ANACONDA challenges the grading services
ANACONDA
Posts: 4,692 ✭
When it comes to grading, I have seen it claimed, on several occasions, that "PCGS is the most consistent". Really? It wouldn't be that hard to determine on an objective basis, if that were true.
I hereby formally challenge any and all of the following -- PCGS, NGC, ANACS, PCI, SEGS, ICG -- to perform the following test (through a neutral, disinterested third party):
Take ten coins, ship them to PCGS for grading. Crack them out, resubmit them. Do that about 10 times.
Take the same ten coins crack them out of the PCGS holders, ship them to NGC for grading. Crack them out, resubmit them. Do that about 10 times.
Compare the results. I'll even supply the coins for free. I want them back, though.
Finally, is consistency what we really want? (That actually ought to be the first question.) ACG might be consistent. I don't really know, but I'm almost ashamed that my name begins with an "A" just like ACG - I wouldn't put my dog into ACG coins, but maybe they're consistent.
I think what we really want is accuracy. Accuracy emobdies consistency. It's a moving target to some degree, but not entirely, and one could write a couple of hundred pages on that one. (I could get started with a couple of hundred pages. Beecha to it.) And one more thing. This concept of technical grading is unwise. "Yeah, that coin looks like crap, but it sho is clean. Must be worth a forturne."
"So, adrian, do you think surface abrasions count for nothing?" Uh....no. But the value of a coin should be MORE related to it's grade. I recently saw an early copper in a PCGS 66 Red holder with significant planchet flaw on the reverse. Maybe you like planchet flaws. Kinda like a bald girl if you ask me. Interesting but not my kinda thang. The coin without the planchet flaws was probably a 66 but I felt sorry for the fellow who was in it at a no-planchet flaw price. "Hey, whatta you talkin' 'bout! IT'S GRADED BY PCGS! YOU'RE TRYIN' TO RIP ME OFF!"
Oh, well. Just a few thoughts. Lotsa holes, but still, a few thoughts. What do you think - any grading service want to take up that challenge? It can be done to obtain objective results. Has this already been done? You gotta be kidding me?
adrian
anaconda.rare.coins
I hereby formally challenge any and all of the following -- PCGS, NGC, ANACS, PCI, SEGS, ICG -- to perform the following test (through a neutral, disinterested third party):
Take ten coins, ship them to PCGS for grading. Crack them out, resubmit them. Do that about 10 times.
Take the same ten coins crack them out of the PCGS holders, ship them to NGC for grading. Crack them out, resubmit them. Do that about 10 times.
Compare the results. I'll even supply the coins for free. I want them back, though.
Finally, is consistency what we really want? (That actually ought to be the first question.) ACG might be consistent. I don't really know, but I'm almost ashamed that my name begins with an "A" just like ACG - I wouldn't put my dog into ACG coins, but maybe they're consistent.
I think what we really want is accuracy. Accuracy emobdies consistency. It's a moving target to some degree, but not entirely, and one could write a couple of hundred pages on that one. (I could get started with a couple of hundred pages. Beecha to it.) And one more thing. This concept of technical grading is unwise. "Yeah, that coin looks like crap, but it sho is clean. Must be worth a forturne."
"So, adrian, do you think surface abrasions count for nothing?" Uh....no. But the value of a coin should be MORE related to it's grade. I recently saw an early copper in a PCGS 66 Red holder with significant planchet flaw on the reverse. Maybe you like planchet flaws. Kinda like a bald girl if you ask me. Interesting but not my kinda thang. The coin without the planchet flaws was probably a 66 but I felt sorry for the fellow who was in it at a no-planchet flaw price. "Hey, whatta you talkin' 'bout! IT'S GRADED BY PCGS! YOU'RE TRYIN' TO RIP ME OFF!"
