Some Data Regarding TPG Grading And Cracking/Resubmitting

To participate in the forum you must log in or register.

Author	Replies: 26 / Views: 3,880	Page: of 2
ProfLiz Valued Member United States 373 Posts	Posted 09/18/2013 2:22 pm It seems like every few months, the topic of TPG consistency comes up, usually in the context of submitting or resubmitting coins for grading. In this post, I want to add some solid data to those discussions. Over the past year or so, I've been slowly phasing out my old Dansco 7070 type set in favor of a slabbed type set. In the process, I've submitted nearly 50 coins to NGC and PCGS. Some of these coins, I purchased raw; others were cracked out of slabs to go in my album. (I also submitted a few miscellaneous coins, not from my 7070 set.) I include the complete data on original and new grades in the table below. Feel free to analyze it and draw your own conclusions. A few things to keep in mind: I do not specify the coins graded in the table, but they include a wide variety of coins, all of which are pre-1930 classic types (no moderns) in circulation strikes, including gold. No denomination or type is disproportionately represented. Remember, this is basically a 7070 type set. I was not particularly good at grading coins when I assembled this set. I bought most of them on ebay or from online coin dealers; I certainly did not have any expert assistance; and I was quite price-driven. That means I expect that these coins are representative of available coins in their grade, not necessarily high-end or select. All coins were submitted raw. I did not try any crossovers of slabbed coins. Some of the conclusions that I find particularly interesting are: Of the 15 raw coins I sent in, 3 received details grades. The others graded from VG08 to MS62. Of course, I only sent in coins I thought had a reasonable chance to grade. However, this does suggest that there are gradable raw coins available in a wide spectrum of conditions on ebay and elsewhere. Of the 32 slabbed coins that I cracked, 3 received details grades. Only one of those started in a PCGS or NGC slab, but it was a doozy, going from NGC MS64 to PCGS genuine. Of the 29 non-details coins, about 40% regraded at the same grade (12 coins). Of the remaining 60%, about half upgraded (8 coins) and half downgraded (9 coins). This is interesting for two reasons: ---First, since over half of the coins changed grades (and did so symmetrically about the original grade), it shows that a grade might be better seen to signify a "grade range." ---Second, there is an assumption that the coins on the open market represent the low end for their grade. The idea is that higher end coins are either impounded in collections or were resubmitted until they upgraded. My data do not show this. If "average" coins were actually low end for their grade, downgrades should be more likely than upgrades, and that is not the case. The symmetry between upgrades and downgrades is also seen when comparing the magnitude of the grade change. The largest upgrades were three grade levels (AU58 to MS62, XF45 to AU55), and the largest downgrades were exactly the same in reverse (MS62 to AU55, AU55 to XF45). There were also three 2-grade upgrades, and three 2-grade downgrades. Again, this suggests that there is no bias toward weak-for-the-grade coins. Although I do not have sufficient statistics to compare grades within or across PCGS and NGC, the overall trends seem to be repeated within the grading services. For example, on 9 PCGS to PCGS regrades, 3 upgraded, 3 downgraded, and 3 remained at the same grade â€" very similar to the statistics as a whole. I do not have sufficient data to make conclusions about other TPGs. However, of 2 ICG slabs, one regraded details and the other downgraded. Likewise, of 2 new ANACS slabs, one got details and the other downgraded. The old, small ANACS slabs have a reputation for being undergraded. In the 3 represented here, they performed pretty much like PCGS or NGC: one upgraded, one downgraded, and one stayed the same. I hope this provides some food for thought! raw ---> NGC AU53 raw ---> NGC XF details raw ---> NGC VG08BN raw ---> NGC AU58 raw ---> NGC MS61 raw ---> NGC MS60 raw ---> NGC AU55 raw ---> NGC AU55 raw ---> NGC MS62 raw ---> PCGS AU details raw ---> PCGS AU details raw ---> PCGS VF20 raw ---> PCGS XF40 raw ---> PCGS AU50 raw ---> PCGS VF30 ANACS EF40 ---> PCGS XF details ANACS AU58 ---> PCGS AU55 ICG VF20 ---> NGC VF details ICG MS64BN ---> NGC MS63BN NGC AU58 ---> NGC MS62BN NGC AU58 ---> NGC AU58 NGC MS60 ---> NGC AU55 NGC MS61 ---> NGC MS62 NGC AU55 ---> NGC AU55 NGC MS62 ---> PCGS MS62 NGC AU55 ---> PCGS XF45 NGC AU58 ---> PCGS AU53 NGC AU55 ---> PCGS AU55 NGC AU55 ---> PCGS AU55 NGC MS64 ---> PCGS genuine NGC MS65 ---> PCGS MS65 old ANACS XF45 ---> NGC AU55 old ANACS AU58 ---> NGC AU58 old ANACS AU58 ---> PCGS AU53 PCGS AU50 ---> NGC AU53 PCGS AU50 ---> NGC AU55 PCGS MS65 ---> NGC MS65 PCGS MS65 ---> NGC MS65 PCGS AU55 ---> PCGS AU55 PCGS MS62 ---> PCGS AU58 PCGS AU50 ---> PCGS AU55 PCGS AU55 ---> PCGS AU53 PCGS AU55 ---> PCGS AU58 PCGS MS65 ---> PCGS MS64 PCGS MS64 ---> PCGS MS65+ PCGS MS64 ---> PCGS MS64 PCGS MS64 ---> PCGS MS64 Report this Post to the Staff

