From hot mess to hot damn, here's a few coins from my collection going through all the grades commonly used with ancient coins. Maybe a useful reference to some and for me a working draft of how I will approach the subject in my next book. Poor: a coin identifiable to both type and ruler but few other details remaining. Fair: A coin grading fair will be heavily worn and/or have other problems but its major features will still be easily recognizable. Fine: The quintessential fine is a coin which is in overall good shape but has been well worn. Sometimes an About Fine, Good Fine or Choice Fine serve to better place the state of preservation. Shown here in progression from aF to F+ Very Fine: This tier represents the bulk of collectible ancient coins. They are well-preserved with little detail missing. As in the previous range, VF coins often carry a plus or minus qualifier. Here an about (or "near") very fine, very fine and choice very fine. (continued on next post due to image limit)
EF: Relatively few ancient coins remain in bona fide extremely fine condition. It describes those coins which show few imperfections and only minimal wear. They would be the equivalent of U.S. coins that grade "about uncirculated". An extremely fine coin could be slightly demoted to the about extremely fine condition when a small amount of wear or other imperfections as scratches, die cracks or slight waviness detracts from its otherwise impressive state of conservation. As mentioned above those coins fully within the extremely fine range should at first glance look nearly perfect And the rarefied few reaching choice extremely fine, a small percentage of EF coins as a whole, should be reserved for those whose surfaces are pristine, are well centered and have had almost no circulation wear. Arguably the highest grade possible for an ancient bronze coin. FDC: The ultimate grade, which stands for Fleur de Coin, is a term unfortunately oft misused by the overly hopeful and dealers looking to maximize profit. The few coins properly deserving of this grade should be essentially perfect in every way with no appreciable wear or defects of any sort. Rasiel
I got this Gothicus, Always thought of it in the VF category, I can see the presence of corrosion in the fine photos.
Grading is of course to some extent subjective. Personally, I would not have graded the Claudius Gothicus - VICTORIA AVG and the Byzantine Histamenon as EF. The Gothicus shows clear signs of wear and no silvering. The Byzantine Histamenon shows significant scratches on the nose of one of the figures. The Probus, however, is a clear case of EF, in my view. Also I would practically not grade any ancient coin as FDC. In my view, FDC is only applicable to modern, machine-minted coins. An ancient hand-struck coin in FDC would have to be a perfect strike from fresh dies, complete mint lustre, perfect untouched and naturally toned surfaces and of course not a hint of wear or die imperfection. But again, grading is subjective.
I saw from reputable dealer some so called FDC denarii but for me you shouldn’t use FDC for antics even if dealers told you that you can find one. The FDC I saw were CHXF or SuperbXF in fact. But the dealer , in Belgium, was honest in the sense that it consider that you cannot see scratches or wear ever at x10 magnifier that was true and the price he ask for that quality was the price for a XF. He told me that I was right to tell it CHXF as with a FDC grading he could sell it for years as no collector want it and are not looking for FDC. So I bought it as a XF in fact it is a CHXF according to an other expert , for 300 USD , I saw the same in CHXF sold for 800 USD at an auction.
There are somewhere between two and two million opinions on the grading of ancient coins. Some, like Ras, merge wear, corrosion and all other 'situations' into one rating somehow equating a slightly worn but smooth coins with an unworn but corroded one. Others, like NGC ancients, reserve the words (fine and friends) for wear and apply some other way of informing the user of 'situations' other than wear. In 1997, I proposed a three part system: http://www.forumancientcoins.com/dougsmith/grade.html Wear - Standard Grading Things that happened before the coin left the mint - Conditions of Manufacture Things that happened after minting - Conditions of Preservation These last two are very slightly different from what the NGC system calls Strike and Surfaces but I did not attempt to assign a value number which I considered like comparing apples to oranges. How many fine scratches equal one gouge? How many letters completely lost to poor centering equal losing half of every letter due to a small flan? If you visit my page above, do click through to the pages with 96 images defining several 'Conditions'. This is one: VF Ragged flan Septimius Severus denarius Some flans were not round and smooth. Some collectors discriminate against these coins. Take care to distinguish this from a chipped flan. There would need to be more than that if the job were done correctly allowing for the countless combinations and degrees of all of the factors presented and the many others I missed. This is why I now subscribe to a two tier grading system: I like it. I don't. Simplistic and non-commercial? I have been called worse. I have vague memories of someone way back when proposing a numbered grading system ranking non-wear factors but I forget the details and the source (Was it Paul Rynearson?). Do any of our long term collectors remember this?