Oh, well. Just a few thoughts. Lotsa holes, but still, a few thoughts. What do you think - any grading service want to take up that challenge? It can be done to obtain objective results. Has this already been done? You gotta be kidding me?
adrian
anaconda.rare.coins
0
Comments
Dragon
You speak the truth, and you won't get any takers on your challenge.
When you buy plastic, you're buying an opinion, that's it.......and opinions change.
Dave
adrian
Dragon
adrian
One thing we have to consider here is turn around time. We are going to send the same ten coins to the PCGS to be graded and returned, crack them out and resend them on ten different occasions. I think that we'll both be sharing coin stories at the rest home by the time this is accomplished.
Also, I wouldn't use the 1834 NGC PR67 25c piece, because after the 7th or 8th time it shows up at the service, someone might start to get suspicious. "Gee, we sure are gittin' lots o' these lately!"
Mike
My Barbers
You wouldn't have a five year LAY-A-WAY plan would you?
Mike
Someone once posted here that this coin sold at auction for something like $35000 days before you started listing it on eBay. I don't know if this is true. Just curious if it is.
I also wonder why it hasn't sold at $150,000 the few dozen times it has been listed, but the asking price is now up to $175,000. Same with that MS68 Eagle.
Take the same ten coins crack them out of the PCGS holders, ship them to NGC for grading. Crack them out, resubmit them. Do that about 10 times.
This experiment would be unfairly biased in NGC's favor.
You'd need to alternate which service you submitted to, so as the coins acquired haze, prints, and nicks from their multiple dangerous journeys, the grades would tend down at both services.
Still want to supply those coins Anaconda?
What's with talking in the third person? Are you confused?
if you have a rifle that consistenly shoots the bullet 1 foot too high, you could have just as good luck hunting w/ it as 1 that is exactly on target. you would just learn to aim 1 foot too low! the gun would be "inaccurate", yet still get the job done by you knowing its nuances and compensating appropirately.
but if the rifle were inconsistent and sprayed those bullets all over the place, you'd be the laughing stock of all your fellow deer hunters (elephant hunters, groundhog hunters, whatever). not to mention a danger to farmer bill's cows.
bottom line: if there are 20 people grading a coin, there will be exactly 20 "accurate" opinions, and each 1 must be respected, because again, a grade IS AN OPINOIN, not an "accurate consistent" FACT. paying pcgs for their opinion makes it 21. its YOUR opinion that should matter most to YOU, since it's YOUR MONEY.
K S
Collecting Dollars
the same type and date, but one with a good strike, one with marks on only
the reverse, one with light circuation, etc. It's not especially valuable to grade
only some sorts of coins accurately/consistently.
a pitcher can be accurate, but if he is also consistent, he will get rocked every time. why? imagine if he consitently put that slider right down the middle, and accurately hit the strike zone. that would be a BAD thing. "accurate" in the baseball sense literally means "INconsistent", because the pitching coach wants him to be inconsistent w/ the ball placement. (hope your not a cubs fan...)
"accuracy" is not definable in numismatic terms across all "graders", because the definition for au-55 likely will be different for you, me and for pcgs (imagine that). sure, there is the ANA official guide, but the simple fact is that it is a guide and not a definitive. and if "accuracy" can't be definitive, using your logic, "consistency" can't be definitive either.
i believe that some percent, say 70%, of observers might accept a au-55 grade for a specific coin, but we can't immediately conclude that the other 30% of observers who diagree are "wrong" or "incosistent" or "inaccurate".
bottom line: i think the biggest problem w/ grading, and it is heavily exacrebated (sp?) by plastic svc's, is that everyone seems to believe that having "different" grading opinoins is somehow a bad thing. RIDICULOUS! why is it bad for you and me and pcgs to differ in opinion on a coin's grade? if everyone in the hobby walked around like robots, agreeing w/ each other on everything , how much fun would coins really be?
it always comes down to: if you learn to grade for yourself and buy coins by how YOU grade, and stay honest w/ yourself, you'll always be happy w/ what you buy.
