Which Dragon cards carry the widest gap between the raw ask and the PSA 10 sold median, on 5 or more verified comps? A wide gap is where grading adds the most, and where a raw copy in gem-ready condition is priced as if it were not.
PSA 10 verified sold median over the raw ask, cards with 5 or more verified PSA 10 comps.
| # | Card | Premium | PSA 10 | Raw ask | PSA 9 | Gem rate |
|---|---|---|---|---|---|---|
| 1 | ![]() Golem ex#91 · Rare | 41.9x | $3,3365 comps | $79.56 | $40014 comps | 7.5% |
| 2 | ![]() Charizard#100 · Rare | 40.8x | $20,0009 comps | $490 | $2,40019 comps | 5.2% |
| 3 | ![]() Muk ex#96 · Rare | 28.1x | $2,8997 comps | $103 | $3487 comps | 9.7% |
| 4 | ![]() Charmeleon#99 · Rare | 19.7x | $2,3255 comps | $118 | $38711 comps | 8.3% |
| 5 | ![]() Charmander#98 · Rare | 19.6x | $6,40010 comps | $327 | $47911 comps | 9.4% |
| 6 | ![]() |
Raw is the canonical printing's TCGplayer ask, the same figure the card page shows. PSA 10 and PSA 9 are eBay sold medians verified per listing, printed with their comp counts; a premium is only computed on 5 or more PSA 10 comps.
Gem rates come from the PSA registry snapshot and are only trusted on lines of 100 or more slabs; a zero beside a real population is a provider gap and is withheld, never shown as 0%. Moves and highs come from each product's own recorded series, trimmed at the price-rule change of 2026-09-07, and ignore cards under $5.
Every figure carries its sample. A short series, a thin comp count or a small registry is printed beside the number rather than hidden, and a card that lacks the figure a reading needs is left off that reading instead of scored on a guess.
| 12.0x | $1415 comps | $11.77 |
| 7 | ![]() Gyarados#32 · Uncommon | 7.8x | $1547 comps | $19.71 | $1305 comps |