What’s going on with TAG?

[discodanman45]

Originally Posted by hoi_ming

Is it Pokecollect or the Omega folks propping up TAG? I forget.

But obviously I have no idea who's subbing all the Pokemon slabs, but it's not a stretch to say it can be a mix of folks, PC folks, streamers, big mystery packers.

Who knows maybe in the end it was Omega the whole time.

Probably a bunch of mystery box and Whatnot streamers. This is not PC folks. Pokémon went from non-existent to 100,000 slabs very quickly. This is mainly some big influencers working with TAG. The PC folks are a very small portion of that 100,000 slabs.

[rfgilles]

Originally Posted by hoi_ming

Is it Pokecollect or the Omega folks propping up TAG? I forget.

But obviously I have no idea who's subbing all the Pokemon slabs, but it's not a stretch to say it can be a mix of folks, PC folks, streamers, big mystery packers.

Who knows maybe in the end it was Omega the whole time.

You missed the good ole days when people would say TAG was a technology only company.

[rfgilles]

Originally Posted by pumpkinescobar

Do you know if the trend of sports cards submissions to TAG has declined from the early days or have sports just not seen the kind of rapid growth as TCG? I've always thought that it would be hard for TAG to take any sort of sports card marketshare without having the ability to grade rookie patch autos.

This....

[Scottish Punk]

Originally Posted by rfgilles

This....

Yeah gemrate data on sports cards has taken a dip even for PSA. Not much, but it is change from the up and up and up for the past two years.

TAG isn't going to be respected in the sports card world for many until they can prove they can do vintage. Not to mention the other stuff people have mentioned. It is what killed HGA. When they couldn't authenticate any big boy cards correctly, no point in trusting them with anything.

TCG and Pokemon specifically is definitely driving ultra modern submissions though. I know nothing of that market. I am just floored that we can get 500-600k of Pokemon graded every month. Print runs must be through the roof on sets. How many sets are released per month?

[rfgilles]

Originally Posted by discodanman45

Probably a bunch of mystery box and Whatnot streamers. This is not PC folks. Pokémon went from non-existent to 100,000 slabs very quickly. This is mainly some big influencers working with TAG. The PC folks are a very small portion of that 100,000 slabs.

TAG is grading ~5K cards weekly (total), which isn't quick. PSA does >100K TCG cards a week.

[JWBlue]

Originally Posted by inaka

That's so funny because I noticed the same thing. I bought my first TAG baseball slab from eBay and it arrived on Saturday. From the photos, I thought the etched "label" area would be great, but that's typically because the photos always have a black background making the text perfectly easy to read. But in-person, it's not as nice in the label area and much more difficult to read white text on a clear background.. The slab plastic itself is ULTRA crystal clear, and looks great.

I checked out the TAG slab for about 5 mins before cracking it open to send to PSA. I've cracked tons of cards in the past, but this was my first time cracking a TAG slab. The plastic of the slab is beautiful and crystal clear, but is also is so brittle it shattered into multiple small pieces unlike PSA, BGS and SGC slabs I've cracked in the past. When separating the layers it literally shattered the top layer, but thankfully the card was fine.

I have to keep this in mind if I ever get another TAG slab to crack and re-sub.

We would love to hear how PSA grades it.

[inaka]

Originally Posted by JWBlue

We would love to hear how PSA grades it.

Sure, no problem. I'll post the results when I submit to PSA.
(Currently stock piling until a get 20 cards for a bulk PSA order)

It was a TAG MINT 9 (947) on the cheap, and I just want a PSA 9.
There's a long shot for a PSA 10, but we'll see in 3-4 months and I'll post the results.

[hoi_ming]

Originally Posted by inaka

Sure, no problem. I'll post the results when I submit to PSA.
(Currently stock piling until a get 20 cards for a bulk PSA order)

It was a TAG MINT 9 (947) on the cheap, and I just want a PSA 9.
There's a long shot for a PSA 10, but we'll see in 3-4 months and I'll post the results.

What was the main issue from the TAG report, was it the centering?

If it was still within 60/40 technically it could still be a PSA if no other big issue.

Is 3-4 months due to the turnaround time?

[JeremyNick]

Originally Posted by Scottish Punk

Yeah gemrate data on sports cards has taken a dip even for PSA. Not much, but it is change from the up and up and up for the past two years.