K S
Accuracy is hitting the right target
Consistency is aiming for the same target each time
Precision is hitting the right place on the right target
Grading is inherently imprecise and a challenge to be accurate. However, it can always be consistent. So long as the rules stay the same.
Neil
Varying the placement on a pitch isn't being inconsistent. It is displaying variety. If they all fall in the strike zone, then the pitcher consistently throws strikes, and I venture to say, THAT IS WHAT IS DESIRED.
Sounds like several of you are trying to apply a generalization to a specific. By what sometimes appears to be a majority of you, PCGS seems to be consistently disappointing!
I'll be satisfied with consistently accurate; consistently timely; consistently fair, but not just consistent.
Keep us all laughing! This place wouldn't be the same without you!
adrian
anaconda.rare.coins (on eBay)
<< <i>Hey gmarguli, NastyNic, cointagious - you guys are great! I love to read your posts. Kinda like watching a fat lady run. Not very educational or thought provoking but always entertaining!
Keep us all laughing! This place wouldn't be the same without you!
adrian
anaconda.rare.coins (on eBay) >>
Heh Heh!.....Ha Ha Ha! LOL! Its the truth!
TBT
TBT
<< <i>How in the sam hill can agreeing on the state or preservation of a coin constitute being a robot? >>
NOT what i said. what i said was that if everyone walked around agreeing on the same grades on the same coins at the same time for the same reasons THEN we would just be a bunch of robots.
<< <i>Varying the placement on a pitch . . . .IS WHAT IS DESIRED. >>
exactly so - the variation is a GOOD thing, just as is variation in how collectors will vary in how they grade coins.
<< <i>I'll be satisfied with consistently accurate; consistently timely; consistently fair, but not just consistent. >>
now that i CAN we agree with, because you used a key word here: FAIR. and by definition, FAIR is a subjective thing. if you feel like you've been fairly treated 100% of the time by a grading service, you should use that svc. all the time, because the consitency and accuracy ARE 100% IRRELEVANT. you feel like your getting what you pay for, a fair opinion, and that is what it's all about.
K S
K S
it's too bad money is in the mix to cause problems. that always seems to be the bottom line whether it's acknowledged or not. if a grade point results in dollar loss the service sucks. if an overgrade results nothing is ever said.
from my line of thinking, if i need a service to assign a grade to a coin for whatever reason, it means i am unsure of the grade and am asking for help. some of you guys seem to know what the coins you're sending in are, grade wise, and then are upset when XYZ Service doesn't agree and you end up on this sounding board complaining. if you send a coin in, accept the opinion or quit sending. if you already know the grade, save the submission fee.
al h.
1) I believe that most people want/need accuracy and consistency. PCGS and NGC could probably be consistent but inaccurate (consistently inaccurate) by grading too liberally or too conservatively. I don't think we want that. And, while accuracy implies consistency, that accuracy needs to be consistent over time (consistent accuracy). Semantics, I know but.....
2) Many dealers (and yes, some collectors, too) don't want accuracy or consistency. I know plenty of sharp buyers who buy conservatively graded coins and upgrade them to accurately graded coins or who buy accurately graded coins and upgrade them to liberally graded coins. Most buyers want to buy conservatively graded coins and most sellers want to sell liberally graded coins.
3) contagious, you state : the only sense that can be made of adrian's proposal is that he has some high end non-pcgs coins for sale. other than that, it's meaningless excercise. if anyone hasn't figured it out yet, ananconda is into subliminal suggestions in promoting his sales on this board".
My Random House dictionary defines "subliminal" as "existing or operating below the threshold of consciousness". If you are conscious/aware of what Adrain is doing, how can it be "subliminal"? Or, are you implying that his suggestions are "subliminal" to everyone on the board except for you and that you are just more aware than the rest of us? Please explain?
