Slots Aren't Rigged - Math Beats Us

Why the casino always wins: the math they don’t explain in adverts
Online casino and bookmaker advertising can sound convincing: “RTP 96%”, “high return percentage”, “fair game”. The numbers look attractive – but almost everyone who sits down at slots or roulette loses money in the long run.
Cheated? Robbed? Slots are rigged!
No! Math beat us! Ergodicity and all that stuff. And no conspiracy theories!
Playing around a bit in Excel, a video blogger showed that more than 3/4 of people leave the game with at least less than their starting bankroll. And fewer than 2% pull out 100 times their starting amount. Although on average (across a large set of players) everything follows the formulas.
In the video below a very simple random process is “simulated” in Excel, which draws an unrelenting conclusion: over time everyone will be eaten.
Read on to understand the gist of this video and the described term-phenomenon for a regular casino player.
The trap of a “profitable” game
Imagine a slot with RTP 96%. It seems the casino takes only 4% – sounds almost fair. But here’s what happens in practice: with a $1.50 bet per spin and a pace of 600 spins per hour even a simple calculation by the mean (which, as we’ll show later, is itself misleading) gives a loss of $1.50 × 600 × 4% = $36 per hour. And that’s before the real math kicks in.
But the real problem isn’t even the size of the casino’s edge. The problem is that most players compare their chances to the wrong number (the mean across all players), not to their own result over time. The difference between these two quantities is the key to everything.
Two kinds of averages – why the casino and the player see different numbers
In math and physics there is the concept of ergodicity. A system is called ergodic if the average across a large number of participants at one moment in time matches the average of one participant over a long period. Simply put: what is beneficial for the whole group on average should be beneficial for each individual in the long run. For gambling this condition is not met – and that changes everything. Gambling is a non-ergodic system, and that is crucial for understanding why players lose even in “profitable” games.
Ensemble average
The casino looks at the picture from above: 10,000 players played today – some won, some lost, and in total the house got its 4-5% of turnover. From that vantage point everything is predictable and stable.
Time average
The player is not the casino. The player looks at himself: he came with $500, played 300 spins and his specific balance follows a specific trajectory. In the long run this trajectory for a multiplicative process (where the result of each round depends on the current capital) almost certainly goes down – even if “on average across the ensemble” the numbers look neutral.
Simulation: 10,000 players and one slot
To see this effect in action, let’s run a thought experiment – the same that underlies simulations in ergodicity theory.
Let’s take a conventionally “fair” game: on a win the player’s capital increases by 50%, on a loss it drops by 40%. The expectation: 0.5 × 1.50 + 0.5 × 0.60 = 1.05, i.e. +5% per round. A profitable game!
We launch 10,000 participants for 200 rounds. Here’s what happens:
| Metric | Result |
|---|---|
| Ensemble average capital | Grows, confirming +5% |
| Share of players who remain in profit | ~26-30% |
| Share of players who lost money | ~70-74% |
| Share of total winnings by top 2% participants | >80% of total capital |
The ensemble average grows, but only because a few lucky players hit a huge jackpot and pull the mean up. The typical player – the median player – is in the red. This is called a fat-tailed distribution: rare extreme events skew the mean and make it meaningless for personal decision-making.
Key reasons for non-ergodicity:
- Multiplicative dynamics: losses and wins multiply (not add), so the sequence of outcomes matters – a streak of losses destroys capital irreversibly.
- Gambler’s ruin: with finite capital and an infinite number of rounds any player will inevitably be wiped out.
- Negative expectation: the house edge is built in (house edge), which makes the time-average even more negative.
Additive and multiplicative games. What’s the difference?
Not all gambling games are equally dangerous in a mathematical sense. The difference is determined by the payout mechanics.
An additive game – you bet a fixed amount each time regardless of your current balance. You win $10 – you lose $10. The distribution of outcomes is relatively symmetric, tails are thin, the mean works as a guide.
A multiplicative game – the result of each round is multiplied by the current capital. That’s how roulette behaves when playing a Martingale system and any strategies where bet size is tied to the bankroll. Slots with a fixed stake are formally additive – but more on that below. In multiplicative processes tails are fat, dispersion is huge, and ruin over the long run is mathematically inevitable.
How the game type affects risk
| Process type | Example | Long-run behavior |
|---|---|---|
| Additive | Fixed $1 bet | Predictable losses, thin tails |
| Multiplicative | Martingale, % of bankroll | Fat tails, ruin inevitable |
| Mixed | Sports bets with flat staking | Depends on discipline of bankroll management |
Casinos often promote strategies like “double your bet after a loss” not because they are profitable for the player, but because they turn the game from additive into multiplicative, accelerating the inevitable.
