PSA 8.5 Grading Anomaly December 2016 - January 2017

Hello all. I hope everyone has a fantastic holiday week coming up here with it being Thanksgiving tomorrow for those of you in the states. Happy Thanksgiving! I hope for the long weekend to generate some thought and discussion on something that has been bothering me for a while now. This has been on my mind for about a year now, and I have been delaying putting in the work to further explore it for too long but the time is now.

Being that I recently posted in my buy thread about regrading I decided that I should further explain my stance here and explain why in my eyes it has validity in certain circumstances ESPECIALLY in the case of the dreaded PSA 8.5. This post has a lot more to do with the 8.5 actually than regrading. I really want to avoid anyone thinking I adorn tin foil hats as well because I tossed out the crazy theory that PSA had a short stint where PSA 8.5’s were handed out unreasonably, unfairly and in astronomical proportions never before seen. Nobody searches for them, nobody collects them or wants to buy them, they often times go for sub PSA 8 prices, the grade is too long to properly title and still fit all the other relevant bits in an eBay listing, the label looks ugly and messy with all those extra characters. I could go on and on.

So please see all my data below from my personal submissions. A quick word on my methodology. I only looked at cards graded PSA 8, 8.5, 9 or 10 in both my subs and in pop reports. All cards below grade 8 were excluded from the samples. You can see that my sample sizes are reasonably large and that I compare them against entire populations of large sets further below.

As you can see above I have graded in my tenure 2,990 total cards PSA 8-10. Of those, 34 have achieved a PSA 8.5. I had a two month stretch consisting of only ~10% of my total submissions where I received a whopping ~76% of my 8.5’s overall. Not only that, but regrading some of these cards led to the below results.

Do note that I have graded 34 PSA 8.5’s yet I only show results of 17 regrades. There are multiples reasons for this. Certain cards were not regraded due to the low value they would yield in PSA 8 or 9 (e.g. base II). They were cracked and added to binder sets. Certain cards I have graded in such high quantities that I couldn’t reliably say whether it went to an 8/9/10. What I can certainly say is that exactly ZERO regraded a PSA 8.5. I know this because of the 34 8.5’s I have received them all on unique card variants except two. Those two cases resulted when both 8.5’s for the given card were received in the same submission.

To put it into greater context here is the PSA 8.5 rates for a few select sets. Choosing Jungle Fossil and Chinese as they are ones that affected me the most. Choosing Base and Burning shadows for old and new large pools of data. What would be perfect is to have these data sets for my given timelines. Unfortunately I do not have that data, all I have is my own and my visual anecdotal evidence whereby a lot of the 8.5’s I see on eBay were submitted in the same time frame having the same cert # range.

Further things to add and some of my own interesting observations.

  • I had 17 submissions before the anomaly window, 6 during and 37 since.
  • My ship dates during the window were from 12-2-16 to 1-19-17.
  • I speculate the anomaly started after 11-16-16 and ended before 2-2-17 as I have large unaffected submission bookmarking my affected submissions.
  • My affected cert #'s are 26,3XX,XXX through 26,8XX,XXX but I feel the anomaly includes all residing in the 26 million bracket.
  • Interesting uptick this year in my personal PSA 10 rate while PSA 8 held steady. I always submit cards that are PSA 9+ worthy in my eyes. Many are pack fresh, but many are also mint+ looking ones that I buy raw as singles or in large collections. I see 9’s as breakeven, 8’s as losses, 10 as my profit and 8.5’s as a PITA that needs to be regraded.

I decided to post this in the article section as I feel it best suits it, but mods please feel free to move if necessary.

10 Likes

If anyone had any large submissions that were graded and marked shipped during the window 11-16-16 through 2-2-17 I would love to see your raw data and analyze to add it to my work above. Also if there are any comments or questions feel free to share here in the thread or PM me.

Another thing that drove me to do this investigation was my recent acquisition of two PSA 8.5 Base 1st Charizards. You guessed it… both having 26,XXX,XXX cert #'s.

2 Likes

I know some members scrape the population pages, but they may not have data that far back.

Many respects to you for this level of thoroughness and clarity. great work and thank you!

Maybe some meticulous, obsessive new employee(s) in there who just couldn’t quite give some cards a 9 and did so for awhile before management noticed the increased use of the “.5”!

1 Like

Could we just give you a number of how many 8.5s we received in that window?

Or do you need to know the grade of every card from each order we sent off?

Ideally I could just get all the raw grades on any orders from that window along with the ship date of the order.

Thinking about it more now though… Perhaps someone much smarter than I could actually save everyone a lot of time and run a scripted cert# lookup for all 1 million cards graded between 26,000,000 and 26,999,999. If I could just get a file with cert and grade that’d be enough but with card info also I could at least separate Pokemon from non Pokemon.

If anyone can deliver that to me pick anything $25 or less from my store and I’ll ship it to you free or get $25 credit towards whatever. Whoever could do this would be the real MVP. @skiwi I know you scrape the Pokemon Pop Report data frequently. Is this something easily done?

