Staking
Why there is no Martingale here.
Doubling your stake after every loss feels like it cannot fail: one winner gets it all back. The record shows what it would actually have cost.
Longest losing run
33
losing bets in a row, to
Next bet, doubling
8,589,934,592u
€85,899,345,920 at €10 a unit
Bank needed
17,179,869,183u
€171,798,691,830 to survive the run
The arithmetic
Start at one unit and double after each loss. After 33 losers you have staked 8,589,934,591 units and lost them all, and the next bet is 8,589,934,592 units. To place it you needed 17,179,869,183 units of bank before the run started, and every exchange and bookmaker has a maximum stake that you would have hit long before.
That run is in the record: 1 August 2026 to 2 September 2026 · 782 bets in 349 races · 33 days. It is not a thought experiment. Over a year of racing the longest run gets longer, not shorter, and each extra loser doubles the number again.
What AXIOM does instead
The same stake on every bet: 2.5u. During that run of 33 losers the strategy lost 82.50u and carried on. The largest drawdown in the whole record is 98.83u, or 39.5 stakes.
A staking plan cannot turn a losing strategy into a winning one. It can only decide how a run of losers is paid for. Level stakes pay for it slowly and survivably; a Martingale pays for it all at once, usually with money you do not have.
Why it is not offered as an option
A tool that offers Martingale is selling a trap that looks like a setting. If someone loses their bank to a plan a product handed them, the product is responsible. So it is not in the simulator, not in the dashboard, and not in any staking plan Furlong Ledger describes. What is offered is level stakes, a fixed amount, a percentage of bank, and Kelly with its warning.
This page is about staking, not about results. Nothing here promises a return, and past results do not indicate future results.