There is a regular at your local card shop who walks out every week with a box under their arm, and a month later you see them selling singles online for more than the box cost. You collected these cards as a kid. You know the sets better than they do. So why are they the one getting paid?
The gap is method. They sort at the kitchen table into two piles: the cards worth listing and the cards worth nothing. They know when a clean card is worth sending away to be graded, and when grading is an expensive way to lose money.
Meanwhile your paycheck stretches less every month. The weekend shop costs more. The job asks for more hours and gives back the same. A business built on knowledge you already have can feel like the one part of your week that belongs to you.
Move with a plan here, and move carefully. Supply is huge, with around 10 billion cards printed a year, so scarcity is something you check and never assume. The edge belongs to whoever learns pricing and grading properly while newcomers are still buying on hype.
Getting started runs from $50 at the low end to $5,000 at the high end. A working month lands anywhere between $100 and $12,000, and it takes 1-6 months before the money coming in covers the money you put out. Start at the bottom of that range and let your records earn the right to spend more.
Today, pick one set you already know well, look up what its singles have actually sold for in the last month, and write those numbers down.
The Pokemon Company reported record net sales of 531 billion yen for the fiscal year ended February 2026, up 29.3% year on year, which is roughly 3.3 billion dollars at the reference rate of 159.39 yen to the dollar on 27 August 2026. Pokemon accounts for more than 68% of hobby retail card sales in the United States and is the top-selling card brand in the world by unit volume.
Grading has grown faster still. PSA graded 3.5 million collectibles in June 2026, a single-month record, of which roughly 2.5 million were cards, up about 74% year on year. The company reportedly runs a backlog of more than 10 million cards, and on 2 June 2026 it temporarily stopped accepting new submissions across four of its most popular service tiers to work that backlog down.
Now the number that should shape your plans more than any of those. As of March 2026, more than 85 billion Pokemon cards have been printed in total, up from around 75 billion a year earlier. Roughly 10 billion new cards entered the world in twelve months.
Hold those together. Enormous demand, record grading volumes, and a manufacturer printing at a rate that makes almost any modern card permanently common. Put side by side, they show you where the money is, and where it is missing.
The money here comes from operations. Speculation is where people lose it. The people who do well are running a small trading business with inventory, throughput and margins. The people who do badly are buying things they hope will go up.
Which of those two people have you been so far, if you have bought cards before? Your honest answer tells you which habits this page needs to change first.
Five separate models sit under the one label, and they need different amounts of capital, different skills, and carry very different risk.
Most guides treat these as one activity. They are five different businesses, and choosing which one you are running is your first decision.
Ten billion new cards a year is the single most important fact on this page.
Scarcity is what makes a collectible rise in value. Modern sets are printed to meet demand, and when demand rises the manufacturer prints more, which is exactly what a well-run company should do, and exactly what stops the product from ever becoming scarce. The Pokemon Company's own results show demand outrunning supply in the short term while total production rises by ten billion units a year. Short-term shortages inside a supply that is abundant by design are a very poor basis for a long hold.
That is why vintage and modern behave so differently. A 1999 card is scarce because the print run ended and most copies were played with, bent and thrown away. A 2026 card will stay common, because tens of millions of copies sit in protective sleeves owned by people who know exactly what they have.
The practical rules that follow are unglamorous.
Treat modern sealed product as inventory to sell. Holding it as an asset is the mistake. Turnover is the business.
Be very sceptical of anyone claiming a current set is a long-term investment. The people making that claim are often the ones selling the set.
If you want real scarcity, you will find it in graded vintage in high grades, and that market needs real capital and real expertise, which makes it a different business from the one most beginners can start.
Is there a sealed box on your shelf right now that you are "holding"? Ask yourself what it would sell for today, and whether you would buy it again at that price.
If you are planning grading arbitrage in 2026, plan around the queue, and leave the theory aside.
A backlog reported at more than 10 million cards, a record 3.5 million items graded in a single month, and four popular service tiers temporarily closed to new submissions from 2 June 2026 mean three things for your business.
