How It Works

How the Rankings Work

Every active game gets a single Hot Pick Score, recalculated with every data refresh, and the homepage list is ranked by that score — highest first. That score is also what shows up as the Rank badge on every card (e.g. "#4"). The only thing you filter manually is ticket price (and, optionally, Jackpot Overlay games — more on that below). The top 2 scoring games statewide are pinned above the list as today's Hot Picks.

A game is excluded from scoring entirely — sorted to the bottom of the list, ranked there by Return Rate alone, and shown as "—" for Rank — if any of the following is true:

This keeps the rankings honest: a game that's technically still active but effectively played out doesn't get to rank above one that genuinely has something left to chase.

Goal-Based Rankings

The Hot Pick Score is a good default for "what's the best game overall," but not every player wants the same thing — some are chasing one big jackpot, some just don't want to walk away empty-handed. So every game's own page shows how it ranks across seven different goals, each one a straight, single-purpose ranking against every other active game (no blending, no weighting) using the same underlying odds math as everything else on this page.

A game can be ranked out of a different total in each category — "Jackpot Watch" only ranks games with a $50,000+ top prize still available, for instance, so a $1 game with a $500 top prize simply isn't part of that particular ranking. Where a game doesn't qualify for a category at all, its card says "Not eligible" and why, rather than showing a misleadingly low rank number.

Big Swing and Steady Cash only consider prize tiers with at least 20 prizes printed at launch — see the note on tier-size weighting in the Return Rate section below for why that floor exists. Jackpot Watch is the one exception: top-prize tiers are almost always small by design (a handful of prizes is normal for a $1M+ payout), so we don't apply that floor there — it uses the exact same math as the Jackpot Overlay badge described further down this page.

Overall Odds

This is the odds printed on the back of the ticket — like "1 in 3.39" — fixed permanently on the day the game launched. It's the Florida Lottery's own published figure and never changes, no matter how many prizes get claimed over the game's life. Because it's static and doesn't reflect anything about what's actually left to win, we don't show it on the homepage cards anymore — you'll find it on each game's own page instead, next to the "Current odds" figure that actually is recalculated from what's left in circulation.

We still color-code it (green / yellow / red) as a quick reference, but treat it as a starting point, not a live read on today's odds. For that, see the "Current odds ↓" column in the full prize breakdown table on each game's page — that figure is recalculated from what's actually left in circulation, tier by tier.

Return Rate

Return Rate answers a simple question: if you bought a ticket for this game today, how much would you expect back on average? Unlike Overall Odds, this is fully dynamic — it's recalculated every data refresh based on what's actually left in the prize pool right now, not what was printed at launch.

The formula has two parts. First, we estimate how many tickets are still in circulation. For every prize tier still paying out, we multiply that tier's own printed odds by how many of that tier's prizes remain — that gives one independent estimate of the ticket count per tier. Rather than treating every tier's estimate as equally trustworthy, we weight each one by how many prizes that tier originally had. A tier printed with only a handful of prizes is reconstructing the ticket count from a tiny sample and swings wildly on a single redemption; a tier with tens of thousands of prizes barely moves. Weighting by original prize count is the statistically correct way to combine several noisy readings of the same unknown number, and it's why a small handful of tiny prize tiers can no longer quietly drag the whole estimate around. We also exclude any tier with fewer than 20 prizes printed at launch entirely, rather than just down-weighting it, so it can never single-handedly move the number. Then:

where the remaining prize pool is the sum, across every tier, of that tier's prize amount times however many of that tier are still unclaimed.

We cap the displayed number just under 100% as a sanity ceiling — no Florida scratch-off is designed to pay back more than it takes in, on average, across its full run. On rare occasion the raw math computes higher than that for a specific game at a specific moment (see Jackpot Overlay below for exactly when and why) — the display cap keeps the headline percentage meaningful and comparable across every other game, while the Overlay stat surfaces that specific situation honestly instead of hiding it.

Jackpot Overlay

Every tier in a scratch-off game gets claimed at roughly the same pace on average, but not always exactly the same pace. Occasionally the smaller, common prizes get bought out much faster than the game's single top prize, which can sit unclaimed while the rest of the game empties out around it. When that happens, whatever tickets are left are statistically more concentrated with that unclaimed jackpot than they were on day one.

We measure this directly by comparing the top prize's current implied odds — estimated tickets remaining ÷ top prizes remaining — against its printed odds at launch:

We only show the badge when a game clears two conditions: at least one top prize must still be unclaimed (a played-out top tier has nothing to overlay), and the ratio must be 3.0x or higher. Below that threshold, the gap is close enough to normal estimation noise that it's not a meaningful signal — most games sit somewhere between 0.8x and 2x just from ordinary variation in how tiers deplete.

You can filter the rankings to Jackpot Overlay games only using the "🎯 Jackpot Overlay" chip on the homepage. It's usually a very short list — that's by design. A rare, genuine overlay is far more useful to know about than a long list of marginal ones.

One honest caveat: this is a statewide, population-level estimate, not a location guarantee. Tickets aren't perfectly shuffled evenly across every retailer — some stores may already be sold out of a given game, others may be holding a disproportionate share of what's left. An overlay ratio tells you the odds are better across all remaining tickets in the state; it doesn't tell you which store has them.

Frequently Asked Questions