TAG isn't going to be respected in the sports card world for many until they can prove they can do vintage. Not to mention the other stuff people have mentioned. It is what killed HGA. When they couldn't authenticate any big boy cards correctly, no point in trusting them with anything.

TCG and Pokemon specifically is definitely driving ultra modern submissions though. I know nothing of that market. I am just floored that we can get 500-600k of Pokemon graded every month. Print runs must be through the roof on sets. How many sets are released per month?

Pokémon is releasing English sets every 3 months. However most English sets are a combination of two Japanese sets that have released in the previous 2-3 months.

So on average you are getting a set released every month and those sets can stay in release rotation and available for 2-3 years.

You can still find 2021 Pokémon English sets getting stocked at Wal Mart now.
So it’s really just a constant flow of new and older cards constantly gettint ripped and submitted.

[Scottish Punk]

Originally Posted by JeremyNick

Pokémon is releasing English sets every 3 months. However most English sets are a combination of two Japanese sets that have released in the previous 2-3 months.

So on average you are getting a set released every month and those sets can stay in release rotation and available for 2-3 years.

You can still find 2021 Pokémon English sets getting stocked at Wal Mart now.
So it’s really just a constant flow of new and older cards constantly gettint ripped and submitted.

So the print runs are enormous? How much of it is really rare? Just amazing demand can sustain that much.

[hoi_ming]

Originally Posted by Scottish Punk

So the print runs are enormous? How much of it is really rare? Just amazing demand can sustain that much.

I didn't realize this until recently but Pokemon is the biggest franchise in the world, like bigger than Mickey, bigger than MCU.

But yeah i think the fact that it's a TCG and the cards are actually used to play helps witb demand. Even common cards have some utility.

[rfgilles]

Originally Posted by inaka

Sure, no problem. I'll post the results when I submit to PSA.
(Currently stock piling until a get 20 cards for a bulk PSA order)

It was a TAG MINT 9 (947) on the cheap, and I just want a PSA 9.
There's a long shot for a PSA 10, but we'll see in 3-4 months and I'll post the results.

Delete

[JeremyNick]

Originally Posted by Scottish Punk

So the print runs are enormous? How much of it is really rare? Just amazing demand can sustain that much.

Print runs are large but it’s very unique.

First there is a very large base of people that play the game, and also a very large base of collectors who only collect raw cards.

But for modern cards in PSA 10, you are usually capped at under $2k So cards don’t get terribly expensive, at least impaired to sports.

The most popular and expensive card from the last complete era of sets is ~$1200 in a PSA 10 right now, with a pop around 12,400. That card was released in 2021.

Most of the really popular, rare cards can be had for under $800 in PSA 10.

[inaka]

Originally Posted by hoi_ming

What was the main issue from the TAG report, was it the centering?

If it was still within 60/40 technically it could still be a PSA if no other big issue.

Is 3-4 months due to the turnaround time?

Here's the centering chart for it, (the back centering is: 52/48)



But visually to me, it looks off. Yes, the PSA "specs" say 60/40 is still within septic for a PSA 10, but we'll see. I have a feeling it will get a PSA 9 at best, but again, not sure.

In terms of PSA turnaround time, when submitting bulk today, the PSA estimates have a return of sometime in October. But I haven't even submitted it to PSA yet, just cracked it out.

[hoi_ming]

Originally Posted by inaka

Here's the centering chart for it, (the back centering is: 52/48)



But visually to me, it looks off. Yes, the PSA "specs" say 60/40 is still within septic for a PSA 10, but we'll see. I have a feeling it will get a PSA 9 at best, but again, not sure.

In terms of PSA turnaround time, when submitting bulk today, the PSA estimates have a return of sometime in October. But I haven't even submitted it to PSA yet, just cracked it out.

So centering wasn't the issue here since it's within 55/45 on the front.

I guess it was the front surface defects?

I had to help a friend so a PSA other recently and it was much faster than their estimate. But their estimate was ridiculously long so it's a big buffer.

[inaka]

Originally Posted by hoi_ming

So centering wasn't the issue here since it's within 55/45 on the front.

Centering is definitely an issue, well, in my onion anyway. The design of this particular card with two (double) thin borders on the right side really amplifies how off-center it is to my eye, being much wider on the top and left sides.

