Solicited eBay Message 1 1

[thehumantorch]

19 hours ago, blazingbob said:

Are you going to sign up for Dylan's grading course?  He is an overstreet advisor.

I'd sign up for that course.  Hope there's a question and answer segment.

[RhialtoTheMarvellous]

I didn't realize this guy was notorious. I bought a small lot of comics from him and they were extremely poor overall with a lot of concealed problems on the backs. I viewed it as my fault (and still do), because there were no pics of the backs of the books on the listing and he made no representation as to condition. I've taken a chance on one or two other auctions like that and came up with some gems, but in this case pretty much every book was in poor condition. I should have known better considering it was a guy that had thousands of sales. The deals I had found were people doing more one off sales of comics on an account where they sold different items.

[blazingbob]

14 minutes ago, RhialtoTheMarvellous said:

I didn't realize this guy was notorious. I bought a small lot of comics from him and they were extremely poor overall with a lot of concealed problems on the backs. I viewed it as my fault (and still do), because there were no pics of the backs of the books on the listing and he made no representation as to condition. I've taken a chance on one or two other auctions like that and came up with some gems, but in this case pretty much every book was in poor condition. I should have known better considering it was a guy that had thousands of sales. The deals I had found were people doing more one off sales of comics on an account where they sold different items.

Imagine a world where there was the ability to "Rate" the overstreet advisor's.  You want the notoriety that comes with being one,  well guess what Diamond will now take feedback on how well you are performing as one that is public record.  Naturally schmear campaigns will be removed but how about some honest feedback on who is deserving and who is not.  Why are the CGC boards the "Feedback forums" of public sellers/auction houses and the place where you have to go to find out something?  

Edited September 28, 2020 by blazingbob

[RhialtoTheMarvellous]

 

1 hour ago, blazingbob said:

Imagine a world where there was the ability to "Rate" the overstreet advisor's.  You want the notoriety that comes with being one,  well guess what Diamond will now take feedback on how well you are performing as one that is public record.  Naturally schmear campaigns will be removed but how about some honest feedback on who is deserving and who is not.  Why are the CGC boards the "Feedback forums" of public sellers/auction houses and the place where you have to go to find out something?  

You've taken us to the original eBay feedback problem. You're placing an aggregate of subjective opinions against one subjective opinion.

The problem now becomes, what percentage of the negative feedback is just ignorant people or people that have a grudge against the person doing the rating versus actual real complaints about his rating skill?

"9.8 rating, I love this guy, he is a great rater!"

"8.5 rating on my previous 9.8, this guy is an insufficiently_thoughtful_person!"

There is no added value to having something like this.

[blazingbob]

1 hour ago, RhialtoTheMarvellous said:

 

You've taken us to the original eBay feedback problem. You're placing an aggregate of subjective opinions against one subjective opinion.

The problem now becomes, what percentage of the negative feedback is just ignorant people or people that have a grudge against the person doing the rating versus actual real complaints about his rating skill?

"9.8 rating, I love this guy, he is a great rater!"

"8.5 rating on my previous 9.8, this guy is an insufficiently_thoughtful_person!"

There is no added value to having something like this.

Well then be marvelous and come up with something that works.

Is there a Hall of Shame for Overstreet advisors?

Overstreet Advisor Dispute system in place if a transaction doesn't go as planned moderated by Diamond?

Lawsuits filed listed under their names?

Bounced check count?

Been in business since?

For some I'd love a last seen date

 

 

 

Edited September 28, 2020 by blazingbob

[RhialtoTheMarvellous]

1 hour ago, blazingbob said:

Well then be marvelous and come up with something that works.

Is there a Hall of Shame for Overstreet advisors?

Overstreet Advisor Dispute system in place if a transaction doesn't go as planned moderated by Diamond?

Lawsuits filed listed under their names?

Bounced check count?

Been in business since?

For some I'd love a last seen date

 

 

 

I think your earlier suggestions are more pertinent than some sort of rating system. If Overstreet gives out certifications or represents that someone is legit then they should own that and have a database that you can search on their site that shows they approve of them and that way they can also remove them if they get complaints.

For the buyer it's always buyer beware overall. Do your due diligence as to what you are getting and take nothing for granted. Make sure that you have options to get out of the deal if you don't get what you expected. If those things don't fall into line then don't do the deal.

 

 

[RhialtoTheMarvellous]

Also, whenever I'm reading this guy's name I'm thinking Dielan.

Although, this is probably more appropriate.
 

 

[Park]

On 9/27/2020 at 9:33 AM, joeypost said:

i hope you are correct about the longevity of his career, but I feel he will stick around a lot longer than anticipated. 

Yeah, Dupcak seems to be plugging right along.

[seanfingh]

1 minute ago, Park said:

Yeah, Dupcak seems to be plugging right along.

Quit buying from him, bro.

[Park]

Just now, seanfingh said:

2 minutes ago, Park said:

Yeah, Dupcak seems to be plugging right along.

Quit buying from him, bro.

But why?

