PSA no longer represents the best ROI (a data analysis)

[FT35]

Originally Posted by 49erRCCollector

OK, that makes sense, then.

There was like 1-2 subs a few years ago, post-Covid that I genuinely think I got screwed on but for the most part the cards are right where I think they should be.

My last sub was:

Eight 10s.
Five 9s.
Six 8s. (two of these I thought could have been 10s...)
One 7 (super condition sensitive, probably a correct grade)

But overall? Probably fair grades.

I see everyone complaining but I haven't really been destroyed unfairly lately.

I’m glad to hear that the hammer hasn’t dropped on you! Those results sound pretty good actually. The cards that I’ve seen the most change (at least on my subs) are anything prior to ultra modern. 80’s and 90’s stuff has been the worst by far. 2000’s are rough too. I haven’t seen much of a change with ultra modern stuff but I don’t sub as much of that to see a large enough sample size.

Funny way of putting things into perspective—I use a loupe to look at stuff but I actually got glasses thinking my eye sight must be why I’m getting low grades! I defaulted to my own fault originally thinking “I must have missed something!” It didn’t help…of all the “glasses subs” early this year, my gem rate hit rock bottom…lol. I’m not subbing until I see things restore to normal for PSA.

[SLGSports]

It seems that PSA still gives out 10s at a decent rate... but only if it's ultra modern cards. Once they've seen a card a bunch of times, they find EVERY issue and your grade suffers. Good luck getting a 10 on a card more than a few years old.

My theory is that they're capturing common card-specific issues and logging them in an effort to speed up future grading. Unfortunately, this introduces severe bias and unfairness for submitters who expect a level playing field, regardless of when a card is submitted.

Either that or their AI models find more issues the more they see a specific card.

[inaka]

Originally Posted by SLGSports

My theory is that they're capturing common card-specific issues and logging them in an effort to speed up future grading. Unfortunately, this introduces severe bias and unfairness for submitters who expect a level playing field, regardless of when a card is submitted.

I think your theory gives PSA too much credit as if they are deeply analyzing notes and card history, when in actuality they are glancing at a card for an average of 30 seconds. My theory is:

Some bean counter raised his hand during an exclusive PSA pop control meeting and said:

"Guys, why are we just doing pop control on a few select key cards from the 80s and 90s, like the Rickey Henderson and Ken Griffey Jr. rookies? Instead, why don't we just apply our same pop control measures to all 80s/90s cards, and then people will be sending in countless Bo Jackson 1990 Score cards (and the like) in droves, gambling on grading to hit a PSA 10. But we're the only ones that hold the key to allow that to happen, so the less 10s we give out, the more the 10 value rises, and the more people gamble to get a PSA 10. You know, like the unregulated gambling we're setting up with PowerPacks™??"

There was brief silence for a moment in the meeting, and then everyone applauded.

Don't tell the graders, but just have Johnson in engineering develop an algorithm filter that flips any PSA 10 grade to a 9 in the QA stage, based on a certain pop control percentage we bake into the system. Easy, and the graders themselves would never even know, and even if they did, they have an NDA.

Because, you know, money.

[SLGSports]

lol, ironically, if Johnson in engineering would +1 to grades, PSA would have a lot of happy customers... and would get a lot more money via upcharges too