Here's the card if you'd like to see what I mean:
(it['s a chrome card which is why the photo is so dark)



https://my.taggrading.com/card/m8428686

For reference, I come from a design background and you guys probably know about the Golden Ratio and all that. Well, in terms of top-to-bottom centering for borders, typically human beings actually are accustomed to a slightly larger/wider border on the bottom based on how we view things in a portrait orientation. It started in the Victorian era when art was hung at an angle, and thus artists used bottom-weighted mattes/frames that were slightly larger at the bottom to compensate. Then we then got used to it, even when they were hung on the wall flat. So in portrait orientation, the human eye is still actually naturally drawn to the area slightly above the actual center, and not the true center.

I think it brings up an interesting point about AI in general. A computer cannot always determine the "center" on certain card designs. So with AI grading, I'm sure there's even a debatable issue on true centering as this is based on the layout of each card design.

So when the bottom border is smaller like it is here, even just 54/46, it really stands out in a bad way. That asymmetrical double-thin border design on the right, also amplifies the smaller margin/border on the right of this card as well, making the centering far from GEM MINT (PSA 10) to me, even if TAG says it's within PSA spec.

So that's why I think this card as best will be a PSA 9 and not a PSA 10, but again, we'll see in the months ahead when I submit to PSA.

[rfgilles]

Originally Posted by inaka

Centering is definitely an issue, well, in my onion anyway. The design of this particular card with two (double) thin borders on the right side really amplifies how off-center it is to my eye, being much wider on the top and left sides.

Here's the card if you'd like to see what I mean:
(it['s a chrome card which is why the photo is so dark)



https://my.taggrading.com/card/m8428686

For reference, I come from a design background and you guys probably know about the Golden Ratio and all that. Well, in terms of top-to-bottom centering for borders, typically human beings actually are accustomed to a slightly larger/wider border on the bottom based on how we view things in a portrait orientation. It started in the Victorian era when art was hung at an angle, and thus artists used bottom-weighted mattes/frames that were slightly larger at the bottom to compensate. Then we then got used to it, even when they were hung on the wall flat. So in portrait orientation, the human eye is still actually naturally drawn to the area slightly above the actual center, and not the true center.

I think it brings up an interesting point about AI in general. A computer cannot always determine the "center" on certain card designs. So with AI grading, I'm sure there's even a debatable issue on true centering as this is based on the layout of each card design.

So when the bottom border is smaller like it is here, even just 54/46, it really stands out in a bad way. That asymmetrical double-thin border design on the right, also amplifies the smaller margin/border on the right of this card as well, making the centering far from GEM MINT (PSA 10) to me, even if TAG says it's within PSA spec.

So that's why I think this card as best will be a PSA 9 and not a PSA 10, but again, we'll see in the months ahead when I submit to PSA.

If it is within the defined criteria for a PSA 10 (centering) it should be a 10 irrespective of subjective impression.

The card didn't gem because of the back corner score (937). However, none of the corners had a score this low.....

[hoi_ming]

Originally Posted by inaka

Centering is definitely an issue, well, in my onion anyway. The design of this particular card with two (double) thin borders on the right side really amplifies how off-center it is to my eye, being much wider on the top and left sides.

Here's the card if you'd like to see what I mean:
(it['s a chrome card which is why the photo is so dark)



https://my.taggrading.com/card/m8428686

For reference, I come from a design background and you guys probably know about the Golden Ratio and all that. Well, in terms of top-to-bottom centering for borders, typically human beings actually are accustomed to a slightly larger/wider border on the bottom based on how we view things in a portrait orientation. It started in the Victorian era when art was hung at an angle, and thus artists used bottom-weighted mattes/frames that were slightly larger at the bottom to compensate. Then we then got used to it, even when they were hung on the wall flat. So in portrait orientation, the human eye is still actually naturally drawn to the area slightly above the actual center, and not the true center.

I think it brings up an interesting point about AI in general. A computer cannot always determine the "center" on certain card designs. So with AI grading, I'm sure there's even a debatable issue on true centering as this is based on the layout of each card design.

So when the bottom border is smaller like it is here, even just 54/46, it really stands out in a bad way. That asymmetrical double-thin border design on the right, also amplifies the smaller margin/border on the right of this card as well, making the centering far from GEM MINT (PSA 10) to me, even if TAG says it's within PSA spec.

So that's why I think this card as best will be a PSA 9 and not a PSA 10, but again, we'll see in the months ahead when I submit to PSA.

