Flatbed Scans vs Phone Photos for Trading Card Grading

Flatbed scans and phone photographs answer different condition questions. A scan is strong at repeatable geometry and even detail across a flat plane. A phone can vary the light and viewpoint to reveal texture, gloss, and relief. Treating either as a complete substitute for in-hand inspection creates blind spots.

What a flatbed scan does well

A clean scanner can capture the entire card at consistent scale with little perspective distortion. That makes it useful for border comparison, printed detail, edge colour, card identification, and repeatable before-and-after records. Both sides can be captured with the same settings.

Disable automatic sharpening, dust removal, colour restoration, document cleanup, and aggressive contrast. Those features may erase subtle marks or create halos at edges. Save a lossless master when possible. Keep the scanner glass clean, but never press a card beneath heavy objects or force a curved card flat.

Scanner light travels in a fixed path. Foil, texture, and gloss can therefore appear dark, blown out, or unlike their in-hand appearance. A shallow dent may disappear because the light cannot rake across it. Scanner-glass dust and scratches can also look like defects unless you clean the glass and repeat the scan after rotating the card.

What a phone camera does well

A phone lets you change the light direction while the card remains still. That is the key advantage for detecting indentations, scratches, scuffs, gloss changes, foil lift, and surface waves. It also captures how the card presents at viewing distance.

The risks are perspective distortion, uneven focus, computational sharpening, reflections, compression, and clipped edges. Keep the normal lens parallel to the card, include a margin, use diffuse light, disable filters, and retain original files. A close ultra-wide image may make one end of the card larger than the other.

Choose by the evidence question

Use a complementary evidence set

For a high-consequence decision, capture a square front and back plus changed-light photographs of both surfaces. The square views can be scans or carefully controlled phone images. The angled-light views should come from a camera. Add close views only where they answer a documented uncertainty.

Name files consistently, such as card-a-front-square and card-a-back-light-left, and keep originals. Never infer that a surface is clean because the scan does not show a defect that requires directional light. Likewise, do not measure centering from an oblique phone image simply because it reveals texture well.

The right method is not the one with the highest advertised resolution. It is the method that exposes the condition feature you need to evaluate, preserves an honest record, and states what remains unseen.