[evilskip]

Said it before and will say it again. Just because somebody is an Overstreet Advisor, it doesn't have anything to do with grading correctly. OPG is looking for Market Reports. There are a lot of Overstreet advisors out there that IMO can't grade for.(Or they've not looked at the grading standards in the front of the book in a while).

[Domo Arigato]

An Overstreet Advisor certificate is not as easy to acquire as some of you think.

Dylan had to eat 27 boxes of Cracker Jacks before he finally got his.

 

[SeniorSurfer]

On 9/27/2020 at 3:11 PM, blazingbob said:

Are you going to sign up for Dylan's grading course?  He is an overstreet advisor.

I could see an Overstreet grading course that would have maybe five respected advisors sitting around a table (or via Zoom in these days) where each takes a book with obvious differing grades, points out their defects, and establishes a (near) consensus grade to educate all watching.  Have it recorded, then posted on YouTube.  I think it would garner much interest, views and be very informative.

By the same token in another world, my Loony Tunes mentality has one of those five being this individual.  I then picture each book having their flaws pointed out (as before) with all the advisors hiding their grades until the big reveal, at which point you will have 4 sane and somewhat similar grades while one will be akin to an Olympic Russian judge's left-field, seeing-eye grade:  "So Ed... What did you determine?" "I say 4.0"  "John?"  "3.5 for me."  "Bob?"  "Definitely no more than 4.5." "Mike?"  "Looks about a 4.0."  "Uh... D_____?"  "That's a solid 7.5 with room to go 8.0 with a press."   (awkward silence with mouths slightly ajar). 

[RhialtoTheMarvellous]

If I were to establish a systematic grading process here is what I would do.

1. Assume that we use the same 10 point scale that is currently used (though IMO it's somewhat absurd and overly complex).
2. Create a master document identifying each possible defect a comic can have and how it affects the grade of the book.
3. Have five different people take the master document and the same exact sampling of books and give them all grades with notes on why they deducted points. This is done individually and in private.
4. Bring all of those people together and see where the differences occur.
   1. Are there things that can be interpreted differently in the master document?
   2. Are there errors in the master document?
   3. Are there things that are just wrong in the document?
   4. Are there things missing in the document?

Repeat the process until all you have is human error on the part of the graders, which you can't eliminate, but you mitigate by having multiple people grade a book.

Now that I think about it, this process might be a good match for a machine learning AI. You collect high res images of different books and their resulting grades and throw those into a model and you could probably get some pretty accurate scores. It would be impractical though as most imagery of comics out there varies pretty widely in terms of quality.

[DanCooper]

2 hours ago, SeniorSurfer said:

"That's a solid 7.5 with room to go 8.0 with a press."   (awkward silence with mouths slightly ajar)

Sadly, in the current comic atmosphere, there are probably more people that would sign up for a course on "How to spot defects on raw books that could be "improved upon" and get you to that higher blue number" than there would be people interested in learning how to grade properly by standards.

Notice I say "people" and not collectors.

 

[theCapraAegagrus]

This thread:

[justafan]

3 hours ago, RhialtoTheMarvellous said:

If I were to establish a systematic grading process here is what I would do.

1. Assume that we use the same 10 point scale that is currently used (though IMO it's somewhat absurd and overly complex).
2. Create a master document identifying each possible defect a comic can have and how it affects the grade of the book.
3. Have five different people take the master document and the same exact sampling of books and give them all grades with notes on why they deducted points. This is done individually and in private.
4. Bring all of those people together and see where the differences occur.
   1. Are there things that can be interpreted differently in the master document?
   2. Are there errors in the master document?
   3. Are there things that are just wrong in the document?
   4. Are there things missing in the document?

Repeat the process until all you have is human error on the part of the graders, which you can't eliminate, but you mitigate by having multiple people grade a book.

Now that I think about it, this process might be a good match for a machine learning AI. You collect high res images of different books and their resulting grades and throw those into a model and you could probably get some pretty accurate scores. It would be impractical though as most imagery of comics out there varies pretty widely in terms of quality.

I was thinking of a good way to tackle this problem and about to reply with my own  but what you've laid out is a pretty good way to start.  However, here are the key areas of conflicts with any grading standard that seems to get in the way a lot and that may need to be defined/addressed early on in the development of that "master defect document".

1. Undetectable pressable/cleanable vs non-pressable/non-cleanable defects.  one of the interesting things in the Overstreet grading guide is that there are multiple representations of defects that are pressable/cleanable but only allowed in lower grades but not upper grades. one example might be large or book length bends (not creases) that can be completely pressed out, multiple non-color break shallow bends and finger bends that don't break color all over the book but with sharp corners, perfect spine, and edges, or mild to moderate soiling all over the front or back covers that is dry cleanable.  Should those continue to be graded at the same level as they are now or considered for elevation in grade considering that FN/VF or VF book could soon be a VF/NM or NM+ with a good thorough clean and press?