Perhaps I can share some data from my data-mining algorithm, to present an overview of the use of these terms. I've got 13 datasets, extracted from SixBid, with a total of ca. 10.000 auction results (~individual coins) of very different categories: Code: Aethelred 1006 Attica tetradrachm 1035 Byblos 110 Caesar denarius 837 Caracalla tetradrachm 1533 Claudius aureus 376 Eravisci denarius 188 Hadrian aureus 886 Nero tetradrachm 1381 Nero_2 denarius 601 Octavian aureus 318 Patraos 479 Sceatta 660 Code: Aethelred Attica Byblos Caesar Caracalla Claudius Eravisci Hadrian Nero Nero_2 Octavian Patraos Sceatta good 11.23 22.61 20 20.19 21.66 17.82 11.17 23.14 12.89 17.80 28.62 17.95 21.06 vg 1.79 0.00 0 0.96 1.83 54.26 3.72 74.04 5.43 47.42 46.54 4.38 8.33 gutes 0.30 0.00 0 0.12 1.83 5.32 0.00 3.61 1.38 1.66 1.57 8.35 0.00 fine 42.25 61.35 66.36 58.18 48.47 41.49 15.96 53.27 49.38 45.59 75.16 35.49 46.82 schön 8.95 0.00 0.00 1.91 8.15 22.07 5.32 11.63 14.99 10.65 5.66 19.21 1.82 very fine 34.89 35.27 20 30.35 31.51 25.27 13.30 24.04 39.46 28.62 40.88 23.17 29.55 aef 0.00 0.00 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 avf 0.00 0.00 0 0.00 0.00 0.00 0.00 0.00 0.00 0.17 0.00 0.00 0.00 vf-xf 0.00 0.00 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ss 0.70 0.00 0 0.00 0.13 0.00 0.00 0.34 0.29 1.33 1.26 0.00 0.00 s/ss 0.00 0.00 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 vz 0.00 0.00 0 0.00 0.20 0.00 0.00 0.79 0.14 0.17 0.00 0.00 4.55 vz-ss 0.00 0.00 0 0.12 0.07 0.00 0.00 0.11 0.00 0.00 0.31 0.21 0.00 ss-vz 0.00 0.00 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.63 0.84 0.00 sch 0.00 0.00 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 gutes sehr schön 0.10 0.00 0 0.00 1.24 4.26 0.00 1.92 1.30 0.67 1.26 7.10 0.00 sehr schön 8.45 0.00 0 0.00 7.37 21.81 4.79 8.69 13.98 7.32 4.72 16.28 1.67 choice 1.19 8.21 4.55 8.36 3.20 3.72 0.53 8.58 1.96 2.50 4.72 1.46 2.42 ch ef 0.00 0.00 0.00 0.60 0.00 0.00 0.00 0.11 0.07 0.00 0.00 0.42 0.15 ef 34.99 72.66 73.64 9.20 37.64 28.72 19.68 59.59 26.00 38.44 38.05 41.54 26.36 xf 1.49 1.84 5.45 5.62 2.22 1.06 0.53 3.50 1.52 1.66 1.89 5.22 2.12 axf 0.30 0.10 0.00 0.36 0.65 0.00 0.00 0.00 0.14 0.17 0.00 1.25 0.15 ef 34.99 72.66 73.64 9.20 37.64 28.72 19.68 59.59 26.00 38.44 38.05 41.54 26.36 xf 1.49 1.84 5.45 5.62 2.22 1.06 0.53 3.50 1.52 1.66 1.89 5.22 2.12 extremely fine 3.48 24.06 43.64 20.55 15.33 13.03 2.13 20.88 3.48 12.98 36.48 12.53 12.12 vorzüglich 6.56 0.00 0.00 0.00 4.89 5.85 0.53 3.27 2.53 3.83 2.52 17.95 3.18. unc 1.99 0.58 4.55 0.84 4.57 1.06 1.06 3.27 2.32 2.16 8.49 2.92 3.64 ngc au 0.50 1.64 0.00 4.66 0.85 0.27 0.00 1.81 0.29 0.17 1.26 0.21 0.00 grade au 0.00 0.19 0.00 0.36 0.13 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ngc ch au 0.00 0.10 0.00 0.48 0.07 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ngc ms 0.30 8.99 0.00 1.19 0.59 0.00 0.00 1.13 0.00 0.00 0.63 0.21 0.15 fdc 0.00 0.00 0.00 0.84 0.52 1.06 0.00 1.69 0.00 0.00 1.89 0.00 0.15 uncirculated 0.60 0.00 0.00 0.12 0.26 0.00 0.00 0.11 0.00 0.00 0.31 0.00 0.00 mint state 0.00 11.88 0.91 0.84 0.59 0.27 0.00 0.79 0.07 0.33 0.63 0.00 0.15 near mint 0.00 7.15 0.00 0.36 0.20 0.00 0.00 0.34 0.00 0.17 0.00 0.00 0.00 Cointalk is not really super with tables. Anyway. Numbers are percentages, and are coded hierarchically (i.e. if the text states "Good very fine"; both "good" will be TRUE, "fine" will be TRUE and "very fine" will be TRUE. In the end, the highest grade is assigned to the coin (so "very fine").