None of that makes grading a bad idea. It means grading is slow and capital-heavy, and you should fund it with money you can leave idle. The people who get hurt are the ones who buy heavily expecting a fast turnaround.
Check current service levels, prices and estimated turnaround times directly with the grading company before you commit money, because these change often, and any figure quoted in a guide, this one included, ages quickly.
Run the sum before you submit anything, because most cards fail it.
For a card to be worth grading, the graded price must beat the raw price plus the grading fee plus shipping both ways plus insurance plus the selling fees on the eventual sale. And it must beat them by enough to pay you for months of waiting and for the risk of a lower grade than you expected.
The grade distribution is what hurts people. A card worth a great deal in the top grade is often worth barely more than raw one grade down. If you expect a top grade and get the one below it, your whole margin can vanish, and the fee is never refunded for a disappointing grade.
Three things follow for you.
Take the best card you own and run that sum on paper tonight, using the grade below the one you hope for. Does it still come out ahead? If not, it stays raw.
When the sum works, it feels wonderful. One well-judged card can pay for a weekend away or a month of car insurance. A dozen of them a year, each one checked against the lower grade first, can make this hobby fund itself and then some. That is what careful arithmetic buys you.
Now that you can work out what grading costs you in fees and waiting, the earlier decision matters more than it did a moment ago. What you paid in the first place decides whether any of that arithmetic comes out in your favour.
Everyone focuses on which cards to buy. The people who make money focus on where they buy.
Retail arbitrage. Buying new product at retail when it is in stock. Thin margins and heavy competition, including from people using automation.
Tournaments and local game stores. Built on relationships, and the source most likely to bring you repeat deals.
Notice that four of those five reward being there in person and doing unglamorous work. That is the lasting edge in this business, and it is why it resists automation: software can tell you a card's price instantly, and it cannot drive to a house and sort a shoebox.
Who do you already know who has a box of old cards in a loft or a cupboard? That first collection often comes from one conversation with someone in your own circle.
Those unglamorous afternoons have a very concrete payoff. If sorting bulk from a house clearance and selling it in lots leaves you 100 a month after what each lot cost, that is the football boots your kid asked for, paid for by a shoebox someone else was throwing away. Small money with a name on it is the right way to start.
Selling, and the fees that decide your margin
So you have bought well and you know what you are holding. What happens between the listing going up and the money landing in your account is the part that quietly takes a share of it.
Where you sell changes your take-home more than most beginners expect.
Large general marketplaces give you the widest audience and charge the most, with final value fees that vary by category and seller status. Specialist card marketplaces usually charge less and reach buyers who know exactly what they want. Direct sales through social platforms and community groups avoid marketplace fees and leave you to manage the payment risk yourself. Local game stores and card shows pay you less, but they pay immediately and take the whole lot.
Always work out your net after fees, payment processing, shipping and packaging before you decide a sale was profitable. A card sold for double what you paid can still lose money on a small ticket once postage and fees are counted, which is the most common accounting mistake in low-value reselling.
Two practical rules. Sell the cheap material in lots, because your time per transaction is the real cost. And ship properly, because a card damaged in transit costs you the sale, the fees and your reputation.
What did your last small sale actually leave you after postage and fees? If you do not know to the penny, that is your next job before you list anything else.
Counterfeits and authenticity
Fakes are a real and growing problem, especially for high-value vintage cards and sealed product, and their quality has improved a lot.
Your defences are straightforward. Learn the physical tells for the products you trade, including card stock, print pattern under magnification, and the light test for the internal layer. Buy from sources with a history, and be wary of new accounts offering a bargain. Be sceptical of resealed sealed product, which is a specific and well-documented fraud in this hobby. And use grading as authentication where the value justifies it, because that is one of the things grading truly provides.
If you cannot authenticate a category with confidence, stay out of it. There is more than enough business in the material you do understand.
Breaking, and the regulatory line
Breaking, where people buy a random slot in a live pack opening, is popular, entertaining and legally sensitive.
Here is the structure: a participant pays a fixed price for an uncertain outcome that may be worth much more, or nothing. That looks enough like a lottery that several jurisdictions regulate it, and the treatment varies widely between countries and between states or provinces within them. Payment processors and platforms also apply their own rules on top of the law.