With regards to AI centering, it's not really AI it's more that's it's automated machine grading, there is some human input involved before the grading, but the scan and the grading is automated with human QA, i think of it like facial recognition. Here's some insight from the CEO he shared in the Discord

'Number 2: We have also greatly advanced Centering. Let me tell you a little about challenges associated with Centering. The Centering of a Bordered card is not an issue, the Centering of a card with 1, 2, or 3 markers on a card surface such as a player name, MFR logo, team name, etc, is no longer an issue.

Number 3: The centering on a completely borderless card, without any identifiers on the front of the card (a very small percentage of cards) requires a number of technical and manual annotation. For example, from a technical standpoint, we search the database of cards graded for every similar card, for every card from the same set, and for various card characteristics that enable the system to identify at least 1 centering point on 1 border to enable us to extract comparable measurements for the remaining borders and thus determine the image centering on the card. With respect to surface defects, false positives are virtually always associated with very small surface defects that may include small scratches, dents, and print defects. Each of which, and in the aggregate, impact on a card grade very slightly. We continue to improve these minor grading elements and in the interim, QA assures us of not relying completely on the system for elements of surface grading that have not reached the "'stand-alone" technology comfort level that we require, and our submitters expect and deserve. Our QA, although limited in application, is performed by very high level and well-experienced grading experts, one of which has grading experience of approximately 3 million cards."

[rfgilles]

Originally Posted by hoi_ming

With regards to AI centering, it's not really AI it's more that's it's automated machine grading, there is some human input involved before the grading, but the scan and the grading is automated with human QA, i think of it like facial recognition. Here's some insight from the CEO he shared in the Discord

'Number 2: We have also greatly advanced Centering. Let me tell you a little about challenges associated with Centering. The Centering of a Bordered card is not an issue, the Centering of a card with 1, 2, or 3 markers on a card surface such as a player name, MFR logo, team name, etc, is no longer an issue.

Number 3: The centering on a completely borderless card, without any identifiers on the front of the card (a very small percentage of cards) requires a number of technical and manual annotation. For example, from a technical standpoint, we search the database of cards graded for every similar card, for every card from the same set, and for various card characteristics that enable the system to identify at least 1 centering point on 1 border to enable us to extract comparable measurements for the remaining borders and thus determine the image centering on the card. With respect to surface defects, false positives are virtually always associated with very small surface defects that may include small scratches, dents, and print defects. Each of which, and in the aggregate, impact on a card grade very slightly. We continue to improve these minor grading elements and in the interim, QA assures us of not relying completely on the system for elements of surface grading that have not reached the "'stand-alone" technology comfort level that we require, and our submitters expect and deserve. Our QA, although limited in application, is performed by very high level and well-experienced grading experts, one of which has grading experience of approximately 3 million cards."

In the case of the Rickey card, centering is straightforward given the borders.

[hoi_ming]

Originally Posted by inaka

Centering is definitely an issue, well, in my onion anyway. The design of this particular card with two (double) thin borders on the right side really amplifies how off-center it is to my eye, being much wider on the top and left sides.

Here's the card if you'd like to see what I mean:
(it['s a chrome card which is why the photo is so dark)



https://my.taggrading.com/card/m8428686

For reference, I come from a design background and you guys probably know about the Golden Ratio and all that. Well, in terms of top-to-bottom centering for borders, typically human beings actually are accustomed to a slightly larger/wider border on the bottom based on how we view things in a portrait orientation. It started in the Victorian era when art was hung at an angle, and thus artists used bottom-weighted mattes/frames that were slightly larger at the bottom to compensate. Then we then got used to it, even when they were hung on the wall flat. So in portrait orientation, the human eye is still actually naturally drawn to the area slightly above the actual center, and not the true center.

I think it brings up an interesting point about AI in general. A computer cannot always determine the "center" on certain card designs. So with AI grading, I'm sure there's even a debatable issue on true centering as this is based on the layout of each card design.

So when the bottom border is smaller like it is here, even just 54/46, it really stands out in a bad way. That asymmetrical double-thin border design on the right, also amplifies the smaller margin/border on the right of this card as well, making the centering far from GEM MINT (PSA 10) to me, even if TAG says it's within PSA spec.

So that's why I think this card as best will be a PSA 9 and not a PSA 10, but again, we'll see in the months ahead when I submit to PSA.