4) nwcs - I agree with your first sentence but not your second one when you say :Grading is inherently imprecise and a challenge to be accurate. However, it can always be consistent. So long as the rules stay the same". The first one pretty much negates the second one and the "rules" for grading are not precise or objective enough to be followed, even by robots. That is part of why grading is inherently inconsistent.
5) Bottom line, as good as it might (or might not) be, grading, even by the best graders in the universe will always be somewhat inaccurate and inconsistent. And, even if it were perfect, there are so many people out there who don't know how to grade, that there would still be complaints, probably, nearly as many as we have now!
<< <i>nwcs - I agree with your first sentence but not your second one when you say :Grading is inherently imprecise and a challenge to be accurate. However, it can always be consistent. So long as the rules stay the same". The first one pretty much negates the second one and the "rules" for grading are not precise or objective enough to be followed, even by robots. That is part of why grading is inherently inconsistent >>
i think what nwcs said is correct. if rules could be well-defined, and remain exacting and consistent, it would be possible to make grading 100% accurate and consistent. but it's a moot point, since it woudl be absolutely impossible to get every single person in the world to agree on an exact set of rules for grading, plus the rules would be so voluminous as to be impossible to follow.
for example, the rules might say:
rule 361-b: "for each nick on the coin measuring more than 3 micrometers, deduct 1/32 of a point from the overall grade, with the following exceptions:"
rule 361-b-1: "if the nick is on the rim, the deduction is reduced by 50%"
rule 361-b-2: "if the nick falls less than one millimeter from the primary device, as defined in rule 88-r, the deduction is reduced by 33%"
rule 361-b-3: "if two nicks fall within one-half millimeter of each other, the nick closest to the primary device, as defined in rule 88-r, is doubled in value.
and on and on with this nonsense. as you can see, the rules could in fact only be executed by a computereized robot, not by humans. IN THEORY though, they could be so defined as to make grading a science. it just isn't realistic.
again, this is a terrific thread, one that every coin collector ougtta read.
K S
I suppose you could program coin grading using these rules, but it would be "technical" grading. You are talking about nicks. That is a relatively easy part. How do you gauge luster? or eye appeal? or level of reflectivity of fields in a proof coin? Too many variables that are subject to interpretation.
NOTE: No trees were killed in the sending of this message. However, a large number of electrons were terribly inconvenienced.
Type collector since 1981
Current focus 1855 date type set
<< <i>4) nwcs - I agree with your first sentence but not your second one when you say :Grading is inherently imprecise and a challenge to be accurate. However, it can always be consistent. So long as the rules stay the same". The first one pretty much negates the second one and the "rules" for grading are not precise or objective enough to be followed, even by robots. That is part of why grading is inherently inconsistent.
>>
Again, it has to do with what is precision and what is consistent. It is possible to be consistently wrong. It is possible to be consistently off-target. Technically, precision can be obtained on an incorrect target, yet for our purposes precision is the degree of accuracy on the correct target. So it is reasonable to say that a process can be imprecise and partially accurate, but consistent.
To follow what you are saying, a human can be inconsistent in their grading with differing levels of accuracy and precision. But with a set of rules, consistency can be gained but not guaranteeing accuracy and precision. What you are referring to is comparing process to processor.
The process can be consistent and the processor be inconsistent. That's what I'm saying in my statement. I was saying that a process can be inherently inaccurate and imprecise, but consistent. But when it comes to a human grader, I would argue that a human is inherently inconsistent which means that precision and accuracy have to be taken on an individual case.
This is standard statistical quality control here.
Neil
Precision is how far you go in "quantizing" what you are trying to measure. Using coin terms that would mean a 10-point MS grading system is more precise than a 5-point MS grading system.
Accuracy has to do with how often you measure the same experiment and get the same results based on the same rules. Accuracy is very similar to Consistancy in this regard.
I agree though it is all semantics...
