also tested adding horses flagged by one of seven CHV angles: a Green FC combination; a tr, dr, v, or + grade suffix; or the kt or tw FC Angle. We counted each horse only once within a leg. To test broader coverage, we then added remaining B-grade horses, ranked by modeled W% and alternating between legs, up to 50 combinations per Double or 100 per Pick 3 sequence. The Pick 3 B-grade comparison starts from the A-grade build; it does not include the seven-angle layer.
Our practical takeaway from this sample is to include the seven-angle horses on the Daily Double tickets, but stop the Pick 3 tickets after the remaining A-grade horses. The angle layer is shown for Pick 3s as a comparison, not as part of that suggested build. The angles are screening clues, not automatic bets in isolation; the race and price still matter.
The analysis covers five Breeders’ Cup Saturdays, 2021–2025: 45 races, 40 possible Daily Doubles, and 35 possible Pick 3s. Every combination in the tables is a flat $1 wager and every winning return is the actual $1 payout. A Pick 3 player can generally buy the same combinations for $0.50 each; that halves both the displayed cost and proportional payout, while leaving the ROI essentially unchanged, apart from possible payout rounding. Consecutive sequences share races, so these are overlapping observations, not 75 independent trials.
Five-year ticket results
| Daily Double build | Hits | $1 combinations | Return | ROI |
|---|---|---|---|---|
| Win Bet | 1/40 | 40 | $13.30 | −66.8% |
| + Top W% | 5/40 | 79 | $58.90 | −25.4% |
| + Top TJR | 8/40 | 191 | $115.90 | −39.3% |
| + Remaining A-grade | 9/40 | 200 | $121.60 | −39.2% |
| + Seven angles — suggested DD build | 11/40 | 295 | $336.96 | +14.2% |
| + B-grade, 50-combination cap | 16/40 | 921 | $432.50 | −53.0% |
The seven-angle layer picked up two more winning Doubles while adding 95 combinations. The full five-layer ticket returned $336.96 on $295 wagered. But that gain is highly sensitive to one $207.76 Double in 2022. It is a reason to keep testing the angle layer, not proof that every flagged horse belongs on every future ticket. Even with the 50-combination cap on each sequence, the B-grade expansion raised coverage to 16 hits but more than tripled the aggregate cost from $295 to $921.
| Pick 3 build | Hits | $1 combinations | Return | ROI |
|---|---|---|---|---|
| Win Bet | 0/35 | 35 | $0.00 | −100.0% |
| + Top W% | 1/35 | 94 | $30.60 | −67.4% |
| + Top TJR | 3/35 | 362 | $406.70 | +12.3% |
| + Remaining A-grade — suggested P3 build | 4/35 | 392 | $422.30 | +7.7% |
| + Seven angles, for comparison | 5/35 | 685 | $475.80 | −30.5% |
| + B-grade from A-grade build, 100-combination cap | 8/35 | 2,843 | $606.60 | −78.7% |
For Pick 3s, the remaining A-grade layer added one hit for only 30 more combinations. The seven angles added another hit, but the ticket grew by 293 combinations and its pooled ROI turned negative. We therefore stop the illustrated Pick 3 build at the A-grade layer. That +7.7% result is also fragile: $406.70 of its $422.30 return came from the 2025 card. Starting from that A-grade build, the capped B-grade version covered more sequences but cost far more than the payouts collected. The caps apply to each sequence, while the cost and return columns sum all five years; none of the core tickets reached either cap before B-grade horses were added.
Does HxS help choose which sequences to play?
Our “Stop Playing Every Daily Double” analysis makes the case for selecting promising sequences instead of buying every Double. We explored HxS in these Breeders’ Cup cards using the DD build through seven angles and the Pick 3 build through remaining A-grade horses. “Lowest HxS 3 to under 4” means every leg scored at least 3, with the lowest leg below 4. The “one leg 3+” screen is broader. These screens overlap and their returns must not be added together.
| HxS screen | DD sequences | DD hits | DD $1 cost | DD ROI | P3 sequences | P3 hits | P3 $1 cost | P3 ROI |
|---|---|---|---|---|---|---|---|---|
| Lowest leg 3 to under 4 | 12 | 5 | $89 | +194.8% | 8 | 2 | $96 | +197.4% |
| Lowest leg 4 to under 5 | 6 | 0 | $54 | −100.0% | 4 | 0 | $40 | −100.0% |
| At least one leg 3+ | 34 | 10 | $264 | +25.4% | 33 | 4 | $382 | +10.5% |
The 3-to-under-4 band is intriguing, but it contains only 12 Doubles and 8 Pick 3s. The 4-to-under-5 band had no hits in this sample. That is too little evidence to turn any of these HxS bands into a dependable rule. The screen is useful here as a question for further testing, not a replacement for examining each leg and its ticket cost.
The 2025 card: a useful reality check
The 2025 Saturday produced eight consecutive Doubles and seven Pick 3s. The figures below show the winning combinations, their actual $1 payouts, and the first step at which the full sequence was covered. “Not covered” means the winner was still absent after the recommended DD or Pick 3 build.
| Suggested 2025 ticket | Sequences hit | $1 combinations | Return | ROI |
|---|---|---|---|---|
| Daily Doubles through seven angles | 5/8 | 45 | $96.60 | +114.7% |
| Pick 3s through remaining A-grade | 3/7 | 88 | $406.70 | +362.2% |
In 2025, the Win Bet, top W%, and top TJR layers had already included every horse that qualified through the remaining A-grade and seven-angle layers. Those later layers added no unique horses, combinations, or newly covered winning sequences on that card. That is the precise 2025 observation; it does not mean A-grade or angle horses were absent from the races, nor does it erase the incremental coverage those layers provided across the full five-year sample.
The 2025 winners also show what the method missed. The $173.30 R5–R6 Double and the $1,479.10 R4–R6 Pick 3 were not covered by these suggested builds. Large payouts can swing a small historical test in either direction, and a profitable year is not a guarantee for the next card.
What we would carry forward
Start with the Bottom Line Win Bet, top W%, and top TJR selections, then check whether remaining A-grade horses add a sensible amount of coverage. For the Daily Double, consider the seven-angle qualifiers as the next layer and measure the added combinations before wagering. For the Pick 3, the historical comparison favors stopping at the A-grade layer rather than adding every angle horse. Even with per-sequence caps, the B-grade expansion cost more than its extra hits returned in this sample.
Use the CHV Glossary to decode the signals, read the earlier win-betting study for the companion analysis, and explore the CHV handicapping application to build and price your own tickets.
These are historical, exploratory results from five cards. They are not guaranteed returns or a standing instruction to make a wager.