If you are considering this model, find out what applies where you and your customers are located before you take any money. Get the answer from the regulator's own material or from a professional. Other breakers may sound confident, and their confidence proves nothing.
The same caution applies to repacks and mystery boxes, which raise the same question.
Would you be comfortable explaining your break format, line by line, to a gambling regulator? If you hesitate at any line, that is the part to check first.
Tax and record-keeping
This is a business with inventory, and inventory businesses have specific record-keeping rules in most places.
Keep a record of what each item cost, when you bought it, what it sold for and what the fees were. Without cost records you cannot work out your profit, and in many tax systems you will be taxed as though the entire sale price were profit.
The rules on how collectibles are taxed vary by country and sometimes differ from ordinary business income. Marketplaces report seller income to tax authorities in many countries and the thresholds change, so find out what applies to you before your first busy year, while it is still easy to set up.
Set aside a fixed percentage of every sale from the beginning, in a separate account.
Learning to grade, which is the actual skill
Everything in grading arbitrage rests on predicting a grade before you pay the fee, and you can learn this without spending anything.
Graders assess four things, and it is worth understanding each one on its own, because a card usually fails on just one.
Centering. How evenly the border frames the image, front and back. This is the most common reason a card that looks perfect gets a lower grade, and it is the easiest to check because you can measure it. Hold a ruler to it. A card that is visibly off-centre will miss the top grade however clean it is.
Corners. Look at all four under magnification and angled light. Corner wear is often invisible flat-on and obvious at an angle.
Edges. Chipping and whitening along the edges, especially on dark-bordered cards, where any wear shows at once. Dark borders are punished hardest, so judge them most sceptically.
Surface. Scratches, print lines, dents, dimples and finish damage. You need a light held at a shallow angle, and this is where most people miss defects, because a card viewed flat under a ceiling light looks flawless.
The way to learn is to predict grades on your own cards, write each prediction down, and check it against graded copies of the same card already for sale. Do that a few hundred times and you will build a usable eye. Skipping this and submitting on optimism is what makes grading arbitrage lose money.
A useful rule for yourself: if you catch yourself hoping a card grades well, and you cannot honestly say you expect it to, keep it out of the submission.
How many of your own predictions could you check this weekend against graded listings? Even twenty will show you whether your eye runs hopeful or harsh.
Running it as an actual business
Once you are past the first few months, the difference between a hobby that sometimes makes money and a business comes down entirely to process.
Know your inventory. What you hold, what it cost, how long it has been sitting. Stock that has not moved in six months is a decision you have been putting off, and that is very common in this hobby, because selling a card at a loss feels like admitting a mistake. In a business it is a normal event.
Set a turn target. How many times a year your capital cycles. A smaller margin on faster turnover usually beats a bigger margin on stock that sits, and beginners keep chasing the wrong one.
Price from sold data, always. Asking prices tell you what sellers hope for. Sold prices tell you what the market did. Almost every overpayment traces back to this.
Batch the boring work. Listing, packing and postage in dedicated blocks, never one at a time. Time per transaction is the real limit on a card business, and batching is the only lever you have on it.
Decide the floor. The minimum value below which you sell in lots. Without one, you will spend an hour selling a card worth very little, which is the most common way this business quietly pays you less than minimum wage.
Reinvest deliberately. Growth here comes from capital and knowledge, and both compound. Taking every pound of profit out keeps the business permanently small, which is a perfectly fair choice, as long as you make it on purpose.
Which card in your stock has sat the longest, and what would it sell for today if you priced it from sold data? Listing that one card this week is a small, clean start on the habit.
Keeping cost records and reinvesting on purpose is the moment your collection becomes stock. You know what you paid for each card, what it sold for and what you kept after fees, and that skill travels with you to any trade you ever try. It starts with one cost column on your next batch.
Why this survives automation
This deserves saying outright, because much of the standard side-income advice is being repriced by generative AI, and this category is holding up.
Software already tells everyone what a card is worth. Price data is free, instant and universal, and has been for years. That wiped out the easy edge of simply knowing prices better than a casual seller online, and the business carried on regardless, which tells you where the value really sits.
