How Long Does $100 Last on Slots? We Simulated 450,000 Sessions at Identical RTP

“96% RTP” tells you almost nothing about your evening. We built three slot models with exactly the same 96.00% RTP, differing only in volatility, and ran 450,000 sessions from a $100 bankroll.

The gap is enormous.

Same payback, 16x the session length

At $1 per spin — 100 spins’ worth of bankroll:

Model (all 96% RTP)	Median spins survived	Still playing at 500 spins	Ever doubled to $200	Ahead at 500 spins
Low volatility	2,315	100%	0.00%	12.4%
Medium volatility	745	63.2%	35.29%	35.9%
High volatility	147	11.2%	11.21%	10.9%

A median of 2,315 spins versus 147. Identical payback percentage; the only difference is how the returns are distributed.

This is the number RTP cannot tell you. RTP is the long-run price. Volatility decides how quickly you reach the long run — and whether your money survives to get there.

You cannot have longevity and a jackpot

The trade-off is absolute in our data:

• The low-volatility model doubled a $100 bankroll zero times in 50,000 sessions. Not rarely. Never. Its biggest multiplier is 5x, so there is no mechanical path from $100 to $200 before the 4% edge grinds it down.
• The high-volatility model pays up to 5,000x and doubled the bankroll 11.21% of the time — but half of its sessions were over inside 147 spins.

If a game lets you win big, it must take your money faster on the way. That is not a design choice a casino makes to be cruel; it is arithmetic. The money for the rare 5,000x comes from the 87.36% of spins that return nothing.

The counterintuitive part: medium beat high

Medium volatility doubled the bankroll far more often than high volatility — 35.29% versus 11.21%.

The high-variance model has the bigger prizes and still won the race less often, because it usually ran out of bankroll before a big multiplier arrived. Extreme variance is not the same thing as a good chance of a big win. With $100 at $1 a spin, you are not buying enough spins to reach the tail.

The lever you actually control

Bet size dominates everything you can change:

Median spins survived	$0.20/spin	$1/spin	$2/spin
Low volatility	5,000+ (capped)	2,315	1,071
Medium volatility	5,000+ (capped)	745	232
High volatility	1,006	147	66

Halving your stake more than doubles your session. At $0.20 a spin, both the low and medium models simply did not run out within our 5,000-spin cap.

What it does not do is make you money. At 96% RTP you pay about 4% of everything you wager, at any stake. Smaller bets buy time, not profit. Our house edge statistics put that 4% next to every other game on the floor.

The 97.95% “hit rate” that still loses

The low-volatility model returns something on 97.95% of spins. It also loses money on almost every one of them, because most of those returns are 0.5x — half your stake back, presented with the same animation as a win.

This is worth naming plainly: a “hit” is not a win. A game can light up on nearly every spin and still be taking 4% of your turnover. When a slot advertises how often it pays, check what it pays, not how often.

What this does not say

• These are models, not specific slots. All three multiplier tables are published below with their probabilities, and each is asserted to exactly 96.00% RTP in code. Real titles differ; the point is the mechanism, not a claim about any particular game.
• No bonus rounds, free spins or features are modelled. Real slots concentrate a large share of their RTP in features, which raises effective volatility above a plain spin model.
• Sessions are capped at 5,000 spins. Where a median hits 5,000, the true median is higher — we report it as capped rather than guessing.
• These are distributions, not predictions. Half of high-volatility sessions ended before 147 spins; the other half went longer, and 11.21% doubled. Your night is a draw from this, not the average of it.

Methodology (reproduce it yourself)

• Environment: Python 3, standard library only (random).
• Seed: random.seed(20260804) — fixed, so every figure is reproducible.
• All models scaled to RTP exactly 96.00%, asserted in code; the 0x bucket absorbs the remaining probability.
  • Low volatility — ×0.5 p=0.30, ×1 p=0.20, ×2 p=0.09, ×5 p=0.012 (before scaling). Hit rate 97.95%, variance 0.60.
  • Medium volatility — ×0.5 p=0.18, ×1 p=0.09, ×2 p=0.055, ×5 p=0.022, ×20 p=0.0065, ×100 p=0.0012. Hit rate 52.39%, variance 21.98.
  • High volatility — ×0.5 p=0.06, ×2 p=0.035, ×10 p=0.010, ×50 p=0.0030, ×500 p=0.00055, ×5000 p=0.00004. Hit rate 12.64%, variance 1,332.79.
• Session: $100 bankroll, flat bets of $0.20 / $1.00 / $2.00, ends when the balance cannot cover the next spin or at 5,000 spins.
• Sample: 50,000 sessions per model per bet size — 450,000 total.
• No cherry-picking: single runs at the stated seed, reported in full.

Volatility is the part of a slot nobody puts on the label, and it decides almost everything about how your session feels. It does not decide what it costs — that number is the house edge, and it is the same whether you last 147 spins or 2,315. Play only what you can afford to lose, where it is legal for you.