jdmern Pillar of the Community United States 1949 Posts	Posted 09/18/2013 2:54 pm That's some really cool info, I've broken out a few coins and resubmitted and found much the same thing Boardwalk Numismatics http://stores.ebay.com/jdmern/ https://boardwalknumismatics.com Report this Post to the Staff
pawpaw34 Valued Member United States 331 Posts	Posted 09/18/2013 2:56 pm Thanks for the info. That was an interesting read. Report this Post to the Staff
Fatman Valued Member United States 362 Posts	Posted 09/18/2013 3:20 pm NGC-64 to PCGS Genuine? That one sounds mighty hokey if you ask me. Just another reason to forgo TPG's imho. But I think I'm just Report this Post to the Staff
bpoc1 Rest in Peace United States 4078 Posts	Posted 09/18/2013 4:42 pm ProfLiz, after following some of the contest sent in here at CCF for grading, ie Large Cents. You have a better idea on grading then most. This information you provided is very interesting. Which needs to be read a couple of times before any conclusions. ( By me ) One question at the start, is there a time line from when you first started submitting? Quote: Over the past year or so, oops! Edited by bpoc1 09/18/2013 5:07 pm Report this Post to the Staff
ProfLiz Valued Member United States 373 Posts	Posted 09/18/2013 5:42 pm bpoc1 - You are very kind! Of course, I tried to pick attractive examples, but the fact is that when I originally purchased most of these coins, I was quite a novice. Plus, they represent many series, not just old coppers (where I've started to learn a bit about grading). Fatman - Yeah, 64 to details really smarts. I can see the few, very light hairlines that earned it the "cleaned" designation. But they are, to me, quite clearly contact hairlines NOT cleaning. I might've downgraded, but not details graded. C'est la vie. Report this Post to the Staff
ProfLiz Valued Member United States 373 Posts	Posted 09/18/2013 5:46 pm P.S. I belatedly noticed that the original post above was my 200th CCF post. Seems appropriate, somehow! Report this Post to the Staff
FadeToBlack Pillar of the Community 1751 Posts	Posted 09/18/2013 5:46 pm 7 upgrades, eh? What was the 64 to 65+? That's a big jump. Report this Post to the Staff
ProfLiz Valued Member United States 373 Posts	Posted 09/18/2013 7:47 pm FadeToBlack - The 65+ is an Oregon commemorative half dollar. A very handsome coin! Of course, it was just as lovely when it was "merely" a 64. ;-) Report this Post to the Staff
FadeToBlack Pillar of the Community 1751 Posts	Posted 09/18/2013 9:35 pm Nice jump. I'm hoping I can find a nice, attractive Oregon in 65 at the Philly show later this week. Not sure if I'm going tomorrow or Friday. Report this Post to the Staff
chetzler Valued Member United States 206 Posts	Posted 09/23/2013 01:34 am Very interesting research! I think it could be even more telling to increase the sample size. I wonder how many CCF members could contribute their own regrading experiences to the data. It would be fun to start a database that all members could add to. Heck, I'd even volunteer--I enjoying designing relational databases. Edited by chetzler 09/23/2013 01:37 am Report this Post to the Staff
Susuman Pillar of the Community United States 595 Posts	Posted 09/23/2013 03:16 am ProfLiz, A couple of questions. 1. Is there any bias, scatter or better consistency based on type of coin (copper, silver or Gold)? 2. Is there any consistency in the re-grades with respect to value? The scatter in your data would indicate that statistically we should be able to do well by buying MS-64 1928-S Peace dollars ($1400 on PCGS Coinfacts), cracking them out, and re-submitting them. From your data, we should have about a 1/3 chance of upgrading a single tier to MS-65 ($20,000 on PCGS Coinfacts). However, the relative populations of the higher end grades for this coin (1,928 @ MS-63, 1,759 @ MS-64, and 48 @ MS-65)would seem to indicate that this is unlikely to happen. So, in this case, either the grading companies 1 )establish artificial glass ceilings for some coins and grades, 2) take extra care with such grades to be more consistent, or 3) correctly recognize that there is an exceptionally clear break in the actual coin populations where there is little movement between the grades from cracked-out re-submissions and the grading. Unfortunately, I have not yet broken out any coins and re-submitted them, though I have a few I think are undergraded that I eventually will. I can also say that I have one small experiment that actually attests to consistency. I had a small envelope with three uncirculated 1856 slanted 5 large cents. The cents were essentially indistinguishable and red except one had just a little bit more spotting. I submitted one to PCGS and it came back MS-64RD. A few months later, I submitted the other 2 to PCGS and they came back MS-64RD and MS-64RB. The RB designated coin was for the one with a little spotting. So, the grades on these three basically seem consistent. In essence, I was pleased that all three coins did come back with the same numerical grade with 2 submissions and the color designations were consistent. Noting that the RD designated coins are condition census, with only 1 higher (MS-65) in population, perhaps the companies are more consistent with higher end coins. On the other hand, there is only a population of 3 lower (MS-63), so perhaps MS-64 is a default grade on these in red color unless there is clear reason to move them higher or lower. Unfortunately, I am not dealing with a statistically valid set to really say much with confidence. I will post pictures of these cents in the Days of.... thread when the time comes. It would be very interesting to develop better statistics on the consistency of regrades. I would be happy to assist in the compilation of data if other members want to contribute their experiences. Report this Post to the Staff
