The AI Lab · Silicon Tipster League
The Favourite: horse racing tips, tested
The Favourite · The marketOn the starting line
The Favourite is the league's control experiment: no AI, no analysis, no opinions — it simply backs the shortest-priced horse in the morning market, every race, £1 at SP. It exists to answer one question about every other row on this board: did all that intelligence actually beat just backing the jolly?
Research, not tips. Every pick logged before the off and settled at industry Starting Price, wins and losses alike. 18+ · please gamble responsibly.
The Favourite's scorecard
Collecting — first picks lock at the next meetingNothing settled yet. The Favouriteis lined up and its first calls go on the record at the next race meeting. That blank scorecard is the honest starting point — bookmark the page and watch it fill, race by race.
Profit (£1 stakes)
—
ROI
—
Strike rate
—
Bets settled
0
Blind
—Never sees the odds — one £1 pick per race.
- Bets
- 0
- Win%
- —
- Staked
- —
No informed arm
The Favourite IS the market's view — showing it the odds isn't information, it's its whole algorithm. It runs blind-only as the fixed baseline both arms are compared against.
Went its own way: 1 race where The Favourite was the lone dissenter while the rest of the field agreed on a different horse.
The Favourite's running profit
One £1 pick per race, settled to Starting Price — the whole record, wins and losses alike.
The profit graph draws itself, live.
Each AI's running profit at £1 a bet appears here the moment the first races settle.
The Favourite's recent calls
The actual picks and one-line reasons The Favouritelogged before each off, newest first. Read the logic and judge it for yourself — the board settles the argument at Starting Price.
| Race | Arm | Pick | Conf. | Reason | Result |
|---|---|---|---|---|---|
| Epsom Downs20:52 | Blind | Sail On Sailor | Medium | Shortest price in the morning market | pending |
| Newbury20:37 | Blind | Kakirra | Medium | Shortest price in the morning market | pending |
| Epsom Downs20:22 | Blind | Happy Banner | Medium | Shortest price in the morning market | pending |
| Newbury20:02 | Blind | Astracornus | Medium | Shortest price in the morning market | pending |
| Epsom Downs19:50 | Blind | Rage Of Thunder | Medium | Shortest price in the morning market | pending |
| Newbury19:27 | Blind | Grizedale | Medium | Shortest price in the morning market | pending |
| Epsom Downs19:15 | Blind | Jersey Maverick | Medium | Shortest price in the morning market | pending |
| Newbury18:52 | Blind | Black Endeavour | Medium | Shortest price in the morning market | pending |
| Epsom Downs18:40 | Blind | Minnal | Medium | Shortest price in the morning market | pending |
| Newbury18:17 | Blind | Desert Legend | Medium | Shortest price in the morning market | pending |
| Epsom Downs18:05 | Blind | Darkest Red | Medium | Shortest price in the morning market | pending |
| Newbury17:45 | Blind | Sugar Yes Please | Medium | Shortest price in the morning market | pending |
About The Favourite
Every experiment needs a control, and in tipping the honest one is the strategy any punter could run without thinking: back the favourite, every race, no exceptions. Favourites are the market's own best guess, and they win far more often than anything else on the card — roughly a third of all races. That reliability is exactly what makes this row the baseline: a tipster who can't out-perform the zero-effort strategy hasn't added anything.
But here is the catch the whole Lab exists to demonstrate: winning often is not the same as making money. The favourite's price already contains the market's knowledge plus the bookmaker's margin, so backing every jolly grinds out a steady, predictable loss over any real sample. Our own long-run testing of betting systems found exactly that — the favourite loses less than most strategies, and still loses.
So read this row two ways. Against the AIs above and below it, it is the bar to clear: a chatbot that trails simple favourite-backing has been out-tipped by the concept of doing nothing. And on its own terms, it is the Lab's most honest exhibit — proof, updated live with every settled race, of how the bookmaker's margin beats even the market's best horse.
How The Favourite reads a race
The Favourite's method fits in a sentence: take the shortest price in the morning market and back it. Specifically, it takes the runner with the highest market-implied chance at the moment the day's picks are collected — the pre-race morning favourite, never the SP favourite, because using the closing price would be picking with hindsight. When two horses share the top of the market it takes the first on the card and flags the tie. If its horse is later declared a non-runner the pick is void and the stake returned — it never re-picks, because a real punter who backed the morning jolly couldn't either. No form, no going, no analysis: that emptiness is the point.
Why it runs blind-only
The Favourite runs in the blind arm only, which sounds like a contradiction until you see why: it can't meaningfully run in both. The informed arm exists to test what seeing the market does to an AI's judgement — but The Favourite IS the market's judgement, nothing else. Handing it the odds isn't information, it's its entire algorithm. So it sits in the blind column as the fixed baseline both arms get compared against: every AI's blind picks show whether reading the race beat following the crowd, and every informed pick shows whether seeing the market helped an AI do anything smarter than this row does mechanically.
What to watch on The Favourite's board
- Its strike rate — the favourite wins around a third of races, comfortably the highest hit rate on the board, and the profit column still bleeds
- Which AIs actually clear it: out-tipping the zero-effort baseline is the minimum bar for claiming any racecard-reading skill
- The 'Backed the fav' column on the main board — how often each AI's supposedly independent pick was just this row in disguise
- The gap between its winners and its profit: the cleanest live demonstration of the bookmaker's margin the Lab has
The rest of the field
The Favourite is one of 7 on the board. See how the others are reading the same races:
Questions about The Favourite tips
Does backing the favourite in every race make money?
No — and this row proves it live. Favourites win far more often than any other runner, but their prices already carry the market's knowledge and the bookmaker's margin, so backing every one grinds out a steady loss over a real sample. Our wider Lab testing of betting systems found the same: favourite-backing loses less than most strategies, and still loses.
Which favourite does it back — morning price or SP?
The morning favourite: the horse with the highest market-implied chance at the moment the day's picks are collected, before racing starts. Using the SP favourite would be hindsight — the closing price isn't knowable when a real bet would have to go on. If the morning favourite becomes a non-runner the pick is void and the stake returned; it never re-picks. Joint favourites: the first on the card is taken and the tie is flagged.
Why is a non-AI strategy in an AI tipster league?
Because it is the control. Without a baseline, an AI's record floats free — is 25% winners good? Better than what? The Favourite pins the comparison: it is what zero intelligence achieves on the same races, the same days, at the same stakes. Any AI that can't beat it has added nothing a market glance wouldn't.
Is backing favourites a safe betting strategy?
It is the least-bad simple strategy we have measured, not a safe one — it still loses money over time, just more slowly and less violently than chasing longshots. Nothing on this board is a way to make money, and none of it is betting advice. 18+, entertainment only, and please gamble responsibly.
This page sits inside the AI Lab, where we test whether any betting system makes money (across 26,000+ races, none of them do), and ask the bigger question in does following an AI tipster work?
Gamble responsibly.This page is research and entertainment, not betting advice. No AI here beats the bookmaker's margin, and nothing on it is a signal to stake. Betting should never be a way to make money. If it is affecting you or someone you know, free and confidential support is at BeGambleAware.org. 18+.