Great contribution @gottaketchumall I noticed this too but never put together the data. All I can say is I’m glad to see them all but gone. Aside from my opinion that they hurt the secondary market, the 8.5 grade is such an undefined position I honestly believe most graders wouldn’t know the difference.

I believe my buddy @pokemonmaster556 had a fair amount of 8.5s come back. The last submission of 2016 I had was Nov 10-15th. I took a break from submissions during the following period after seeing the sharp rise in the half-point grade. All my orders since this June have been 9s or 10s

I did have a run of those horrid 8.5s lol. I believe I stopped grading for a while cuz of that guy… maybe 20 or so in 2 submissions got that 8.5… if have to look at it and count em all.

1 Like

This is sensational work. I specifically remember this grade being prominent while people were grading evolutions. I wonder what made them flush out so much. Maybe they all looked a policy during this period where if grader 1 gave an 8 and grader 2 have a 9 an 8.5 was given.

Interesting work. Well done.
Not sure how much it’ll help seeing as this past year I had much fewer submissions but maybe this will help some

Dec 2016 7(8.5s)/389
Mar 2017 0/120
May 2017 2/407
July 2017 1/301
Sep 2017 1/100
Oct 2017 0/100
Those are all the ship back dates.

Eddie graded thousands of cards during that time and got so many 8.5s that he called me all the time to complain about it. He could supply you with a big sample lol.

Keep in mind that 9o-95% of mine I thought were strong 10s so that may throw off the percentages a bit.

1 Like

Awesome thanks for adding these! We are getting to a point now where I think it is becoming exceedingly safe to say that it wasn’t simply a standard ebb and flow in grading whereby 8.5’s naturally came at an increased amount. It does indeed seem to have been an actual paradigm shift in the way that they graded cards for whatever reason.

I plan to dig up my old statistics book and run some actual statistical models on these large data sets to actually prove that these numbers don’t fit together in “normal” random data sets.

I wonder if PSA would ever have any discussion about this as it seems it must have begun and/or ended with some discussion of the lack or excess of the 8.5 grade being given out.

I suppose this is easily done, however in the naive method you are going to grab PSA’s attention. The naive one would be sending 1 million requests for the data, it kind of only seems reasonable if there is a bulk way of doing this, and I’m not sure if that exists.

For my scraper I scrape per set and hence why it doesn’t seem to draw lots of attention.

Ha @skiwi yeah I figured there would likely be a good way to go about it and a poor way to go about it. Maybe I’ll just e-mail them and ask if they have a database file they could share or something haha. They obviously have one with all cert #'s. Doubt that will get anywhere but worst they can say is no. I mean it’s public information anyways anyone can search it. I plan to call them at some point and ask them what was up with the 8.5 grades when I get together a little more proof that it definitely wasn’t just random that so many came all at once. Again I doubt that will lead to any kind of an actual response or admission but it can’t hurt to ask.

I see two possible ways you could go… One is scraping the page itself, the other would be looking to which endpoint the app calls when checking for a cert number, both however seem to be meant for single queries though.

And it’d be interesting to see what kind of reply you would get, as long as you don’t remind them that they need to grade more 8.5’s! :grin:

Got back a submission about a week ago where I sent a Japanese Jolteon GS promo that I took right out of the wrapper and was perfect everywhere. Never been so sure of a 10 in my life, the card was perfect… comes back 8.5. Lord help us all…

1 Like

I don’t have an issue with the grade at a fundamental level or anything personally. I just have an issue with how they handed them out like Halloween candy for a month or two apparently. If a card is truly an 8.5 then so be it. It is what it is I guess. But the fact that so far I have regraded 17 or more of them with zero repeat-ability is a huge problem. If anyone has ever regraded one to achieve an 8.5 the second time around I would love to hear about it. Especially if it came from DEC 16-JAN 17.

I actually have a 6.5 Articuno that I really don’t get either. Honestly any half grades down that low you are really splitting hairs IMO. I doubt any half grades all the way through the scale are really repeatable at all. I would love to see if they have any internal metrics on it. They must have certain times where they submit single small batches of cards across several or all graders. It really would be fascinating to see some of those results. Too costly and time consuming for anyone to do it personally and of course you would have the issue of potential handling damage over time.

1 Like

I’ve never regraded so I’m just curious, if a grade comes back higher than the former grade do they refund anything (grading service) considering you had to pay extra money for them to get it right when it should’ve been right the first time?

Yes, this is interesting. I’ve never regraded a card but considering the ones I have probably deserved 8s or at best weak 9s I don’t mind getting them;)

@pokematt if you crack and regrade then the answer is an obvious no. If you review it inside the case and it becomes a higher grade though I am not entirely sure. They definitely charge the grading fee if it doesn’t change grades but they may waive it if it does indeed go higher. I think that is a rare occurrence.

@garyis2000 I understand your distaste for regrading in general as a practice. I am interested in your thoughts on regrading the 8.5’s from this time period I mentioned though? Clearly they gave them out at higher rates as myself, you and Ed noticed. Knowing that and the information I showed on how high the 8.5 rates were compared to historical rates do you think it is improper to crack and regrade the 26,XXX,XXX 8.5’s?

1 Like

Just what I was looking for, thanks @gottaketchumall