This is what i got with the edge centering tool

[inaka]

AI centering is an interesting discussion point, but I guess I might not be expressing myself well. My reference is what this TAG card will actually receive from PSA based on my real-world experience, and not based on listed specs.

For example, PSA also has a half point scale. So in theory they can give out a 6.5 or a 7.5, etc. But a real world example is, they use this primarily for vintage, and if you submit a 2024 ultra-modern card, you will likely NEVER receive a PSA 5.5 Grade. PSA will instead simply give out a PSA 5 or PSA 6 on the card, not a PSA 5.5 grade, even though their specs say they do offer half grades. (PSA 8.5s? Sure, that's the one they still wheel out on occasion even for ultra-modern, but even that is once in a blue moon). With SGC, they also offer half grades, but they actually use them, in my real world experience. Both companies specs say they offer half grades, but only one company actually uses them as intended.

There's no PSA spec out there saying they only give out half grades to vintage, but that's what PSA does. They just don't give mid hand grades for ultra modern. I'm sure someone can find the rare isolated example, but that's the exception to the rule.

So with centering, since these are human eyes at PSA, in this real world example with the TAG Rickey card, I don't see that TAG card getting a PSA 10 based on centering, even if it falls within their listed 60/40 PSA specs. Just don't see it happening, based on the multiple points I outlined previously. And maybe that highlights the difference with AI and human grading.

Hope that makes sense.

[rfgilles]

Originally Posted by inaka

AI centering is an interesting discussion point, but I guess I might not be expressing myself well. My reference is what this TAG card will actually receive from PSA based on my real-world experience, and not based on listed specs.

For example, PSA also has a half point scale. So in theory they can give out a 6.5 or a 7.5, etc. But a real world example is, they use this primarily for vintage, and if you submit a 2024 ultra-modern card, you will likely NEVER receive a PSA 5.5 Grade. PSA will instead simply give out a PSA 5 or PSA 6 on the card, not a PSA 5.5 grade, even though their specs say they do offer half grades. (PSA 8.5s? Sure, that's the one they still wheel out on occasion even for ultra-modern, but even that is once in a blue moon). With SGC, they also offer half grades, but they actually use them, in my real world experience. Both companies specs say they offer half grades, but only one company actually uses them as intended.

There's no PSA spec out there saying they only give out half grades to vintage, but that's what PSA does. They just don't give mid hand grades for ultra modern. I'm sure someone can find the rare isolated example, but that's the exception to the rule.

So with centering, since these are human eyes at PSA, in this real world example with the TAG Rickey card, I don't see that TAG card getting a PSA 10 based on centering, even if it falls within their listed 60/40 PSA specs. Just don't see it happening, based on the multiple points I outlined previously. And maybe that highlights the difference with AI and human grading.

Hope that makes sense.

It is not a matter of "AI centering", more being objective over subjective.

We will see when it is graded by PSA. If it gets a 10, then in the opinion of the grader the centering is fine.

[inaka]

Originally Posted by rfgilles

It is not a matter of "AI centering", more being objective over subjective.

We will see when it is graded by PSA. If it gets a 10, then in the opinion of the grader the centering is fine.

You're right. If it comes back a PSA 10, then the centering wasn't an issue. But it's PSA. It could also get a PSA 6 for no reason at all. LOL

Fairly recently I had a card come back graded from PSA that came back as a PSA 6. It looked perfect to me. I cracked it open, sent it right back to PSA and the exact same card came back a PSA 10.

I've done this is the past with PSA 8s that became PSA 10s, but never a PSA 6.

This is the appeal of AI grading for me. To rule out the ridiculous inconsistencies of grading. So that's a big appeal of AI grading like TAG being implemented, but then AI too will introduce unintended consequences, and the intricacies of vintage, die cuts, or centering without borders, etc. are extremely challenging for AI grading companies like TAG as well. So it's a give/take in some scenarios.

[hoi_ming]

Here's random example of someone grading 100 Pokemon with TAG. Don't think it's for mystery packs or streaming.

https://www.instagram.com/reel/C-biS...ljNWg3cGpmcg==

[rfgilles]

Originally Posted by hoi_ming

Here's random example of someone grading 100 Pokemon with TAG. Don't think it's for mystery packs or streaming.

https://www.instagram.com/reel/C-biS...ljNWg3cGpmcg==

That can’t be it. There must be a more convoluted explanation rather than an actual customer.


Sent from my iPhone using Tapatalk