Every Charizard's PSA 10 Premium, Ranked

There is already a price answer for this species. How much is a Charizard card worth walks the ask ladder across 114 English printings and reports a median PSA 10 premium of 10.22x across the 91 of them carrying a recorded PSA 10 sale. This is the other half of the question: not what a Charizard costs, but what grading one adds, printing by printing, ranked. On the 82 printings where we hold both a raw ask and a PSA 10 sold median the multiple runs from 1.45x to 60.37x, a forty-fold spread inside one Pokemon. The obvious explanation is price, and price turns out to be the weaker one. What separates the top of this list from the bottom is how many times the slab has actually sold.

114 English priced products carry the word Charizard. Nine are shared cards, five Reshiram & Charizard GX printings and four Charizard & Braixen GX, and this study excludes any card naming two species because a tag team's price is not attributable to either. That leaves 105. Of those, 82 carry a PSA 10 sold median as well as an ask, 101 carry a PSA 9 sold median, and 23 carry no recorded PSA 10 sale at all. The published price study's 114 and this study's 105 are the same catalog read under two stated rules, and a recompute can pick either.

The 23 with no recorded PSA 10 sale are excluded rather than scored, because absent is not zero, and they are not a rounding detail. Charizard from Plasma Storm asks $1,150.00, Charizard from Base Set asks $919.98, Charizard from Stormfront asks $737.50, and Charizard from Base Set 2 asks $512.90. Four of the most recognisable cards in the hobby have no PSA 10 sale on our tape, and their absence tilts this study away from the vintage top of the species.

The two thin bands are close enough to each other that their order should not be read as meaningful: 26.38x on 17 printings against 31.63x on 13. The step that matters is the one below them. Once a Charizard printing's PSA 10 median rests on a hundred or more verified sales, the multiple collapses to single digits and stays there. The rank correlation across all 82 printings is -0.590, and the mechanism is not mysterious. A card whose slab has hundreds of recorded sales has a raw price that already knows what the slab is worth, because the two markets watch each other daily. A card whose PSA 10 has ten does not.

The obvious counter is that comp depth is a proxy for price, with cheap modern cards trading more. On this species it is the reverse. The median raw ask FALLS as comps rise, from $286.21 in the thinnest band to $28.81 in the deepest, so the deeply traded Charizards are the cheap ones, and they are the ones with the smallest premium. The same result on the same tape for a whole family of cards is in every Eeveelution ranked, where Umbreon carries a median 121 comps and the lowest multiple of the eight.

Sorting the same 82 printings by raw ask does not produce a ladder. The under-$10 band sits at 17.52x, the $10 to $50 band at 9.85x, the $50 to $200 band at 19.90x and the $200-and-up band at 7.46x, and that non-monotone shape is printed as it came out rather than tidied with a different set of cuts. The rank correlation between the ask and the multiple is -0.209, which is a real but weak tilt.

Where price does dominate is the cash. The rank correlation between the raw ask and the CASH gap, the sold median minus the ask in dollars, is +0.759, and the last column of that table climbs cleanly from $93.30 to $1,969.56. A $6.45 Charizard ex multiplying 37.71x adds $236.75. A $694.67 Mega Charizard X ex multiplying 2.45x adds $1,005.33. The multiple is the percentage and the gap is the paycheque, and on this species they point in opposite directions.

These are the printings where a slab adds the most relative to the card. The comp column is the caveat on every one of them, and it is also the point: eleven of the twelve rest on fewer than fifty verified sales.

And these are the printings where the slab adds least. Eleven of the twelve rest on 100 or more verified sales, six of them on 400 or more, and one on 1,983. This is the same list a modern collector would name as the liquid Charizards, which is the finding stated the other way round.

The first and most serious is that this is a noise artefact. A median over ten verified sales is a noisier estimate than one over a thousand, and noise widens the top tail of any ratio, so thin-comp printings could carry high multiples for a purely statistical reason. The test is to run the identical construction across the whole English catalog, where each comp band holds hundreds of printings rather than a dozen. Across the 4,677 English products carrying both figures, the ladder is clean and monotone: under 25 comps 21.28x on 2,881 products, 25 to 99 comps 12.22x on 1,039, 100 to 299 comps 7.52x on 521, and 300 or more 4.55x on 236. Noise does not produce a monotone decline across bands that size.

