The Welsh Grand National trends are mostly true, but tested against the Betfair price across nine runnings and nearly 1,500 runners in similar marathon chases, none of them has been a reliable way to make money. The closest thing to an edge is the route through Chepstow's own trial, which has produced four of the last eight winners, and even that is too small a sample to call proven.
“A race to savour in between Christmas and New Year.”
Simon Walton, RaceMetrics founder
How to read the numbers. Every figure is measured against Betfair SP, which carries no bookmaker margin, so it is the fairest price to judge a group of horses against. A/E (actual over expected) is winners divided by the winners the prices said there should be: 1.00 is par, 1.10 means 10% more winners than the market expected. The ± figure is one standard error, roughly how far the number could wander by luck alone. A result inside two standard errors of 1.00 is not a finding.
Level stakes is the profit or loss from a 1-point win bet on every horse in the group, at Betfair SP after 2% commission and at industry SP.
The race
The Welsh Grand National is Chepstow's marathon handicap chase, run over about three miles and six and a half furlongs on 27 December. It is often run on soft or heavy ground and the weather has moved it more than once: the 2017 running took place on 6 January 2018 and the 2020 running on 9 January 2021. We found every running by course, month, race type, trip and class rather than by title, so both January dates are included.
Our race records start in 2018, which gives nine runnings and 163 runners.
| Running | Winner | Age | Weight / OR | SP | Betfair SP | Last run | Days off |
|---|---|---|---|---|---|---|---|
| 2017 (run Jan 2018) | Raz De Maree | 13 | 10-10 / 140 | 16/1 | 25.7 | Before our records | 32 |
| 2018 | Elegant Escape | 6 | 11-8 / 151 | 3/1 fav | 4.5 | 2nd, chase, Newbury | 26 |
| 2019 | Potters Corner | 9 | 10-4 / 145 | 8/1 | 12.2 | Won, hurdle, Chepstow | 20 |
| 2020 (run Jan 2021) | Secret Reprieve | 7 | 10-1 / 134 | 5/2 fav | 3.7 | Won, chase, Chepstow | 35 |
| 2021 | Iwilldoit | 8 | 10-0 / 136 | 13/2 | 10.0 | Won, chase, Chepstow | 23 |
| 2022 | The Two Amigos | 10 | 10-1 / 127 | 16/1 | 19.5 | 3rd, chase, Haydock | 27 |
| 2023 | Nassalam | 6 | 11-3 / 145 | 9/2 fav | 7.1 | Won, chase, Chepstow | 18 |
| 2024 | Val Dancer | 7 | 10-6 / 126 | 8/1 | 10.5 | Won, chase, Carlisle | 54 |
| 2025 | Haiti Couleurs | 8 | 11-13 / 154 | 5/1 | 6.9 | Pulled up, chase, Haydock | 35 |
Weight is the weight carried, after any jockey's claim. Betfair SP is shown as decimal odds.
Three favourites have won in nine runnings, and the winners came from a fairly narrow price band: none bigger than 16/1 at SP. Eight of the nine had run within five weeks.
The trends, checked against the winners
| The claim | Our count |
|---|---|
| Every winner had won a chase over three miles or more, counting the Welsh National Trial | 8 of 8 that we can see |
| Nearly every winner finished in the first four last time | 7 of 8 that we can see |
| Most winners carried 11st or less | 6 of 9 |
| Most winners were aged 6 to 8 | 6 of 9 |
| The trial winner often goes on to win the National | 4 of 8 winners ran at Chepstow last time; 3 of them won the trial |
The trends hold. Our records do not reach back to Raz De Maree's earlier runs, so he is left out of the first two counts. Potters Corner's Chepstow run was a hurdle, not the trial.
Inside the race: what the market expected
| Welsh Grand National, nine runnings | Winners | Runners | Expected winners | A/E | Level stakes, Betfair SP | Level stakes, SP |
|---|---|---|---|---|---|---|
| Every runner | 9 | 163 | 9.2 | 0.98 | −39% | −52% |
| Had won a chase over 2m7f or more | 8 | 125 (77%) | 7.4 | 1.08 ±0.37 | −42% | −51% |
| Finished in the first four last time | 7 | 82 | 6.0 | 1.16 ±0.41 | −18% | −32% |
| Carrying 11st or less | 6 | 114 (70%) | 5.4 | 1.11 ±0.43 | −29% | −45% |
| Aged 6 to 8 | 6 | 79 (48%) | 5.5 | 1.08 ±0.42 | −46% | −55% |
| Ran at Chepstow last time | 4 | 21 (13%) | 1.5 | 2.66 ±0.82 | +55% | +21% |
The Welsh National has been a hard race to back. Every runner at Betfair SP lost 39% of stakes, and every trend group except one lost too. The staying-chase trend is the clearest case: it fits eight of eight winners because it fits three-quarters of the field.