Why players “irrationally” keep playing
Psychologists have for decades explained the propensity to gamble by cognitive biases: illusion of control, availability bias (“I saw my neighbor win a jackpot”), the near-miss effect. All of that is real – but it’s not the whole picture.
Researcher Ole Peters, within so-called ergodic economics, offered another explanation: the player behaves completely rationally – he maximizes what he perceives as his time average. The problem is that the casino methodically substitutes his reference point.
Advertising shows the “ensemble average”: someone won a million, someone hit the jackpot, average RTP 96%. The player unconsciously transfers this to himself – and decides as if he were immediately 10,000 people, not one specific person with a finite bankroll and finite time.
Practical takeaway – how to think about bets and bankroll
Classical expected utility theory (von Neumann-Morgenstern) implicitly assumes ergodicity – it evaluates bets by the ensemble mean. This is the root of the “paradox” of gambling: a game that seems profitable by expected value can in practice lead to ruin. The work of Ole Peters and Murray Gell-Mann showed that people intuitively adjust their behavior to the time average, not the ensemble average.
Professional poker players and sports bettors know: a profitable game kills you with the wrong bet size. That’s why they use the Kelly criterion – a formula that defines the optimal fraction of the bankroll to wager.
If the probability of winning is 60% and the odds are 2.0 (i.e. on a win you get back double your stake, net profit equals 1x), the Kelly criterion recommends betting no more than 20% of your bankroll at a time. Bet more – even with positive expectation you’ll eventually go bust. This is not caution, it’s the math of multiplicative processes.
For casino games, where expectation is negative by definition (house edge from 1% in blackjack to 25% in keno), the Kelly criterion gives an even harsher answer: the optimal bet tends to zero. The only way to “beat” a non-ergodic game with negative expectation is not to play long.
Inequality of winnings is math. Not luck!
The distribution of winnings in casinos always follows the same pattern: the vast majority of players lose money, a few scoop large prizes. This looks like unfairness or a marketing trick – but in reality it’s an inevitable mathematical consequence of multiplicative dynamics with fat tails.
According to gambling math data, the house edge for slots reaches 15%, for keno up to 25%, while for blackjack with correct strategy it drops below 1%. The difference is colossal – but even 1% over a long enough horizon and with multiplicative mechanics leads to inevitable depletion of the bankroll.
“Lucky” players who regularly win large sums generally do not stay ahead over a span of several years. Their stories are used in advertising precisely because they are rare – just as plane crashes make the news, not the millions of safe flights. The ensemble average lies to you again.
Play responsibly!
Gambling is fundamentally a non-ergodic system. Looking at the “average payout” of a group of players and extrapolating it to yourself is a fundamental mistake that casinos and gambling platforms exploit. So accept as a given that for your money there’s only one road in a casino – out of your wallet – and play smart.
Total Articles: 1164
Total Articles: 102
New articles in casino blog
New Questions and Answers
Publication details
| Parameters | Posted on our site |
|---|---|
| Date added | March 31, 2026 at 9:03 AM |
| Last revision | June 7, 2026 at 8:53 PM |
| Views | 1154 |
Article Author
| Parameters | Slots Aren't Rigged - Math Beats Us article prepared and published |
|---|---|
| Published by |
|
| Article Author |
|
The article is complete nonsense.
Author, when you’ve been stuck in the grind for several years flushing hundreds of deposits a month with zero return, I’ll take a look at you and you’ll see for yourself that there’s some rigging going on.
For example, the same Vavada and I can drink vodka until I go blue. Example: Vodka – 32 deps, 3 withdrawals, and the last 11 deps had no bonus games at all, so the article is total bullshit.
Wow, thanks for the guide, really solid
in the long run we’re all in the red 🙁
The article’s great! Even though the math screws us, the casino won’t let you win more than you’re supposed to, so a bit of “rigging”, for me, is present in them anyway
you might as well say that geometry fucks us too, it’s math though
Interesting article to think about, but still, we won’t find out about all those hidden pitfalls.
Math has been screwing me since childhood, and here at the casino it’s not a life but one big 2+2*2.
The casino doesn’t cheat – it just waits.
You can win in the short run, in the long run – the math will take it all.
The longer you play the closer to zero. Always.
Cool, interesting article, especially if you wrote all of this yourself – respect.
Thanks for the helpful article
Very interesting 🙂
Good luck to you and big wins