Florida scratch-off odds, ranked
Every game currently on sale, ranked by what it returns per dollar today rather than at launch. Prize data from September 13, 2026.
Ranked by return, most reliable estimates only
“Returns” is cash back per $1 spent, with annuity prizes discounted to what a winner can actually collect today. It is below $1 on every game, always — that is how lotteries work. The column on the right is how much to trust the number.
| Game | Price | Returns per $1 | Sold | Confidence |
|---|---|---|---|---|
| $50 | $0.86 | 85 % | $0.84–0.87 | |
| $50 | $0.81 | 65 % | $0.80–0.82 | |
| $50 | $0.79 | 40 % | $0.79–0.80 | |
| $50 | $0.79 | 55 % | $0.78–0.80 | |
| $50 | $0.79 | 9 % | $0.78–0.79 | |
| $20 | $0.77 | 42 % | $0.77–0.77 | |
| $20 | $0.76 | 29 % | $0.76–0.77 | |
| $20 | $0.76 | 76 % | $0.75–0.76 | |
| $20 | $0.75 | 76 % | $0.73–0.78 | |
| $10 | $0.75 | 26 % | $0.75–0.75 | |
top prize gone |
$30 | $0.75 | 62 % | $0.75–0.75 |
| $10 | $0.75 | 80 % | $0.75–0.75 | |
| $10 | $0.75 | 49 % | $0.74–0.76 | |
top prize gone |
$20 | $0.74 | 90 % | $0.73–0.76 |
| $10 | $0.74 | 19 % | $0.74–0.75 | |
| $10 | $0.74 | 72 % | $0.72–0.77 | |
| $20 | $0.74 | 74 % | $0.72–0.75 | |
top prize gone |
$20 | $0.73 | 84 % | $0.71–0.76 |
| $10 | $0.73 | 76 % | $0.70–0.76 | |
| $10 | $0.72 | 57 % | $0.70–0.75 | |
| $5 | $0.71 | 60 % | $0.68–0.75 | |
| $5 | $0.71 | 40 % | $0.69–0.72 | |
| $5 | $0.70 | 67 % | $0.68–0.73 | |
| $5 | $0.70 | 84 % | $0.70–0.71 | |
| $5 | $0.70 | 25 % | $0.69–0.71 | |
| $5 | $0.70 | 40 % | $0.68–0.72 | |
| $2 | $0.69 | 61 % | $0.66–0.74 | |
| $5 | $0.69 | 36 % | $0.68–0.71 | |
| $3 | $0.69 | 52 % | $0.65–0.74 | |
| $3 | $0.69 | 64 % | $0.65–0.74 | |
| $5 | $0.69 | 53 % | $0.65–0.73 | |
| $2 | $0.68 | 28 % | $0.67–0.70 | |
| $5 | $0.68 | 63 % | $0.65–0.72 | |
| $2 | $0.68 | 86 % | $0.64–0.73 | |
| $3 | $0.68 | 57 % | $0.64–0.72 | |
top prize gone |
$5 | $0.67 | 79 % | $0.63–0.72 |
| $1 | $0.61 | 24 % | $0.60–0.63 | |
| $1 | $0.61 | 36 % | $0.59–0.63 |
Later in their run — wider margin (16)
Mostly sold. Still worth reading, but the return figure could move by a tenth to a quarter either way, so treat the ordering as a rough guide rather than a ranking.
| Game | Price | Returns per $1 | Sold | Could be |
|---|---|---|---|---|
| $30 | $0.86 | 94 % | $0.77–0.97 | |
| $10 | $0.75 | 94 % | $0.61–0.97 | |
| $5 | $0.71 | 93 % | $0.57–0.93 | |
top prize gone |
$20 | $0.70 | 88 % | $0.58–0.86 |
| $5 | $0.68 | 83 % | $0.58–0.82 | |
| $2 | $0.68 | 74 % | $0.62–0.76 | |
| $5 | $0.67 | 82 % | $0.58–0.81 | |
| $2 | $0.67 | 73 % | $0.60–0.75 | |
| $3 | $0.66 | 68 % | $0.60–0.74 | |
| $2 | $0.65 | 68 % | $0.57–0.75 | |
| $2 | $0.65 | 84 % | $0.54–0.80 | |
| $2 | $0.63 | 86 % | $0.53–0.78 | |
top prize gone |
$5 | $0.63 | 83 % | $0.51–0.83 |
| $2 | $0.62 | 79 % | $0.52–0.78 | |
top prize gone |
$5 | $0.61 | 90 % | $0.51–0.78 |
| $1 | $0.59 | 66 % | $0.52–0.66 |
Nearly sold out — not rankable (11)
These average over 90% sold. So few tickets remain that the estimate of how many is left swings by more than a quarter, and the return figure swings with it. Several of them would top the ranking above if we sorted on the number alone — which is exactly what every competing app does, and why their top picks are the ones you should trust least.
| Game | Price | Returns per $1 | Sold | Could be |
|---|---|---|---|---|
| $5 | $0.71 | 92 % | $0.53–1.06 | |
| $5 | $0.71 | 90 % | $0.53–1.07 | |
| $2 | $0.69 | 88 % | $0.54–0.94 | |
| $5 | $0.67 | 89 % | $0.53–0.90 | |
top prize gone |
$20 | $0.67 | 97 % | $0.51–0.99 |
| $5 | $0.65 | 83 % | $0.49–0.98 | |
top prize gone |
$2 | $0.62 | 87 % | $0.50–0.83 |
top prize gone |
$10 | $0.61 | 94 % | $0.42–1.11 |
| $2 | $0.60 | 82 % | $0.46–0.85 | |
| $2 | $0.60 | 84 % | $0.47–0.84 | |
| $5 | $0.57 | 94 % | $0.28–2.30 |
Where these numbers come from
Florida publishes, for every scratch-off game, the ticket price, the odds at launch, and how many prizes remain at each level. Tickets printed is prizes printed multiplied by the launch odds, so what is left can be estimated from what has been claimed.
That estimate is the weak link, and it is the one every competitor hides. States publish prizes claimed; none publishes tickets sold. We infer it from the small prize tiers, which are cashed at the counter within days, rather than pooling every tier — large prizes must be claimed at a lottery office and lag by weeks, which makes a game look less sold than it is. Then we publish the range, because a single number here would be false precision.