We’ve been pulling 86 days of order data from a single Direct seller with around 30,000 cards in inventory, and a few pricing quirks stood out. Some are current, one kicks in June 18. Figures below are rates and proportions rather than raw counts so they’re useful at any inventory size, though the leakage rate at the end is especially small-sample and should be treated as directional.
Listing Below the Floor Costs You Nothing
Since February 2026, Direct listings below $0.40 get corrected up to $0.40 at checkout. Whatever number sits on the price tag, the buyer pays $0.40 and you receive $0.40. So a card listed at $0.01 and a card listed at $0.40 generate the same revenue when they sell, and the only thing that differs between them is the listed price itself, which presumably feeds into how TCGplayer’s system ranks or selects inventory in ways we don’t have great visibility into.
Whether that’s intentional floor gaming or legacy pricing that predates the February policy, 49.2% of Direct transactions in our sample settled at exactly $0.40, so nearly half of our Direct volume is floor-corrected. At 10,000 Direct orders a month, that would be roughly 4,900 transactions where what you listed and what you received were two different numbers.
One piece of that ranking we’ve since heard about from a TCGplayer source who asked to stay anonymous: when multiple sellers are tied at the floor, the store that’s had the card listed the longest wins the sale. We can’t independently verify that, but it lines up with the mechanic — once every sub-$0.40 listing collapses to $0.40 at checkout, the system has to break ties on something other than price, and listing tenure is a reasonable candidate. If true, undercutting bulk inventory buys you nothing on the demand side; the copies that have been sitting longest are the ones quietly winning the floor sales.
The $3.00 Fee Cliff
The other current quirk is the $3.00 fee cliff. Direct orders under $3 run on a simplified split: the seller gets 50% of item value with no SRC calculation involved. At $3 and above, the full fee stack kicks in: 8.95% commission, variable SRC, plus 2.5% + $0.30 per transaction. Using TCGplayer’s published SRC example of $1.26 for a small Standard card order:
| List Price | Fee Structure | Seller Net | % Kept |
|---|---|---|---|
| $2.99 | 50% flat (sub-$3) | $1.50 | 50.0% |
| $3.10 | Full stack + SRC | $1.19 | 38.2% |
| $3.50 | Full stack + SRC | $1.54 | 44.0% |
| $4.00 | Full stack + SRC | $1.98 | 49.5% |
SRC estimated at $1.26 from TCGplayer’s published example for a small single-card Standard shipment. Actual SRC varies.
A card at $3.10 nets you $0.31 less than the same card at $2.99, and you don’t claw back to the $2.99 net until somewhere around $3.45. Anything sitting just over $3 is probably bleeding margin compared to where it could be priced.
The New $2.49/$2.50 Boundary
That cliff has about five weeks left. On May 13, TCGplayer announced that on June 18 the Direct Reimbursement Invoice and Store Your Product fees move from the order-based SRC model to a per-item model. The old model was hard to reason about because the same listing could net differently depending on what else was in the buyer’s cart. The new one is per-item, which fixes that.
Under the new rate card, listings from $0.40 to $2.49 keep the 50% flat split with commission and transaction fees waived. Listings at $2.50 and up pay a flat $1.12 per item plus 8.95% commission and 2.5% transaction fee. The fixed $0.30 transaction fee that currently applies to commission-based Direct orders goes away. Worked across a price range:
| List Price | Fee Tier | Seller Net | % Kept |
|---|---|---|---|
| $2.49 | 50% flat | $1.25 | 50.0% |
| $2.50 | Item fee | $1.09 | 43.8% |
| $2.68 | Item fee | $1.25 | 46.8% |
| $5.00 | Item fee | $3.31 | 66.2% |
| $50.00 | Item fee | $43.16 | 86.3% |
Excludes Pro Tool fees if applicable.
TCGplayer says more than 95% of Direct sellers should see lower fees under the new structure based on YTD performance, and that’s probably right. On higher-value cards it’s hard to complain: a flat $1.12 on a $10 or $50 card is small and predictable, and dropping the fixed $0.30 transaction fee helps.
The new wrinkle is what happens around $2.50. At $2.49 you keep 50%, or about $1.25 before cost basis. At $2.50 you pay the $1.12 item fee plus 8.95% commission plus 2.5% transaction fee, which leaves about $1.09. So $2.50 isn’t automatically better than $2.49, and you don’t recover the $2.49 net until roughly $2.68 (before any Pro fee).
That should pull pricing behavior in two directions. Cards that can’t justify going higher will cluster just under $2.50. Cards with enough demand and margin to absorb the $1.12 will sit safely above it. The worst spot is barely over the line, where the listing looks higher but the net is lower.
For true bulk under $1, the new model is still not generous. A $0.40 card losing $0.20 to the 50% fee is rough, and waiving commission and transaction fees in that band doesn’t change much about that. The new structure makes the damage more legible; it doesn’t make low-value bulk a good business.
An Experiment We’ve Been Running
We’ve been running a small, deliberately weird test around the floor. Direct eligible floor inventory was listed at $0.01 with marketplace shipping fixed at $19.99, while everything outside Direct stayed on standard competitive marketplace pricing. The inflated shipping line on the marketplace side keeps standard purchases unattractive, so we can watch what sub floor listed prices actually do at the system level without drowning in shipping loss orders.
Over 86 days, that setup produced roughly 1 marketplace leakage order per 1,000 floor priced SKUs: buyers who came in through the standard channel and paid the $19.99 shipping rather than routing through Direct. The cards involved skewed bulk uncommons and commons across The List reprints, Commander Legends, Foundations Jumpstart, and a long tail of lower value Standard sets.
That rate comes from a single seller’s small sample and shouldn’t be treated as precise, but it scales cleanly. At 200,000 floor priced SKUs and the rate holding, you’re looking at roughly 200 leakage orders over the same 86-day window. Whether that’s interesting depends entirely on what you’re charging for shipping.
We have theories about what’s driving the behavior, but we’re more curious whether other sellers have seen anything similar. If you’ve experimented with sub floor listings, watched orders show up at unexpected shipping prices, or noticed any other strange routing around floor-priced inventory, we’d like to hear about it.
Drop us a line at analyst@mtgsold.com.
Order data and pull sheet analysis from a single TCGplayer Direct seller, February - May 2026. Current and upcoming fee structures from TCGplayer’s published documentation and May 13, 2026 seller announcement. Floor policy from TCGplayer’s seller blog announcement, January 8, 2026.