Odds & Math
How Pack Odds Work — and How to Read Ours
Every pack product — retail booster, repack, mystery box, live break — is a probability distribution wearing foil. The wrapper is marketing; the distribution is the product. If you cannot see the distribution, you are not buying a pack, you are buying a story. Here is how to read the numbers when a seller actually publishes them.
A "1 in N" line is a per-slot statement, not a promise
The numbers in this section are a worked example, not Waffles odds. Ours are computed live from the cards actually left in the packs and published in full on the odds page — they move when inventory moves, which is precisely why they are not retyped here.
Suppose an odds table says the $100+ hit is 1 in 200. That means each purchase has a 1/200 chance — 0.5% — of that outcome. It does not mean the 200th pack pays out, and it does not mean 200 packs guarantee a hit. Independent trials do not remember each other. Across 200 packs at 1/200, the chance of at least one hit is about 63% — and that ≈63% holds for any reasonably large N, because it is just 1 − (1 − 1/N)N converging on 1 − 1/e. Not 100%. Anyone who has chased a retail case for a specific alt-art already knows this feeling; the math just formalizes it.
The second thing to check is what the odds are attached to. A pack with one "hit slot" and three filler slots has very different economics from "every card could be anything." On Waffles, each pack is exactly 4 cards: one hit slot drawn from published value bands, and the rest fitted to a value target. That structure is printed on the odds page because without it, a headline number is uninterpretable.
Expected value: the number most sellers won't print
Expected value (EV) is the average market value a pack returns over many rips. Every pack product ever sold has an EV below its price — retail boosters included. Wizards and The Pokémon Company do not sell $4 packs containing $4 of average singles value; the gap funds the game, the shops and the fun. Repacks and break sites are no different, except most of them hide the number.
We print ours: roughly 0.9× the pack price, every tier. Buy a $50 pack and, on average across many packs, you will open about $45 of market value. The remaining margin is the business, stated plainly. If a seller implies their packs return more than they cost, one of three things is true: the values are inflated, the odds are wrong, or the business is losing money on purpose. None of those age well.
Variance is the actual product
An EV of 0.9× does not mean every pack returns 0.9×. Most packs land below the average and the occasional big hit pulls the mean up — that is what a right-skewed distribution looks like, and it is precisely why ripping is fun. The honest framing: you are paying a known margin for variance. That is fine — people happily pay for variance at every card shop on earth — but you should know the shape of the distribution you are buying. Rare, transparent long shots (our God Pack is a flat 1 in ??? in the app, rolled once per purchase at every tier, with the real figure printed on the odds page) belong in the published table too, not in fine print.
Questions any pack seller should be able to answer
- What exactly is in the pool right now? A list with market values, not vibes.
- What are the per-band odds for my tier? Computed from live stock, since odds change as chases are pulled.
- What happens to stock when a card is pulled? One copy should mean one copy.
- What is the EV relative to price? A number, not "great value!"
If the answer to any of these is silence, the distribution is the secret — and the secret is never in your favor. We built Waffles so all four answers are one click away. Every figure is computed from the stock the dealer actually draws from, and a band with nothing behind it says so rather than quietly redistributing itself. Go check it.