The problem with coin grading is that EACH person has a different set of standards. Therefore what is Accurate to one person is NOT Accurate to the next guy with a different standard. Therefore since the graders at the services are HUMANS they each will have a different set of standards. Most people will say that PCGS has 3 graders (or is it 2 graders) and a finalizer. I can tell you there is NO WAY that the vast majority of coins will get 3 graders and a finalizer. They'll get ONE at best. They have too many coins that come thru to do that. Hence, it just comes down to someone's opinion. As far as I'm concerned MY opinion is just as good as anyone else's, especially if I am spending MY money on a coin.
jom
Go BIG or GO HOME. ©Bill
<< <i>A rather time consuming and costly proposition just to prove what we already know regarding consistency from PCGS or any other service. >>
I've already done enough crackouts and re-submits to PCGS to know that they are not consistent, at least in my series.
Russ, NCNE
Grade and "Dang, that's one pretty coin" are two different animals.
peacockcoins
<< <i>It took me twenty years to learn:
Grade and "Dang, that's one pretty coin" are two different animals. >>
Now we are talking about animals? Boy, I am getting confused. But it does make me want to go hunting.
Anaconda,
I was taken out to dinner the other night by Win Carner and he spoke so highly of you.......I would like to either hear from you or meet you.
stewart
adrian
And I don't want MS65 81-S Morgans that any idiot can grade but something that will really put the certification companies to the acid test. How bout some brilliant white coins that have huge price jumps between grades? Do you have any of my favorites like the 79 CC? And no lock coins please. That would make it too boring. How bout some shot m65s? With a $10K difference in price that would be a very interesting experiment.
I would like to take 10 93-S Morgans in MS65 and try it. Again, not lock coins but lock, shot, and no way coins. That would be quite interesting.
I heard that there are some S mint Buffs and Indian Eagles that are tough to grade and we could throw in some early 1800's Bust Halves and see if the different graders can tell the difference between a coin struck with a hand press & open collar and one with actual circulation wear. We're getting into Market grading now and I would rather keep it technical because I want the graders to determine the grade and let ME determine the value. Down with Market Grading!!
The reason I picked Morgans is because I would know if the different companies got the grade right because I can grade Morgans with my eyes closed but I'm kinda iffy on some series I have no intererst in, like Indian Head gold and might not know the difference between an ACG shot 68 and an ANACS lock 65.
After we get that done we can do round #2 where we check the companies for consistency in their grading of Monster Toners, Morgans with DMPL fields, SLQs with Full Heads, and FSB Mercs.
This would be a very interesting challenge indeed and if carried out would make the front page of all the coin rags and be talked about for years!
Darn you sneakysnake for posting threads that make me think!
For some life lasts a short while, but the memories it holds last forever.
-Laura Swenson
In memory of BL, SM, and KG. 16 and forever young, rest in peace.
<< <i>Hey gmarguli, NastyNic, cointagious - you guys are great! I love to read your posts. Kinda like watching a fat lady run. Not very educational or thought provoking but always entertaining!
Keep us all laughing! This place wouldn't be the same without you!
adrian
anaconda.rare.coins (on eBay) >>
If it is not thought provoking, then why do you avoid answering my simple questions?
What's with talking in the third person? Are you confused?A person with your education should be capable of answering these.
I'm busy, never argue with fools and don't cast your pearls before swine. Your questions are too simple and your logo not subtle enough. Someday you may be a worthy adversary.
adrian
<< <i>Karl,
I suppose you could program coin grading using these rules, but it would be "technical" grading. You are talking about nicks. That is a relatively easy part. How do you gauge luster? or eye appeal? or level of reflectivity of fields in a proof coin? Too many variables that are subject to interpretation >>
luster could be mechanically evaluated as well, by measuring surface reflectivity at varying angles and through spectroscopic analysis. ditto for proofs. eye-appeal is a biproduct of all the other factors involved, ie. bulllseye toning vs. blotchy RB coins, etc. technology would make these measurements feasible. but certainly not practical, as we'd end up having to ship all our coins to NASA for definitive technical analysis....
coins in space?
K S
K S