StableBet
Professor Furlong and Pascal at the AI Lab
THE AI LAB
THE AI LAB DAILY · THE MARKET ON SATURDAY, 19 SEPTEMBER 2026

Saturday 19 September 2026 steamers and drifters: Miletus at Newmarket

← Every day·Today's card·The day so far on Sunday, 20 September 2026·Yesterday's results

This is the market piece for Saturday, 19 September 2026, kept as an archive. The latest market pieces are here.

The market for Saturday 19 September 2026 has moved since the morning show. The market has seen Miletus, Nostrum and Paws For Thought shorten sharply, while Saucy Jane, Emperor Aurelius and Majestic Ruler have drifted, with Ayr, Chester, Gowran Park, Navan, Newbury, Newcastle and Newmarket still to race. Of the day's 53 races, 46 were still to be run when the opening show was set against the morning market, the next of them the Ladbrokes 'Get Racing Bet Builder Winnings Boosts' Doonside Cup Stakes (Listed) at 13:55 at Ayr, and the moves below are drawn from those races alone. The morning market is the price the day's predictions recorded for each runner before the first show; the opening show is the middle bookmaker price in the Lab's first snapshot of the day, exchanges left out. A steamer is a runner whose price has shortened between the two, a drifter one whose price has lengthened, and a move is a description of the market's opinion, never advice. Every runner, with the model's estimated chance beside the price, is on the racecards page.

Steamers: Miletus at Newmarket

The biggest shorteners since the morning market, largest first: Miletus has shortened sharply for the Turners Handicap at 14:25 at Newmarket, Nostrum has shortened sharply for the Turners Parks Group Ltd Handicap (Div 1) at 15:00 at Newmarket and Paws For Thought has shortened sharply for the Kidsbank Chester Handicap at 15:46 at Chester. The race the market has moved most for is the Turners Handicap at Newmarket, and the card below sets the model's estimated chance for each runner beside the chance the morning market implied.

Stablebet model (pre-race estimated win chances)

Newmarket 14:25 · Turners Handicap · Class 4 · 7f

Latest output →

Pre-race estimate. Published 19 September 2026, before the off. This is what the model made of the race beforehand, kept as a record. It was never published as a tip.

The model rated Leadman its most likely winner at 10%, ahead of Cherry Cobbler (10%) and Bintola (8%). It read as a competitive, open race with no dominant runner.

HorseSPModel chanceMarketvs Market
Leadman
Toby Moore / M Appleby
11.00
9.8%
6.9%Market 2.8pp longer
Cherry Cobbler
C R Lyons / S C Williams
3.75
9.6%
20.3%Market 10.7pp shorter
Bintola
Lewis Edmunds / Harry Charlton
13.00
8.3%
5.9%Market 2.4pp longer
Rogue Allegiance
Harry Davies / James Owen
5.00
7.7%
15.2%Market 7.5pp shorter
Nanoscience
David Probert / A M Balding
8.00
7.1%
9.5%Market 2.4pp shorter
Golden Strike
J Garritty / R A Fahey
8.00
6.9%
9.5%Market 2.6pp shorter
Miletus
J Mitchell / J Keane
51.00
6.7%
1.5%Market 5.2pp longer
Hardy's Hero
H Crouch / W J Haggas
11.00
6.7%
6.9%In line with market
Lady Mariko
R Havlin / Dylan Cunha
51.00
6.1%
1.5%Market 4.6pp longer
Hungarian
Benoit D L Sayette / Lemos Souza
51.00
6.1%
1.5%Market 4.6pp longer
Paladin
Callum Shepherd / K Frost
51.00
6.0%
1.5%Market 4.5pp longer
Raggaas
Billy Loughnane / S & E Crisford
9.00
5.9%
8.5%Market 2.6pp shorter
Kiniro
Paddy Bradley / Dan Horsford
34.00
4.7%
2.2%Market 2.5pp longer
Berlinetta
George Bass / J Channon
11.00
4.3%
6.9%Market 2.6pp shorter
Lunario
Jack Dace / L A Dace
34.00
4.2%
2.2%Market 2.0pp longer

These were the model's estimated win chances going into the race, not tips. The model is calibrated to real results (when it says 25%, about 25% win) but it does not beat the market, so read it as an independent second opinion on the race. See how accurate it is →

Model: ensemble-v1.0 · Generated Sat, 19 Sep 2026The LabMethodology

The model rates Leadman its most likely winner, and the market has come for Miletus since the opening show. The model's estimate was made before the first show and does not move with the market, so the gap between its column and a steamer's price is a difference of opinion, not a verdict on either; the track record settles every one of its picks at starting price. The course itself is covered in the Newmarket guide.

Drifters: Saucy Jane at Newmarket

The biggest lengtheners since the morning market, largest first: Saucy Jane has drifted sharply for the Turners Parks Group Ltd Handicap (Div 2) at 15:45 at Newmarket, Emperor Aurelius has drifted sharply for the Turners Of Soham Novice Stakes at 16:55 at Newmarket and Majestic Ruler has drifted sharply for the Turners Of Soham Novice Stakes at 16:55 at Newmarket. The race with the biggest drift is the Turners Parks Group Ltd Handicap (Div 2) at Newmarket, shown below with the model's estimated chance beside the morning market's.

Stablebet model (pre-race estimated win chances)

Newmarket 15:45 · Turners Parks Group Ltd Handicap (Div 2) · Class 4 · 6f

Latest output →

Pre-race estimate. Published 19 September 2026, before the off. This is what the model made of the race beforehand, kept as a record. It was never published as a tip.