2. Tear vs Stains: this debate may be as old as grading itself. Overstreet is pretty clear on where these fall but CGC and other grading companies treat these very differently.  From collector to collector, everyone has a difference on if given a choice between a comic with a small stain to a small tear which they would prefer and the degrees of acceptability at different grades.  Then there's the issue of stain type (if you can even determine) to classify whether they are cleanable or pose a potential for increased deterioration of the paper.  Same for tear type. Some small tears 1/4"-1/8" are mangled affecting the appearance whereas I've seen and even missed 1" tears that were so perfectly fine that they are invisible to the naked eye in hand unless you open the cover and see the edges separate.  Some small tears are in danger of becoming chipping or pieces missing where as other cut through covers and story.

3. Defect naming conventions: Overstreet provides some definitions and labels but it is far from complete.  To define the defects you have to be explicit to cover all possible common defects.  See #4 below:

4. Spine tick/bump/bend/crease/stress: Whether they break color or not, these are often used interchangeably and incorrectly.  Furthermore, if a book has spine wear that comes in the form of 1/2" bends but only part of each bend (say 1/4") is a sharp crease and only part of that crease (1/8") show color break, what do you call them?  1/2" spine bends with 1/8" color break?  1/2" spine bends with 1/4" crease with 1/8" color break?  After a press you may end up with only the 1/8" color break spine ticks and maybe some light spine stress lines.

5. Quantifying accumulation and eye appeal: Of all the issues with grading this one is the hardest and most subjective. If you look in the grading guide, you'll see interesting examples of NM books with specific defects but you KNOW that there are more defects on those covers that you cannot see than what the photos are showing you even if you are able to see the ones they are pointing at with arrows.  One example is spine ticks, you may only see the color break but what about the different sizes of their associated non-color break crease or bend lengths leading away from them.  And you know most of those covers have lots of shallow bends that are only visible if you angle the light a certain way.  If they were to represent that in the grading guide (haven't seen a visual example in the guide) what sizing and quantity, or percentage of coverage of the cover is acceptable in various grades. 

Overstreet also shows a graph with a guide on acceptable quantity of defects in the different grades.  What it doesn't show is the severity of each defect in each grade.  It does state that a book with multiple smaller defects may grade higher than a book with a single severe defect. But where do you draw the line?  Would a single diagonal non-color break cover bend up to 4" be graded higher than 4 different 1" bends on the front cover

6. Front cover defects vs Back cover defects: CGC certainly has, in the past and probably currently, given more weight to defects on the front cover than on the back.  this is where the whole spine re-alignment issue surfaced but there are normal non-manipulated examples.

7. Grader state of mind:  I know I'm totally guilty of this and have observed this when I've gone back and examined some books weeks or even days later.  this may be the cause of the greatest swings in subjectivity when grading. Depending on my mood, comfort, or even the book itself, I have seen areas where I've totally over or undergraded a book and gone back and asked myself what was I on when I considered that defect acceptable in that grade or why I was so harsh on the book for something that even overstreet considers NM.  What must we do to either clear our mind, use as a grading sheet, use the same type of lighting, or meditate to help us focus to ensure that the way we grade and assign the grade remains consistent?

One last thing: Overstreet appears to be a very strict standard above what even CGC or other grading company's allow.  One competitor in Australia appears to be the only one that claims to strictly follow Overstreet's standards.  However, over time we have even seen Overstreet bend towards CGC's standards in certain areas.  As the goal post moves from industry leaders, should we follow or try to get them to follow what makes sense to us?

Solve those 7 and probably others I've missed and you can probably cut that Master Defect Document into stone for all time. That's all I got for now but I'd be interested to hear what you all think about those issues and how you grade around them.

 

[blazingbob]

First off Grading is what is in front of you now,  not what the book can become.  That is referred to as a proscreen by a person who does pressing.

I don't care what can come out,  that "What it can be" is only confirmed when it is done by the presser and in a holder or when the book is done and is regraded raw.  

Anybody who is selling books on a "Wanna/outta-be" grade needs to question their selling motive.  If you want to sell "Wanna/outta be" grades then spend the money and get it pressed.  

1. Undetectable pressable/cleanable vs non-pressable/non-cleanable defects.  one of the interesting things in the Overstreet grading guide is that there are multiple representations of defects that are pressable/cleanable but only allowed in lower grades but not upper grades. one example might be large or book length bends (not creases) that can be completely pressed out, multiple non-color break shallow bends and finger bends that don't break color all over the book but with sharp corners, perfect spine, and edges, or mild to moderate soiling all over the front or back covers that is dry cleanable.  Should those continue to be graded at the same level as they are now or considered for elevation in grade considering that FN/VF or VF book could soon be a VF/NM or NM+ with a good thorough clean and press?

Edited September 29, 2020 by blazingbob

[RhialtoTheMarvellous]

I'm thinking that I could create grading software as a machine learning project. I will need a lot of front and back images of comics that have been CGC graded (need the actual grade number as well) to train the model though.

[bc]

6 minutes ago, RhialtoTheMarvellous said:

I'm thinking that I could create grading software as a machine learning project. I will need a lot of front and back images of comics that have been CGC graded (need the actual grade number as well) to train the model though.

While that sounds great - how about the interior defects (like pages & stamps missing, centerfold & wraps detached, tears, etc.)?

-bc