What no machine can do is turn up. Driving to a house to buy a collection. Valuing a shoebox of unsorted cards in twenty minutes with cash in your pocket. Sorting bulk. Judging a card's surface under angled light. Packing it so it survives the post. Building a relationship with a shop owner who calls you first.
Every one of those is physical, local and unglamorous, which is exactly the kind of work that holds its value while remote deliverable work gets cheaper. The trade-off for you: it does not scale the way a digital product does, and it needs you to be somewhere specific.
There is a knock-on effect worth noticing. As more remote work gets automated and more people look for alternatives, more people will try physical resale businesses too. The edge will keep moving toward whoever does the least pleasant part of the job, which in this business means sorting bulk and valuing collections quickly.
So the skill you build this year keeps paying. The collectors you meet, the shop owners who start calling you when a collection comes in, the eye that values a binder in minutes: those take time, and people starting next year will be behind you on all of them. Start learning now, and do it slowly with small money.
The categories beyond Pokemon
Pokemon dominates, and other markets exist alongside it, sometimes with better margins exactly because fewer people chase them.
Sports cards. The older and larger of the two traditions, with a different collector base and a strong link to live events. The 2026 World Cup has pulled soccer collectors into grading in numbers the hobby has never seen, which is a demand event with a defined timeline. Treat it as a season, and expect it to pass.
Other trading card games. Magic: The Gathering has the longest continuous history in the category and a truly scarce early period, with a large secondary market driven partly by playability as well as collecting. That double demand makes its prices behave differently from purely collectible markets.
Non-English printings. Japanese Pokemon cards, for instance, have different print quality, different grading outcomes and a different buyer base. You can often source regional printings cheaply in their home market and sell them where they are scarce, which is the same cross-border logic that runs through the rest of this site.
Sealed vintage. The highest-value and highest-risk end, where authentication matters most and resealing fraud is a documented problem. Stay away from it as a beginner, whatever the circumstances.
The practical advice is the same as for Pokemon. Pick one narrow area, learn it properly, and expand only when you are truly fluent, because spreading yourself wide is what leads to overpaying.
Which one game or era do you already know better than most casual sellers? That is your narrow area, and it is usually closer to home than you think.
Capital, and how much you actually need
You have now seen the categories and how the work runs week to week. What is still open is how much of your money has to sit still while you learn, and whether $50 or something nearer $5,000 is the honest starting figure for you.
You can start with very little, and the model changes as the number rises.
Under 100 dollars. Bulk sorting and small raw flips only. At this level you are learning the operational side: what postage costs, how fees work, how long things take. Expect a poor hourly rate and treat it as paid training.
A few hundred dollars. Single-card raw arbitrage with a real spread, and small collection lots. Turnover matters more than margin at this level, because you are learning which categories move.
One to a few thousand. Collection buyouts become possible, which is where the best margins are, and grading becomes viable on selected cards. This is roughly where the business starts paying you for the time it takes.
More than that. Sealed vintage, high-value graded material and volume buying. A different risk profile entirely, and losses at this level are the kind that hurt.
The mistake at every level is putting in money you need. Card prices move on sentiment, product releases and attention, and there are stretches where the sensible move is to hold your inventory and wait out a falling market. You can only do that with money you can leave alone.
If your whole card budget vanished next month, would the rent notice? If the answer is yes, the budget is too big, so cut it until the answer is no.
Card money kept in its own account is the rule that lets you sleep through a falling market. Grow the stock only out of profits, so the rent and the food shop never sit inside a binder waiting for prices to recover. When your household budget never touches the inventory, holding through a slow month is a choice you can afford.
Doing this outside the United States
Most writing on this subject assumes an American seller, and if you live elsewhere the picture changes in ways that bring opportunities as well as obstacles.
Grading logistics dominate the economics. The major graders sit in a few countries, so international submissions mean postage both ways, insurance on both legs, customs paperwork and often import duty when the cards come back. That can add more to your cost per card than the grading fee itself, and it makes grading low and mid-value cards uneconomic from many countries. Bulk submissions through a group submitter, where several people's cards travel together, are the standard workaround, and worth finding before you decide grading is impossible for you.