The model rated Akabusi its most likely winner at 14%, ahead of Old Is Gold (14%) and Saucy Jane (11%). It read as a competitive, open race with no dominant runner.

HorseSPModel chanceMarketvs Market
Akabusi
Grace McEntee / M Fenton
13.00
14.4%
6.4%Market 8.0pp longer
Old Is Gold
Billy Loughnane / A M Balding
2.88
13.7%
29.0%Market 15.3pp shorter
Saucy Jane
Harry Vigors / Jack Morland
4.50
10.9%
18.5%Market 7.6pp shorter
Quantum Power
Molly Keegan-Price / Tom Clover
5.00
10.9%
16.7%Market 5.8pp shorter
Sierra Sands
Callum Shepherd / E A L Dunlop
9.00
10.7%
9.3%Market 1.5pp longer
Kohana Nation
David Probert / C Hills
11.00
10.4%
7.6%Market 2.8pp longer
Alaminos
H Crouch / E Bethell
13.00
10.1%
6.4%Market 3.7pp longer
Waistcoat
J Mitchell / R Burdon
17.00
9.7%
4.9%Market 4.8pp longer
Enricher
Lewis Edmunds / W Muir & C Grassick
67.00
9.2%
1.2%Market 8.0pp longer

These were the model's estimated win chances going into the race, not tips. The model is calibrated to real results (when it says 25%, about 25% win) but it does not beat the market, so read it as an independent second opinion on the race. See how accurate it is →

Model: ensemble-v1.0 · Generated Sat, 19 Sep 2026The LabMethodology

A drift is the market's opinion changing, and the reasons range from a non-runner or a change of going to a stable's own confidence, none of which the price itself explains. The model does not react to it: its estimate is the same one the morning card carried, which is why a drifter can sit above the market in the left column and below it in the right. The horse tracker follows any runner on the card to its next entry, whichever way its price went.

The cost of backing early

Taking a price before the off means taking the bookmaker's margin at that moment, and the margin is paid whichever way the price then moves: a steamer rewards the early price and a drifter punishes it, but the excess the book keeps over a whole field is built into both. The table below is the latest published margin for each book the Lab tracks, compiled from the previous evening's prices, so it describes the card as it was sold then rather than the lunchtime board.

BookmakerRacesOverroundTakeout
1 SmarketsSBK59215.67%13.55%
2 BetTom51623.62%19.11%
3 Bet36575323.94%19.32%
4 LiveScore Bet shared feed59824.63%19.76%
= Virgin Bet shared feed59524.65%19.77%
6 William Hill77925.63%20.40%
7 Unibet77526.89%21.19%
= Quinn Bet76727.07%21.30%
= Hollywood Bets75327.18%21.37%
= 7Bet64227.20%21.38%
11 Paddy Power80327.49%21.56%
= Betfred74127.51%21.57%
= 10 Bet72527.62%21.65%
14 Coral shared feed80628.04%21.90%
= Ladbrokes shared feed80628.07%21.92%
= Star Sports80728.19%21.99%
17 Midnite78328.86%22.40%
= Boyle Sports79628.99%22.48%
19 Bet Victor shared feed57029.44%22.75%
= Betano shared feed56829.61%22.85%
21 Dragon Bet80729.98%23.06%
= BetWright80730.02%23.09%
23 Betfair Sportsbook80230.36%23.29%
24 BresBet21732.04%24.27%
25 BetMGM80732.41%24.48%

First pre-race book of each day, 2026-08-27 to 2026-09-20, across 809 races every listed book priced in full. Exchanges and spread firms are excluded: they do not quote comparable win odds. Overround is how far past a fair book the prices sit; takeout is the share of stakes the book keeps. On 23 days of data, books marked = are level with the one above rather than behind it. Books marked "shared feed" are one half of a pair that does not price independently, so there are fewer independent books here than rows.

What betting costs has the working behind the margin, and the Tipster League shows what taking a price every day does to the AI tipsters over a season, losses included.

Elsewhere in the Lab: How a Horse's Form Becomes Its Odds (And How Accurate the Price Really Is).

Common questions

Which meetings race on Saturday 19 September 2026?

Navan and Gowran Park in Ireland, and Newbury, Newmarket, Ayr, Chester and Newcastle in Britain. Every card is on the racecards page with the model's estimated chance beside each runner, and Navan, Gowran Park, Newbury, Newmarket, Ayr, Chester and Newcastle still have racing to come.

What is a steamer?

A runner whose price has shortened since an earlier market, here the morning market recorded in the predictions file. The opposite move, a price lengthening, makes a drifter.

Does a steamer win more often?

The Lab makes no such claim. A shortening price is the market's opinion strengthening, and the track record settles the model's own picks at starting price rather than at the price they opened at, so the answer for any system is a running figure over every day, not a rule.

Where do the prices come from?

From the Lab's own snapshot of the bookmakers' boards, taken in the late morning and compared with the morning market the model priced against. Exchange prices carry commission and are left out.

How are the AI tipsters doing?

The Tipster League settles every AI pick at starting price and publishes the running total, losses included.

A word on all of this

None of the above is a tip, and no system the Lab has tested has made money over the whole record. Bet only what you can afford to lose. If gambling has stopped being fun, BeGambleAware has free, confidential help.

Every figure on this page renders live from our published feeds and nothing here is a tip. For whether the model holds any edge, see our track record. 18+, please bet responsibly.