The second is that the raw ask is the problem rather than the slab: thin-comp cards may simply be listed too cheaply. The PSA 9 leg tests that, because it never touches the ask. On the 82 printings carrying a raw ask and both grades, the median PSA 10 sells for 5.20x the PSA 9, and the rank correlation between comp depth and that ratio is -0.641. The premium between two grades compresses with liquidity exactly as the premium between raw and slab does, and the ask plays no part in it.

The third is selection: 23 printings are excluded for having no PSA 10 sale, and those are by definition the least liquid of all. If they were included at whatever multiple they would carry, the thin end of this ranking would grow rather than shrink, so the exclusion works against the finding rather than for it. It is still a real hole, and it is why the article ranks the Charizards that trade in slab form rather than all Charizards.

Raw is an ASK and graded is a SOLD median, so every multiple here divides a listing by a transaction. A printing whose sellers list high reads low on this ladder for reasons that have nothing to do with grading. The comp count is a lifetime count of verified sales we have recorded, not a rate, so an old card with a long recording history and a new card that traded hard for one month can land in the same band.

No fee, shipping, wait time or gem rate appears in any figure here, so nothing in this article is a return. A 60.37x multiple on a card that gems one time in twenty is a worse trade than a 5x multiple on a card that gems four times in five, and is grading Pokemon cards worth it is where that arithmetic lives. This is one day's cross-section on a species whose modern printings move fast.

The claim is that comp depth orders this ranking better than price does. A recompute on the same day's tape returning a comps-against-multiple correlation weaker than the ask-against-multiple correlation would overturn it. A recompute of the catalog-wide comp ladder that came back flat or non-monotone would reduce the Charizard result to a small-sample artefact. The cash claim is separate and should be tested separately: a recompute returning a negative rank correlation between the raw ask and the cash gap would break it. Related work: how much is a Charizard card worth, how much a PSA 10 adds, every Eeveelution ranked, and the species premium, where Charizard finishes third of 302 on raw asks alone.

Sources: the `cards` catalog, the daily `price_history` tape and `graded_snapshots`, all read on day 20718, September 22 2026. Products are deduplicated on the TCGplayer product id; graded keys are looked up across every catalog id claiming the product plus the product id itself. RAW is an ASK, the most recent daily market figure. GRADED is a SOLD median, the latest `graded_snapshots` row for the grade with a median above zero, carrying its count of verified completed sales. Population: English priced products naming Charizard and no other species, by the canonical-key matcher described in the species premium. 114 English priced products name Charizard at all; 9 name a second species (five Reshiram & Charizard GX, four Charizard & Braixen GX) and are excluded; 105 remain; 82 carry a PSA 10 sold median and are the study population; 101 carry a PSA 9 sold median; 23 carry no PSA 10 sale and are excluded rather than scored as zero. The multiple is the PSA 10 sold median divided by the raw ask, per printing; the cash gap is the same two figures subtracted, in dollars. Both band tables are closed intervals and each sums to 82; the comp bands are cut at 25, 100 and 300 verified sales and the ask bands at $10, $50 and $200, and both are printed as computed, including the non-monotone ask table. Correlations are Spearman's rank correlation with average ranks for ties across the 82 printings. The catalog-wide check applies the identical comp bands to the 4,677 English products carrying both a raw ask and a PSA 10 sold median, and its four rows sum to 4,677. Ranking ties break on the card id ascending; medians are upper-middle on even samples and quartiles are nearest-rank. The published price study this extends matched on the word Charizard without the single-species rule and reported 114 printings and a median premium of 10.22x over the 91 of them with a recorded PSA 10 sale; this study's 105 and 11.19x differ for that reason and for the product-level deduplication, and both rules are printed so a recompute can reproduce either. Not financial advice.