Card Show Flip Math: How to Price Your Buys So the Numbers Work Before You Leave the Table

A common mental shortcut is to buy at 50% of comp. That is not wrong as a starting heuristic, but it is imprecise enough to get you in trouble on high-value cards or thin markets. A better frame is to work backward from your target net margin.

Assume a worked example with labeled assumptions: you find a raw card with recent eBay sold comps clustering around $80. You estimate platform fees and shipping at 15% of sale price, so your net after fees is roughly $68. If you want a 25% margin on your cost (a reasonable target for raw flips where you are taking condition risk), your maximum buy price is $68 divided by 1.25, which is about $54. If the seller wants $60, the math says walk away or negotiate. If they want $45, you have room.

For cards you plan to grade, the math is harder — and the stakes are higher than they used to be. With PSA's cheapest open tier now at $79.99 per card (verify current pricing and tier availability at psacard.com, as tiers can reopen or change), you need a meaningful spread just to cover the submission cost before you see any profit. A card that comps at $50 raw and $130 in PSA 9 looks attractive on paper, but after the $79.99 fee, return shipping, and platform fees on the sale, the margin is essentially gone and the outcome uncertain. A card that comps at $40 raw and $300 in PSA 10 with a low pop-10 is a more defensible bet — if it actually grades a 10. The gap between a PSA 9 and PSA 10 comp often spans hundreds of dollars, which means a wrong read on condition can turn a profitable submission into a loss at current fee levels. This is exactly where a tool like CardGrade earns its keep — photograph the card, get a grade prediction in 60 seconds, and make a data-informed decision before you commit. The grading-roi guide covers how to think about expected value across grade outcomes.

The biggest practical challenge at a show is that good comp research takes time you do not always have. A seller is watching you. Other buyers are nearby. Here is a workflow that takes under 90 seconds once you practice it.

Open eBay on your phone, search the card name plus set plus card number, filter to Sold listings, and sort by most recent. Ignore the outliers at the top and bottom — look at the cluster in the middle of the last 10 to 15 sales. That cluster is your comp. If the sales are spread wildly ($30 to $150 for the same card), that is a signal the market is thin or condition-sensitive, and you should either pass or apply a larger discount to account for the uncertainty. TCGplayer is useful for raw commons and uncommons where eBay volume is low — check current market price there. For high-value raw cards, 130point.com aggregates recent sales data and can surface comps faster than manual eBay searching.

One thing to watch: recency matters more than volume. A card that sold for $200 eight months ago and $60 last week is a $60 card. Always sort by recent. Also note whether the sold comps are raw or graded — mixing the two is a common mistake that inflates your perceived comp on a raw card.

Not every card you buy at a show belongs in a submission. The decision tree is roughly: does grading add enough value to justify the cost and wait time, or is the raw flip faster and more certain?

Grade candidates are cards where the PSA or BGS population is thin at high grades, where the price jump from raw to a 9 or 10 is large relative to the submission fee, and where the card you are holding looks like it has a real shot at that grade. With PSA Regular now at $79.99 per card as the entry point — and the four cheaper Value tiers paused indefinitely as of June 2, 2026 — the bar for a grade candidate is meaningfully higher than it was even a few months ago. A card that comps at $50 raw and $130 in PSA 9 leaves essentially nothing after fees. A card that comps at $60 raw and $350 in PSA 10 with a thin pop-10 is a real candidate — but only if the card grades a 10. The reading-population-reports guide explains how to interpret pop data without over- or under-weighting it.

It is also worth noting that alternative grading options are constrained right now. TAG, which had been an affordable AI-assisted option, closed its Basic, Standard, and Express submission tiers in May 2026 as it hit capacity — only its higher-cost Priority and Walkthrough slots remain open, so it is not a practical low-cost alternative at the moment. The same demand surge that overwhelmed PSA's Value tiers spilled over to TAG and CGC as collectors rerouted submissions. CGC is still accepting submissions but runs approximately 65 working days at Economy and around 120 working days at Bulk; check current turnaround estimates at cgccards.com before planning around those timelines. The practical reality is that cheap, fast grading is not available right now across any of the major services — the old sub-$35 bulk math is gone for the foreseeable future, and the minimum realistic cost to slab a card is $79.99 at PSA or a comparably priced lane elsewhere.

For the grade candidates you do identify, CardGrade lets you photograph the card at the show and get a predicted grade across 47 inspection points before you commit to the buy price. If the model says likely PSA 7, you now know the grading math almost certainly does not work at $79.99 per card — a PSA 7 rarely justifies that fee. If it says likely PSA 9-10, you have data to support paying closer to your ceiling. This is not a guarantee — it is a faster, more informed decision that becomes more valuable the higher the submission floor goes.

For the raw flips — cards with solid comps, no obvious grading upside, or where you just want to turn inventory fast — the priority is listing speed. A card sitting in a box at home is not making money. CardDealer identifies a card from a photo in about 1.4 seconds, pulls a price from a three-source blend, and generates a publish-ready listing for eBay, Shopify, or TCGplayer. For dealers coming home from a show with 80 cards to process, that workflow compresses hours of manual listing into a fraction of the time, at a flat fee with no percentage of sales taken.

If you are running a table at the show rather than just buying, your economics get more complex. You now have a fixed cost (table fee, which varies widely by show — $50 to $500 is a realistic range depending on event size and location) that you need to cover before you are profitable for the day. Know your break-even in dollar-of-sales terms before you set up.

A simple way to think about it: if your table costs $150 all-in and your average margin on raw card sales is 30%, you need to move $500 in sales just to break even on the table cost. That is before you account for the time you spent sourcing, pricing, and traveling. This is not an argument against running a table — shows can be excellent sourcing and selling environments — it is an argument for going in with realistic targets and not treating a slow show as a success because you had fun.

Pricing your table inventory is its own discipline. Cards priced too high sit; cards priced too low leave money on the table. A reasonable approach is to price at or slightly above recent eBay sold comps for raw cards, since show buyers expect to negotiate and you want room to move. If you are moving graded slabs, know the current eBay sold comp for that specific grade and recent sale date — not the asking price on active listings, which is often aspirational. The ebay-vs-tcgplayer-vs-whatnot guide covers platform-specific pricing dynamics in more detail.

The operators who do well at shows consistently are running a process, not just showing up and hoping. Before the show: know your fixed costs, have your friction percentage written down, and have your phone workflow ready for comps. During the show: run the buy math on every card before you commit, use CardGrade on anything you are considering submitting, and do not let excitement override the numbers. After the show: list fast, because time is carrying cost on your inventory.

One underrated discipline is tracking your show P&L separately from your overall card business. If you bought 35 cards at a show, spent $400, and netted $520 after all fees and costs over the following 60 days, your show ROI was 30% over two months — annualized, that is a real business. If you cannot calculate that number, you are operating on vibes rather than data. The card-business guide on this site covers how to think about your card operation as a real business with trackable margins.

Shows reward preparation and punish impulsiveness. The math is not complicated — it is just math that most people skip because they are excited to be at a show. Do the math first.