Should You Even Grade It?

Grading feels like free money: pay $20, get back a number, sell for more. But the uplift is a conditional, not a constant, and the card you're submitting is usually a maybe, not a banger. Here's the expected-value math, the break-even by raw price band, and the honest test that survives your own optimism.

Based on ~1,045,000 graded eBay sold comps (via Scrydex), pulled June 2026 and almost all from the past year. Treat it as a popular-card snapshot, not the whole market, so read these as directional findings.

---

The fantasy and the invoice

Every flipper has done it. You pull a clean card, hold it to the light, and the math runs itself: raw it's worth $20, a PSA 10 of this goes for $70, grading is twenty bucks, so that's thirty dollars of free money sitting on my desk.

The problem isn't the arithmetic. It's that the arithmetic is missing three of its four terms. You priced the upside as if it were guaranteed, ignored the time your cash sits frozen, and quietly assumed the card comes back a 10. None of those are safe assumptions, and two of them are the difference between a good submission and a $25 donation to PSA.

So let's actually do it. Not the back-of-the-napkin version that justifies what you already wanted to do, but the real one, with the conditional uplift, the break-even, and the term everyone leaves out. The good news: the rule that falls out the other side is simple enough to apply in the ten seconds you're holding the card.

What a 10 is actually worth over the raw

Here's the thing the fantasy gets right: the uplift is real, and it scales hard with the value of the raw card. The aggregate figures below are median PSA 10 prices minus the same card raw, bucketed by what the raw sells for (USD, recent sold comps):

Raw under $10 → +$32. A ~$6 raw becomes a ~$64 PSA 10. The dollar gap looks great as a multiple, more than 10×, but hold that thought, because a multiple isn't a margin.

Raw $10–25 → +$50. Raw $25–50 → +$81. Raw $50–100 → +$154. The gap widens steadily as the raw climbs.

Raw $100+ → +$705. A ~$449 raw becomes a ~$3,761 PSA 10. This is where grading stops being a hobby and starts being a business decision.

Two honest caveats. These are medians across many cards, not a quote for yours, and a specific card can sit well above or below its band. And every one of these numbers is the uplift if the card grades a 10. Cross out that condition and the whole table is fiction.

The cost side nobody fully tallies

The sticker cost is easy: roughly $15–25 per card at common service levels, plus shipping. Call it $20 all-in for a single card, less per card if you bulk-submit, more if you pay for speed.

But the cash fee is only the visible cost. The two invisible ones:

Time. Standard tiers run weeks, sometimes months. That's not free. It's your capital locked in a queue, unable to be flipped, compounded, or spent on the next pull. More on that below, because it's the cost flippers most consistently price at zero.

The downside leg. If the card comes back a 9 instead of a 10, you didn't get the uplift in that table. You got the much smaller PSA 9 premium, and you still paid the full fee and waited the full weeks. The fee is certain. The uplift is not. Treating them as if they live on the same line of the ledger is the core mistake.

Break-even, plainly

Strip it to the bone and ignore the grade risk for a second (we'll add it back, and it only makes the case stricter). Grading is worth it when:

uplift > fee. That's the floor. With a ~$20 all-in cost, look at the bands:

Sub-$10 raws (+$32): after a $20 fee you clear about $12, and that's the best case, before grade risk, shipping rounding, and selling fees eat it. You are working for almost nothing and taking all the risk.

$10–25 raws (+$50): clears ~$30 best-case. Marginal. Survivable only if your 10-rate is genuinely high.

$25–50 raws (+$81) and up: this is where it turns sharply positive. +$81, +$154, +$705 against a flat ~$20 fee, so now the fee is noise and the decision is about grade risk, not cost. The break-even isn't gradual; it has a knee in it, and the knee sits around $25 raw. Below it you're scraping; above it the economics open up fast.

The hidden cost: the premium is a conditional

Here's the line that should be tattooed on every submission form: the 4× premium is a conditional, not a guarantee, and you only collect it if the card 10s.

Think of grading as buying a call option. You pay a fixed premium (the fee) for the chance at a payoff (the 10 uplift). If the card grades, you exercise and collect. If it comes back a 9, the option expires nearly worthless: you keep a card worth a little more than raw, minus everything you spent. You don't get the fee back for guessing wrong.

So the honest test isn't uplift > fee. It's uplift × (your real 10-rate) > fee + the downside of a 9.

Run it on a $20 raw in the $10–25 band: uplift +$50. If your true 10-rate is 50%, your expected uplift is $25, barely above the $20 fee before you've even subtracted the cost of the cards that came back 9s. If your 10-rate is 30%, expected uplift is $15, and you are losing money on average by submitting. The card that looked like free money is a negative-EV bet the moment you multiply by reality.

Now run the same 30% rate on a $449 raw (+$705): expected uplift ~$211 against a $20 fee. Still wildly worth it. High-value cards survive a brutal 10-rate; cheap cards don't survive a generous one.

Your 10-rate is lower than you think

Which brings us to the most expensive number in this whole analysis: the one you make up.

Two biases conspire here. First, overconfidence: you remember your 10s vividly and quietly forget the 9s, so the rate you carry in your head is the rate from your best submissions, not your average one. Ask any flipper their 10-rate and the answer is suspiciously round and suspiciously high.

Second, the gambler's fallacy in reverse: "my last three came back 9s, I'm due a 10." You are not due anything. Each card is graded on its own surface, centering, and edges; the queue has no memory of your luck. A card that's off-center is off-center whether or not the previous one was.

The practical fix is unglamorous: write down your real results. Cards in, 10s out, by yourself, by card type. Most people who do this discover their honest 10-rate is meaningfully below their gut number, and that one correction flips a pile of "obviously worth it" sub-$25 submissions into clear passes. Your ledger already knows this; you just have to look.

Opportunity cost: the capital you froze

One more term, and it's the one that's literally invisible because nothing happens: your money just sits there.

When you submit a $20 card and wait eight weeks, that $20 (plus the fee) is locked. It can't buy the next underpriced lot, can't ride a card that's heating up, can't do anything but wait in a building in California. For a flipper whose whole edge is turning capital over, frozen capital is a real cost even when the grade comes back perfect.

This is why bulk-submitting marginal cards is quietly corrosive. Twenty $15 cards at the break-even edge tie up $300 of stock plus $300+ in fees for two months, and the best outcome, every single one a 10, clears thin. Any 9s and you've locked serious capital to lose money slowly. The same fees and freeze spent on a handful of $100+ raws would have made multiples more, with cash to spare for the next pull.

Opportunity cost doesn't show up on the PSA invoice. It shows up as the deals you couldn't take because your money was in line.

The rule of thumb that survives all of it

Put the four terms together (uplift, fee, your real 10-rate, frozen capital) and the guidance collapses into something you can run while holding the card:

Don't grade sub-$15 raws on hope. The uplift barely clears the fee even before grade risk, and grade risk is the whole game. The 10× multiple on a cheap card is a multiple, not a margin.

Grade your bangers, not your maybes. Above ~$25 raw the fee becomes noise and the EV stays positive even at an ugly 10-rate. That's where grading is supposed to live. A $100+ raw justifies grading almost regardless of how the surface looks; a $12 raw justifies it almost never.

Use your real 10-rate, not your optimistic one. Multiply the uplift by the honest number and re-decide. Most marginal submissions don't survive the multiplication, and that's the point.

Grading isn't free money. It's a call option with a fee, a wait, and a strike price set by a grader who's never met your optimism. Buy it on the cards where the option pays even when you're wrong, and pass on the maybes you were only ever grading on hope.