The exception is the Chepstow route. Twenty-one horses came to the race off a run at the course, and four won, against 1.5 expected. That is a big ratio on very few horses. Nine races cannot settle it, so we widened the test.
Across similar races
The comparison set is every British and Irish Class 1 to 3 handicap chase of three and a half miles or more, run from December to February since 2018: 123 races and 1,476 runners. For the staying-chase trend we used races from July 2020, so that our records hold enough of each horse's earlier career.
| Handicap chases of 3m4f and over, Dec to Feb | Runners | Strike rate | Rest of field | A/E | Level stakes, Betfair SP | Level stakes, SP |
|---|---|---|---|---|---|---|
| Every runner | 1,476 | 8.3% | 0.99 | −4% | −23% | |
| Had won a chase over 2m7f or more (from July 2020) | 734 | 8.3% | 11.4% | 0.94 ±0.12 | −16% | −29% |
| Finished in the first four last time | 721 | 10.1% | 6.6% | 0.99 ±0.12 | +2% | −17% |
| Carrying 11st or less | 842 | 8.0% | 8.8% | 1.02 ±0.12 | +1% | −18% |
| Aged 6 to 8 | 617 | 9.9% | 7.2% | 0.95 ±0.12 | −11% | −25% |
| Ran at the same course last time | 192 | 10.9% | 7.9% | 1.18 ±0.24 | −7% | −22% |
- The staying-chase trend is not real. Horses that had already won a long chase won 8.3% of the time. Those that had not won 11.4%, and beat their prices (19 winners from 15.5 expected). The difference is under one standard error, so it is not a system either way, but it is no reason to rule a horse out.
- Weight is not real. Horses carrying 11st or less are most of the field and win slightly less often than the rest.
- Recent form and age are real but priced in. A top-four finish last time lifts the strike rate from 6.6% to 10.1%, and the market knows: 0.99. Horses aged 6 to 8 win more often than older ones and are, if anything, a shade over-bet.
- Running at the same course last time is the best of them. These horses won 10.9% of their races against 7.9% for the rest, with an A/E of 1.18. That is under one standard error from par, and it lost at both prices. The market has priced the course route in these races in general. In the Welsh National itself it has done better than that, but on four winners.
The verdict on each trend
| Trend | Verdict |
|---|---|
| Had won a chase over three miles (or the trial) | Not real. It describes almost every runner. |
| Finished in the first four last time | Real but priced in. |
| Carrying 11st or less | Not real. |
| Aged 6 to 8 | Real but priced in. |
| Ran at Chepstow last time (the trial route) | Real. It has paid in this race, on four winners. In similar races it is mostly priced in. |
What we looked for and did not find
We checked two things no trends table covers. Horses that usually make the running have done a little better than their prices in marathon chases, an A/E of 1.25 from 16 expected winners, and the held-up horses won 1 Welsh National from 2.4 expected. And horses returning after running in the previous year's race have won none of the last eight runnings, from 26 runners and 1.7 expected winners. Neither is anywhere near strong enough to use. We mention them so you know they were tested.
How to use this on the day
- Build the comparison in the System Builder. Race Type Chase, Handicap Yes, Distance (furlongs) 28 or more, Class 1 to 3, Month December to February.
- Test the course route. Add Same Course as LTO set to Yes. For the Welsh National alone, add Course Chepstow and LTO Course Chepstow, and use LTO Race Name to pick out the trial.
- Check the other trends one at a time: LTO Finishing Position 1 to 4, Age 6 to 8, Weight Carried 154lb or less, Distance Wins (today's trip) 1 or more. Watch the A/E, not the strike rate.
- On the day, open the race's Race Guide and racecard. Check where each runner ran last, how it is usually ridden and its price. RMP is an opinion to test against the market, not a tip, and it has no history in this race yet.
- Save the system as Active and its qualifiers will show on the day for this race and every similar one.
The take
The Welsh Grand National trends describe the winners accurately and have not been a way to beat the price. The staying-chase and weight trends fit nearly every runner, and recent form and age are already in the odds. The one line worth carrying into the race is the Chepstow route: four of the last eight winners ran at the course last time, from only 21 runners. Treat it as a reason to look closely at those horses and their prices, not as a rule.
Questions we get asked about this
Does the Welsh National Trial winner win the Welsh Grand National?
Often lately. Secret Reprieve, Iwilldoit and Nassalam all won the trial at Chepstow before winning the National, and four of the last eight winners ran at Chepstow last time. Those 21 runners produced 4 winners against 1.5 expected. It is a small sample, so it is a lean rather than a rule.
Do light weights win the Welsh Grand National?
Six of the last nine winners carried 11st or less, but light weights are 70% of the runners. In similar marathon handicap chases, horses on 11st or less win slightly less often than the rest, with an A/E of 1.02.
When is the Welsh Grand National run?
On 27 December at Chepstow, over about three miles six and a half furlongs. Bad weather has moved it into January, as in the 2017 and 2020 runnings.