Import duty and VAT apply to cards. Collectibles crossing borders are taxed like other goods in most systems, and buyers often forget this when sourcing from overseas marketplaces. Work out the landed cost. The listing price is only the start.
Regional price differences are the opportunity. The same card can trade at very different prices in different countries, because local supply, local collector numbers and shipping costs differ. Sourcing where a printing is common and selling where it is scarce is the clearest edge open to you as a seller outside the main markets, and it is the same cross-border logic that runs through everything else on this site.
Payment and platform access vary. Some marketplaces restrict seller registration by country, and payment processors differ in what they support. Settle how you will receive money before you build inventory, because a stock of cards you cannot sell through your preferred channel is an expensive lesson.
Local scenes are underserved. In many countries there is far less competition for local collection buyouts than in the United States, because fewer people do this professionally. That is the highest-margin sourcing channel, and it rewards being physically present, which no overseas competitor can be.
The general rule: the further you are from the main grading and buyer infrastructure, the more you should lean toward raw trading and local sourcing over grading, and the more your knowledge of local supply is worth.
Common mistakes
Buying modern sealed as an investment. Ten billion new cards a year is the answer to this. Modern sealed is inventory with a shelf life, and the people promoting it as a store of value are often the ones selling it.
Grading cheap cards. The fee is roughly fixed and the spread grows with value, so grading low-value material reliably turns your money into slabs worth less than the process cost.
Modelling only the best-case grade. If the deal needs the top grade to work, you are betting. Model the grade below it.
Ignoring fees and postage on small sales. Doubling your money on a low-value card can still be a loss after postage, packaging, marketplace fees and payment processing. Work out the net, always.
Buying broad instead of narrow. Knowing a little about everything is how people overpay. One game, one era, priced from sold listings, ignoring asking prices.
Treating a shortage as scarcity. A set being hard to find this month says nothing about its supply in three years, especially when the manufacturer is printing to meet demand.
No records. Without cost records you cannot work out your profit, and in many tax systems you will be assessed as though the whole sale were gain.
Scope of this guide
The Pokemon Company figures are company results for the fiscal year ended February 2026, and the cumulative print figure is stated as of March 2026. Companies restate and update these.
The PSA volume, backlog and tier-suspension details come from trade coverage of PSA's own announcements during 2026. Service levels, prices and turnaround times change often, and you should check them directly with the grader before you commit money to a submission.
The dollar conversion of the Pokemon figure uses a single reference rate on one day, and the yen moves, so the yen figure is the reliable one.
Nothing here is investment advice, and the central argument of this page is specifically that the production numbers contradict treating modern cards as an investment.
Who should skip this
If you cannot leave money idle for months, skip grading specifically. Buy and sell raw instead, where your capital turns over in weeks.
If you dislike admin, skip this entirely. It is inventory, listings, packing, postage, fees and records, and the card knowledge itself is a small fraction of the work.
If stories of a card selling for a fortune are what drew you in, be honest with yourself that you want the lottery, and the business is something else. Those stories are real, they make a poor strategy, and the enormous cumulative print runs mean the modern equivalent is unlikely.
If you are a collector who loves the hobby, think carefully about whether turning it into a business will survive the parts you will start to resent. That is a common and underrated cost.
A realistic first ninety days
If you are still reading, the question has stopped being whether to try this. Here is what the opening stretch looks like when you are the one doing it.
Days 1 to 14. Pick one narrow category and learn its prices properly: one game, one era, even one language. Breadth is what makes beginners overpay. Track sold prices and ignore asking prices, because asking prices are fiction.
Set the amount you can afford to lose entirely and treat it as your whole budget.
Days 15 to 45. Buy and sell raw only, at small ticket sizes, to learn the operational side: listing, packing, postage costs, fees and how long things really take to sell. Complete at least ten full transactions. The point is to learn what your net margin truly is after everything.
Practise grading assessment on cards you own without submitting anything.
Days 46 to 75. Try one sourcing channel that takes effort beyond clicks: a local listing, a collection lot, a bulk purchase you sort yourself. This is where the lasting margin lives, and it is the step most people skip.