basebal21 Bedrock of the Community 13014 Posts	Posted 09/23/2013 03:51 am Quote: However, the relative populations of the higher end grades for this coin (1,928 @ MS-63, 1,759 @ MS-64, and 48 @ MS-65)would seem to indicate that this is unlikely to happen. That MS 64 population is likely artifically high from people trying to crack them and get a 65 given the huge increase in vale just like the 63 population is probably pumped up a bit too for the same reason. The top populations on PCGS for coins like that generally dont see a lot of movement. Borderline coins will get the lower grade when it comes to situations like that, but the pop will climb from time to time if you look at it over long periods so they havent just set a number and said no more. With that said though the higher the potential value the tighter/pickier they seem to be with the grades. I would say for a true test you would need to send a bunch of coins in crack them and resubmit them right away assume no damage was done in the crack out. No offense to Profliz at all and its always interesting seeing how coins can change, but once theyve sat in an album for a while they can change for better and for worse. Given that eye appeal is a significant part of the grade I may be in the minority but I generally dont expect coins to come back the exact same after sitting in an album a significant period of time. For what its worth I've tried it twice with immediate resubmits (once for a coin in my collection and once for a friend) both times they came back the exact same. Report this Post to the Staff
Susuman Pillar of the Community United States 595 Posts	Posted 09/23/2013 05:13 am Quote: That MS 64 population is likely artificially high from people trying to crack them and get a 65 given the huge increase in vale just like the 63 population is probably pumped up a bit too for the same reason. I suspect that you may be correct in this. Unfortunately, because of crack-outs and resubmissions, it is not possible to distinguish this with respect to my original question without having genuine knowledge of the real numbers of coins submitted, regraded, and how many times they have been regraded. Unfortunately, this statistical data set would be nearly impossible to collect unless you have a lot of detailed contributions from submitters or you are wealthy enough to develop your own data set (or you are PCGS and take and compare photos of every coin) Quote: ..... but the pop will climb from time to time if you look at it over long periods so they havent just set a number and said no more. Of course, didn't mean to imply otherwise Report this Post to the Staff
Broseph Pillar of the Community United States 979 Posts	Posted 09/23/2013 07:18 am This is like a scientific case study of coin grading. Very interesting! Report this Post to the Staff
CoinsKelly Pillar of the Community United States 3453 Posts	Posted 09/23/2013 07:52 am Since these were coins that you purchased a while ago I wonder if the grading standards shifted. It would make sense that a coin sent in/cracked out today would receive the same grade next week but coins cracked out a decade or so ago might have been subjected to slightly different grading standards. ProfLiz, how long between your purchase/crack out and recent submission? I hope you start a ProfLiz 7070's thread. Report this Post to the Staff
	Replies: 26 / Views: 3,880	Page: of 2

Last 15 Replies

To participate in the forum you must log in or register.

All Forums Category: US Coins and Currency Forums Forum: US Classic and Colonial Coins

Silver Gold Bull US is the United States leading bullion dealer. We deliver gold coins, gold bars, silver coins, silver bars and precious metals to your door. Order today!

The First Dollar, your premier source for bullion and rare coins. See our bullion list on the site and the coin inventory on ebay.

View Last 100 New Topics

Disclaimer: While a tremendous amount of effort goes into ensuring the accuracy of the information contained in this site, Coin Community assumes no liability for errors. Copyright 2005 - 2026 Coin Community Family- all rights reserved worldwide. Use of any images or content on this website without prior written permission of Coin Community or the original lender is strictly prohibited.
Contact Us | Advertise Here | Privacy Policy / Terms of Use

Coin Community Forum

It took 0.46 seconds to rattle this change.