Four numbers in.
Every stake out.
Tell it your bankroll, how many bets you are planning, how many you expect to win, and the odds you are getting. It sizes every bet so that hitting your win count finishes the run in profit — whichever bets happen to land.
Watch instead — 2:26
The whole method, in two and a half minutes
Every figure on screen is computed by the calculator's own engine.
Order stops mattering
Win the first two or the last two — same finish. That is the whole trick.
Stakes move for you
Small early, larger as bets run out, smaller again once a win is banked.
Losses are priced in
The plan assumed them from bet one. A bad start is not a broken plan.
Same bets, same win rate, different result
Five bets, two winners. Bet the same amount each time and your result depends on when the winners arrive. Two early wins then three losses can still leave you down — the losses ate more than the wins brought in.
Same five bets, same two winners, and the finish is identical in every order. Each stake is sized against what is still left to do, so a late winner is worth exactly as much as an early one.
At every step the engine asks one question: given what has happened, what stake keeps the original target exactly reachable?
Your opening stake is $53.70. If you are used to flat betting that feels small — it is meant to. Six bets ahead, three chances to hit the target, no reason to risk more yet.
| Bet | Stake | Result | Bankroll |
|---|---|---|---|
| Bet 1 | $53.70 | loss | $146.30 |
| Bet 2 | $59.07 | win | $217.18 |
| Bet 3 | $64.97 | loss | $152.21 |
| Bet 4 | $79.42 | win | $247.52 |
| Bet 5 | $87.36 | win | $352.35 |
One equation, solved before you start
Call T[n][k] the guaranteed finish when n bets remain and
k wins are still needed. Two cases are free:
T[n][0] = 1
Target met. Nothing left to risk.
T[n][n] = on
Every bet must win, so every bet takes the lot.
Everything else comes from making both branches land in the same place:
(1 + f(o−1)) × T[n−1][k−1](1 − f) × T[n−1][k]Solve for the stake fraction f:
f = ( T[n−1][k] − T[n−1][k−1] ) ÷ ( T[n−1][k−1] × (o−1) + T[n−1][k] )
Your stake is that fraction of your current bankroll. The table is built once at setup, so every later recalculation is a lookup — which is why the next stake appears the instant you tap a result.
If you reach a point where every remaining bet must win, the formula puts your whole bankroll on it. That is not a bug — if every bet left has to land, no amount of spreading keeps the target alive. A safety cap is the only thing that changes it.
Your average was a guess. The market is not.
You planned at 2.20. The first selection comes in at 1.85, the next at 2.40. A calculator that ignored that would be wrong by bet one.
The reference table stays fixed at your original average — that table is the target you committed to. Only the immediate stake is recalculated at the real price. Re-averaging after every leg would quietly move the goalposts; this way the plan you agreed to stays the plan.
What the safety cap really costs
Uncapped, a Masaniello cycle has exactly two endings. There is nothing in between, because the all-in rule means a failed run always finishes at zero.
It does not change what the cycle is worth. At fair odds both versions carry the same expected value — 1.0000 and 1.0000 times your bankroll. Anyone claiming a staking rule improves your expectation is selling something.
Fifty per cent is a sensible default: it rarely interferes with a healthy run, and it stops a bad stretch consuming everything before the cycle has had a chance.
A staking plan cannot make a bad bet good
That break-even rate is 45.5% — which is exactly 1 ÷ 2.20, the rate the odds themselves imply. Not a coincidence, and not specific to this cycle: it comes out the same for every combination you can enter, capped or not.
So the only number that decides whether you make money is the gap between your real win rate and 45.5%. That gap is the edge your selections supply. What Masaniello adds is making sure that when the edge is real, the order your winners arrive in cannot rob you of it.
Five ways people break their own plan
Nudging the first stake up
It looks too small, so you add a bit. Every stake after it was calculated assuming you did not.
Moving the target mid-run
"One more win will do." The guarantee rests entirely on the target you set at the start.
Entering odds you wish you had
Feed it the real price and it manages your risk. Feed it fiction and it just hides the risk from you.
Turning the cap off while winning
Protection you switch on after the damage starts was never protecting anything.
Chasing outside the system
The cycle ends badly, and the next few bets go out with no plan at all. This is the one that does real damage.
| Method | Built for | Fails when |
|---|---|---|
| Masaniello | A fixed run with a win target. Losses expected and priced in from bet one. | You miss the win count — then, uncapped, the bankroll is gone. |
| Kelly | Endless independent bets, each sized to your estimated edge. | Your edge estimate is wrong, which it usually is. |
| Martingale | Doubling after a loss until a win recovers everything. | A streak runs slightly long — and it always eventually does. No ceiling exists. |
| Rollover | Chaining wins toward a big number. | Any single leg fails. No partial credit for getting close. |
| Flat | Simplicity. It genuinely cannot blow up. | Nothing — but it does not care when your winners land, which is the problem Masaniello exists to solve. |
Quick answers
Is it free?
Yes, permanently, and no account is needed to use it. An account only exists so your cycles follow you between devices.
Must every bet match my average odds?
No. Enter each real price when you record the result. That stake is recalculated at the real number; the target stays where you set it.
What if I lose more than expected?
The moment more wins are needed than bets remain, the cycle closes rather than running on false hope. With a cap you keep whatever it preserved.
What happens to a postponed match?
Record it as a void. Your stake comes back and it does not use up one of your planned bets — the selection simply never happened.
Only football?
Any sport. The maths only cares about bet count, win count and decimal odds.
Why can't expected wins equal total bets?
Then every bet must win, which is just an accumulator with your whole bankroll on each leg. There is no staking decision left to make.
Does it guarantee a profit?
No — be wary of anyone who says otherwise. It guarantees a specific finish if you hit your win target. Whether you hit it is down to your selections, which is why the break-even rate is shown before you commit.