Work out your real hourly rate, including sorting, listing and packing time. It usually comes as a shock, and it is the number that tells you whether to carry on.
Days 76 to 90. Only now consider grading, on a small number of higher-value cards where the spread justifies the fee and the wait, after checking the current service levels and turnaround directly with the grader.
Decide which of the five models you are actually running, and stop doing the others.
When you work out that hourly rate at day 75, what number would make you carry on, and what number would make you stop? Write both down on day 1, so the decision is already made when the shock arrives.
Waiting feels safe and it quietly costs you. Every month you only browse, someone else builds the relationship with the shop owner and the collector down the road. You can start with the cards in your own drawer and a notebook. Price ten of them tonight, list the one with the clearest sales history, and let that first sale teach you more than another month of watching.
What the demand data is actually telling you
It helps to separate the parts of the current boom that are structural from the parts that are events, because they mean different things for you starting now.
Structural. Pokemon's 29.3% revenue growth to 531 billion yen in the year to February 2026 reflects a franchise that has widened its audience, partly through a mobile product: Pokemon TCG Pocket took more than 1.6 billion dollars in gross player spending in its first eighteen months and passed 150 million downloads. A digital product that introduces huge numbers of new people to the card game is a real expansion of the collector base, and it looks set to last.
Event-driven. The 2026 World Cup pulling soccer collectors into grading is a demand event with an end date. Pokemon's thirtieth anniversary year is likewise a specific period of extra attention. Both are real, and both will fade back toward normal demand.
Capacity-driven. Record grading volumes partly reflect graders processing more, alongside collectors submitting more. A backlog above 10 million cards shows demand outrunning capacity. It is a bottleneck, and a weak market signal.
Why separate these? Because if you enter now assuming current conditions are permanent, you will decide differently from someone who assumes attention settles down. The second assumption is the safer one, and it points you toward the operational business on this page, and away from piling up inventory in the hope that today's prices are a floor.
One more caution about the growth figures. Manufacturer revenue and secondary market value are two separate numbers. The Pokemon Company selling more cards means more supply reaching collectors, which is good for its shareholders and does nothing automatically for the resale value of what you already hold. Those two numbers get quoted as if they were the same, and they point in opposite directions.
The honest summary
The market is real and growing quickly. The Pokemon Company posted 531 billion yen of net sales for the year to February 2026, up 29.3%, and PSA graded a record 3.5 million items in June 2026, of which roughly 2.5 million were cards, up about 74% year on year.
Supply is growing too, and that is the part the promotional material leaves out. More than 85 billion cards have been printed in total as of March 2026, roughly 10 billion of them in the last year, which makes modern product inventory to be sold on.
The grading queue is a real business constraint right now. A backlog reported above 10 million cards and four popular tiers closed to new submissions from June 2026 mean your capital sits idle for an unpredictable period.
The money is in sourcing and operations. Buy below market from people who have not priced their cards, do the sorting nobody wants to do, sell efficiently, and keep records. That is a real small business, it resists automation because it needs you physically present, and it has almost nothing in common with the version of this hobby that gets promoted to you.
One closing test for any decision you make in this business. Ask whether you are being paid for work or hoping to be paid by a price move. Sorting a collection, valuing a lot accurately, packing well and selling efficiently are work, and they pay reliably in proportion to the effort. Holding a modern box for three years is a hope, and 85 billion printed cards is the reason it is a poor one.
The people who last in this hobby as a business are, almost without exception, the ones who treat it as logistics with a knowledge component, and who stopped trying to predict the market. That description is less exciting than the one used to sell it to you, and it is the one that matches the numbers.
If you want a single starting instruction, here it is. Buy one unsorted bulk lot in a category you have studied for a fortnight, sort it yourself, list what is worth listing individually, sell the rest in lots, and write down every cost, including your hours. At the end you will know your real margin, your real hourly rate, and whether you can stand the sorting, which is the part that decides whether anyone carries on. That one exercise teaches you more than any amount of reading about which set to buy, and it costs less than a single graded card.