Question regarding resubmitting your graded cards to PSA.
Yankees70
Posts: 427 ✭✭✭
My best friend recently had 15 vintage cards (pre 1981) graded by PSA. He received ample PSA 7's for cards that he and many others thought had a decent chance of receiving a 9. He recently returned to the hobby like I have after being away for ample years. He submitted hundreds of cards last decade and knows how to grade. He wants to resubmit some of them to PSA immediately. My question is should he wait and if so how long?
PSA claims the graders have no idea who's cards they are grading and if that's the case it won't hurt to resubmit them. My buddy does not care about the grading fees. Still I have recently heard multiple dealers at conventions say that resubmitting the same cards within a short period of time is a waste of money as the chances of any grades changing for the good is minimal at best. .
I would appreciate your opinions
Comments
I would recommend cracking out and re-subbing to PSA or even better SGC for comparison….
If your buddy doesn't care about the fees, crack them out and send them back in immediately.
When they come back, he will have the results he needs to proceed.
If the cards don't come back with better grades, I would suggest trying a different grading company.
If the cards "need" to be PSA graded, keep sending them in until the results get better, or he goes broke.
If he's just getting back and initially into grading then he has a long, difficult road on the path to determining which he thinks is best. It's weird out there these days. You can find plenty of threads here about confusion on recent grades and others about people moving to other companies.
He's going to have to try some different things and figure out what he prefers. Good luck to him and I hope his grades get more in line with his expectations.
In my opinion, it doesn't matter how soon or delayed they get resubmitted. If it's not a '52 Mantle or T206 Honus, nobody will remember those cards. My opinion is grading results depend heavily on whether you get a longtime vintage grader handling your order or a new hire brought in after the Covid shutdown. If your friend submitted a lot of cards in years past, then they should have a good idea of the traditional PSA grading standards and my comment would apply. If they have limited submission history, then I'm not sure if your friend's grading expectations are realistic or not and therefore resubmitting may not result in better grades. It could be worth a shot to resubmit them and hope to get an experienced vintage grader. Unfortunately, the vast majority of the graders in the room are relative newbies due to the massive surge in volume PSA has experienced. You probably have a 10-20% change of getting a vintage grader.
I appreciate the opinions guys. Question - Is it correct that the PSA graders NEVER have an idea on who submitted the cards? If that's correct it would make sense to resubmit them as he could get a bump on some of them. You would think they have some idea otherwise how would they explain the difference in grades?
Example
A collector submits 15 cards and receives ten 7's and five 8's.
He then immediately sends all 15 back and this time he receives five 7's, eight 8's, and two 9's.
This would make PSA look very inconsistent which is not a good sign for the company.
Every grading company is inconsistent, grading is somewhat subjective and all people make mistakes.
I don't know for sure, but why would they care who submitted or resubmitted the cards? I don't think they have the time to try to track every single card sent in?
PSA has been overloaded with submissions for 2(?) years and it doesn't seem to be slowing down. They are in business to make a profit.
Either you "need" to have your cards in a PSA holder or you don't, if you can't get the results you feel you deserve with PSA, try somebody else.
I'd make a very definitive guess that PSA would love it if any collector would keep cracking out cards they recently graded and send them in to be graded again. They probably wish everyone would do that 2 or 3 times for every card in their inventory.
With literally millions of the same cards out there it would make almost 0 sense to have a person check scans from old grades before regrading. It would be a colossal waste of time that would be better spent doing actual grading and (more importantly) making a profit for PSA by efficiently using the time on premises to do their job.
At least if I were running the show...
Most people new to submitting overestimate the condition of the cards they submit so those results are not that surprising.
That said, with the volume of cards submitted to PSA, I doubt cards cracked out and resubmitted would receive anything but a fresh look the second time around. But I wouldn't expect markedly different results.
Collecting 1970s Topps baseball wax, rack and cello packs, as well as PCGS graded Half Cents, Large Cents, Two Cent pieces and Three Cent Silver pieces.
I'm pretty sure that they don't care about optics, given the last 5 years of issues and only seeing submission numbers increase exponentially.
