Pokémon Pull Rates Explained

A Pokémon pull rate is an estimate of how often a card, or a type of card, turns up per pack, worked out by counting what came out of a known number of opened packs. The official product pages for current English sets say what a pack contains but give no odds, so every pull rate you see is somebody's count and is only as good as the packs behind it. This guide explains how to read those counts, how far they can wobble, and what the arithmetic can and cannot tell you before you open anything.

What a pull rate actually measures

A pull rate is a frequency: the number of times an event happened divided by the number of packs opened. Before the number means anything, you need to know exactly which event was counted. "Pulled a special illustration rare" and "pulled the special illustration rare of one named Pokémon" are different events, and the second is far less common than the first.

The first is a rarity-category event. The official rarity guide on the Pokémon TCG site lists the categories in use now, including double rare, ultra rare, illustration rare and special illustration rare, plus mega attack rare and mega hyper rare, which feature Mega Evolution Pokémon ex. You can learn to tell these apart from the symbol in our rarity symbols guide.

The second is a specific-card event. A rate for a whole category does not tell you the rate for one card inside it. Dividing the category rate by the number of cards in the category only works if every card in that category is equally likely, and nobody outside the printer can confirm that.

Check the unit too. "One in 20 packs" and "1.8 per box" describe the same kind of thing at different scales, and a figure quoted per box only makes sense if you also know how many packs the box holds.

Read the sample behind an estimate

We do not yet publish a documented sample of our own, so this guide gives no set-specific rates. What follows is the standard we would expect any published estimate to meet, ours included. If an estimate does not show these details, treat it as an anecdote.

The last row matters more than it looks. Openings that get posted and shared tend to be the lucky ones, so a figure built from collected highlights will lean high.

Why estimates vary

Even if every pack were filled by a perfectly fair random process, two people opening the same number of packs would get different counts. That is sampling variation, and a small sample can produce a figure well away from the true rate without anyone doing anything wrong.

Statisticians handle this with a confidence interval, a range that the true rate plausibly sits inside given what was observed. The table below uses the Wilson method for a proportion, as set out in the NIST/SEMATECH handbook of statistical methods. The hit counts are invented to show the effect of sample size; they are not counts from any set.

The first three rows show the same 5%, but only the third pins it down. A one-box sample of 36 packs cannot tell a rate of one in ten from one in fifty with any confidence. The last row shows that a box with no hits is not evidence of a rate of zero; by the same method the true rate could still be close to one in ten.

Estimates also differ because people measure different things. One count pools English and Japanese packs, another counts only booster displays, another counts reverse holos of rares as hits. You will also read claims that boxes are packed to guarantee a set number of big hits. The official product pages we checked make no such statement, so do not plan around one.

Use a clearly labelled probability example

This example uses an assumed rate of one hit in every 20 packs (5%). That number is made up for the arithmetic. It is not the rate for any set or any rarity. Swap in a figure from a documented sample if you have one.

It also assumes every pack is independent, meaning what came out of one pack tells you nothing about the next. If packs in a box are sorted in some way, that assumption fails, and the real chances could be higher or lower than shown. With those assumptions, the chance of at least one hit in a number of packs is 1 minus (0.95 multiplied by itself once for each pack).

The pack counts in the second column come from the official product listings for Phantasmal Flames: six in a booster bundle, nine in an Elite Trainer Box (11 in the Pokémon Center edition), and 36 in a booster display. Our product formats guide covers what else comes in each.

Even 72 packs at this assumed rate is not a certainty. A specific card is a much longer shot. If the same assumed category held 20 cards that were equally likely (another assumption), one named card would come up about once in 400 packs, and a full 36-pack display would give roughly an 8.6% chance of seeing it. The arithmetic describes chances and nothing else. It is not a reason to open more packs.

Set-specific rates and official contents

Official statements and opening counts are different kinds of fact, and they should never be blended. The official product listings for Phantasmal Flames, Perfect Order and Chaos Rising each say that every booster pack holds 10 cards, 1 Basic Energy and 1 Pokémon TCG Live code card, and each adds that cards vary by pack. The official rarity guide says every pack contains at least one common, uncommon or rare card. That is the full extent of what the publisher states about pack contents on those pages.

Everything else you read about a set, such as how often its special illustration rares appear, comes from people opening packs and counting. That can be useful, provided the sample is shown in the way described above. It is an observation, not a promise.

For what is actually in a set, start with its contents, for example our Phantasmal Flames guide. Knowing how many cards sit in each rarity tells you how thinly a category rate is spread across individual cards.

Our set pages bring together what we know about each set. If we publish a pull-rate estimate for any set, it will come with the sample table above filled in.

Choosing packs or a specific single

If there is one card you want, buying that card as a single is the only way to be sure of getting it. The probability example shows why: even a full display gives a modest chance at one named card under generous assumptions. You can choose the exact card and see its condition before you commit, and our singles section is where we will list them when we open.

Packs make sense when the opening is the point, or when you want a spread of a set's cards rather than one in particular. Go in expecting most packs to hold nothing you were hoping for, and treat a hit as a bonus. Our booster packs page is where we will stock loose packs from current sets.

You can do both: open a modest number of packs for the fun of it, then fill the gaps in a set with singles, rather than opening until the last card turns up, which the arithmetic above shows can take a very long time.

Our chase card guides, such as the one for Pitch Black, rank the cards people open packs for without quoting odds, for the reasons above.

If you are weighing up which sets and products to open, our guide to the best Pokémon packs to buy is the next place to look.