Rake Is Not RTP, and 5% Is Not 95%
Why the two numbers are not interchangeable, and what a rake actually costs you over a hundred duels.
The two numbers measure different objects
Return to player is a property of a machine. You put money in, the machine decides an outcome against a paytable, and over a very large number of spins it returns some proportion of what it took. A 95% RTP slot is a statement about the paytable: the machine is built to keep 5 cents of every dollar that passes through it, and the player has no way to alter that.
A rake is a fee on a pot. Two players stake the same amount, the stakes form a pot, and the house takes a percentage of that pot before paying the winner. Nothing about a rake describes what proportion of your money comes back, because the house is not your counterparty. Your opponent is.
The distinction matters because the numbers look comparable and are not. A 5% rake and a 95% RTP are not two ways of saying the same thing, and treating them as equivalent will make you badly misjudge what a session costs.
What actually happens to a Blinko pot
Take the smallest tier, as set out in what playing actually costs. You stake 10.00 USDT. Your opponent stakes 10.00 USDT. The pot is 20.00 USDT. The house takes 5% of the pot, which is 1.00 USDT, and the winner receives 19.00 USDT.
Run that through both sides of the table. The winner put in 10.00 and got back 19.00, a profit of 9.00. The loser put in 10.00 and got back nothing. Between the two of you, 20.00 went in and 19.00 came out. The missing 1.00 is the rake, and it left the table regardless of which of you won.
There is no paytable here and no house edge on the draw itself. The blast point is not weighted against you, and you can check that for any round. What is weighted against you is arithmetic: you are risking 10.00 to win 9.00.
Winning half your duels is not breaking even
This is the part that surprises people, and it is the single most useful number on this site.
If two evenly matched players duel repeatedly, each wins about half the time. But a win returns 9.00 and a loss costs 10.00. Over a hundred duels at 50/50, you win fifty times for 450.00 and lose fifty times for 500.00. You are down 50.00 having played exactly as well as your opponent.
Solve for the win rate that actually breaks even and you get w × 9.00 = (1 − w) × 10.00, which gives w = 10/19, or 52.63%. That is not a rounding artefact. To leave the table where you started, you must win more than half your duels — not half.
The general form is w = 1 / (2(1 − r)) where r is the rake as a fraction. At a 5% rake that is 52.63%. At 10% it would be 55.56%. The rake does not change your odds of winning a round; it changes what a fair win rate has to be.
Why you cannot convert one into the other
People sometimes reason that a 5% rake is 'like' a 95% RTP, so a hundred duels at 10.00 should cost about 50.00. That happens to land close for an evenly matched player, and the coincidence is misleading.
The rake is charged per settled pot, not per unit staked. Your exposure depends on how many rounds you play and at what tier, not on some proportion of a total that flows back. A player who wins 60% of duels is profitable at a 5% rake and would still be losing on a 95% RTP machine, because on the machine there is no opponent to be better than.
That is the real asymmetry. On a machine, skill is irrelevant and the return is fixed. In a duel, skill moves your win rate, and the rake sets the bar that skill has to clear. Whether you clear it is an open question. Whether the rake is charged is not.
What this means for a session
For an evenly matched player, the expected cost of a duel is half the rake on that pot, because you pay the whole rake on the pots you win and none on the pots you lose, and you win about half. At the 10.00 tier that is 0.50 per duel: a hundred duels costs about 50.00 on average.
Playing longer does not recover it. Each additional duel adds another 0.50 of expected cost. The distribution around that average is wide — you can be up several hundred after a hundred duels, or down considerably more than 50.00 — but the average drifts one way only.
None of this makes the game unfair — see where blast points actually land. The blast point is honest and you can check it yourself. It makes the game expensive, which is a different property, and one worth knowing before you decide how long to sit down for.