There are vids on the YouTube of guys cracking out serial numbered cards and resubmitting them and getting bumps in grade. Some say they have resubbed multiple times, until they got the grade they were hoping for.
When paying for a service where thousands of dollars are at stake, inconsistencies cannot take place and subjectivity should be held to a published standard.
@Yankees70
I would way until you see clear and convincing evidence of improvement in the grading of vintage before you resubmit.
I would think the average submission from Joe off the Street is comparatively anonymous but I would also think that high volume customers are not.
Curious about the rare, mysterious and beautiful 1951 Wheaties Premium Photos?
https://forums.collectors.com/discussion/987963/1951-wheaties-premium-photos-set-registry#latest
This is not the case.
Psa has become a total joke with their grading unbelievable. I’ve been collecting for over 55 years, been sending in cards since PSA basically started, they have gone off the rails now in grading.. I’ve talked to dealers at the last national that have said the same thing. They’re over correcting the past BS. Ive had enough not another friggen dime. It’s a joke. I will post my cards when I receive them. PSA has become ridiculous.. 9’s are now 7’s no matter how nice … they play it safe and hand out a 7.
I do enjoy poking fun at PSA, but I think what we're observing is more of a reflection of the hobby and some of the lumps you experience with scaling the grading business.
Then there's the "AI" story about grading companies using "AI" for grading, which is complete BS. I can't stand the term Artificial Intelligence anymore and I'll keep my opinion short on this.
You can use Supervised Learning to determine algorithmically whether a card is a 10... with the assumption that you can generate a perfect image of the issue you want to use for training. If you're using actual cards for this... well... Chicken. Meet Egg.
You can use Unsupervised Learning to determine algorithmically if there are anomalies (again with the assumption that your trained dataset is pristine). The trick is determining with statistical analysis what impact on overall quality of the card those anomalies add up to. Good luck with that. It's like thinking you can take 5 components of human beauty from different individuals and compile them into a single example and expect good results. You get more Trolls than Models.
I think we're seeing the results as we should expect.
If anyone is interested here's a direct path to the maths.
https://en.wikipedia.org/wiki/K-means_clustering
An image is just a multi-dimensional array.
I highly doubt TAG or PSA or whoever is using anything like that for any image analysis they might be doing. I suspect they are using something like a Siamese (or Twin) NN. I am only guessing, but this is based partly on how much they appear to be offering in salary for the relevant positions. Note: there's also Image GPT which uses a LLM -- also having the same candidate bias hurdle.
Anyways and regardless. There are lots of great tools out there, but they are primarily for the purpose of classification. I don't know exactly why I decided to ramble about AI, but I see it coming up so frequently in discussions about grading and I guess it just boiled over for me. There's no one doing AI grading successfully.
TAG purports (themselves) to have been awarded 100+ patents related to AI grading. I've not been able to find them all, but the ones I did find were what I expected. Obviously they know that very few collectors are going to look up their patents... but we all should as it is enlightening. It also reflects what I mentioned earlier. They are classifying images based on the observance of a range of defects... Those defects in each case (centering, corners, surface, edges) are subtracted from 1000 and that determines the grade simply based on thresholds in each of those categories. Easy to tell if a card is a 10, but not so easy to determine if a card is a 6 or 8 as there's 1 perfect image and 6,250,000 9s and 3,906,250,000 8s and as you can see the permutations of 7s, 6s, 5s, etc. they grow exponentially. Unless TAG and others are behind the NVDA sales I don't know how they are grading.
I failed to keep it short. This is just my opinion... It's not worth much.
Well .. you’re speaking way above my head
). I do understand the just of your post. If.. I don’t exactly know if they’re using AI, they shouldn’t on vintage cards from 70’s and earlier. Here’s my theory. You can’t use a modern day technology to grade/judge a card that was printed in the day of rotary phones.. if you’re picking up what I’m laying down. I guess what I’m getting at, is every card printed today from the box/pack however should be a 10 with today’s technology.. not sure if there is actually a 10 from the 70’s and earlier, that would compare to a modern day 10 today.