That is interesting. I am quite particular when it comes to condition, but a "ragged flan" is no problem in my view (as opposed to a chipped flan). I find these "ragged flans" quite attractive, as they can add to the character of a coin.
If I understand that the data mining is a pourcentage of the grading words on a large number of auction I suppose that it is in % ? If my analysis is fine FDC or mint state account for marginal word use reflect that there are in fact no true FDC coins. what will be more interesting I think will be to have a list by specific coin that reflect the best known grading you can have for this coin. IE Trajan Sesterce RIC xxx. The best you can find is a ChVF... and so on, what do you think of that method combined to data mining ?
That seems to be the case. As for Ras's examples, I'd rearrange many of the coins and as for that "FDC" example... what?! Surface discoloration, small flan edge splits, die wear (flow lines, poorly rendered dotted border), slightly sloppiness of letter engraving, ho-hum reverse style (although the portrait is indisputably wonderful), wear or incomplete die fill on the high spots of the reverse... how can that possibly be considered "FDC"?? It's a lovely coin, surely EF, but no way is that "FDC".
I would say that is absolutely ridiculous were it not for the fact that almost no coins are listed by dealers graded 'Good' (AKA not good enough to be VG). Good is used for two different purposes. Your data mining needs to be able to distinguish between 'Goods' followed by another grade term (F, VF. EF) and 'Goods' meant to stand alone indicating a low grade coin. Similarly you will encounter hyphenated grades meaning between (G-VG, F-VF) and split grades where the listing indicates the two sides of the coin are not equal (F/G). Simply mining words without meaning is, well, meaningless. Automated data mining reminds me of books written by professional writers who knew nothing about the subject before they were assigned the project. The book can end up spreading more ignorance than it educates. Of course there is the other problem of experts who have no idea how to communicate their knowledge to human beings who do not already know the subject. I suppose that these efforts might eventually grow and develop rather like Google of 2020 is better than Altavista of 1995. Still, Primum non nocere is a concept for use in all activities and not limited to doctors.
You can't simply describe an ancient coin on a scale from poor-FDC without also describing its surfaces, centering, strike, flaws, etc. Here are two coins, nearly identical but for state of preservation. In the first, we have less wear, with good centering and a strong strike, but rough surfaces. Its patina is jet-black. Simply calling it gVF or aEF overestimates its desirability: And here's one with more wear, good centering but a weak strike on the upper part of the reverse. Yet it has smoother, glossy surfaces and a bottle-green patina. It has more eye-appeal, in my opinion (in hand) than the first coin, though it would sell for less money than the first example: Grade is only part of the story. I like Doug's grading system, for it takes into account many features that make a coin interesting/undesirable.
I have previously shown these two coins which I feel sums up the problem as I see it. Which is the better coin to retain for my collection? I have never been able to accept either one as my only coin of the type. gouges, scrapes, second rate style 'honest?' wear, lacking legends, better style I admit curiosity who bought the much better one that sold last month. It was pretty if you like 'smoothed' (their words not mine) coins. Anyone here?
Here my XF Trajan Denarius RIC 58 , do you agree with the rating ? From a famous European dealer Elsen in Brussels Belgium. Sometime I think I was silly to buy it 250 USD .... but I love it
In my view this is a VF or gVF, but again, grading is to some extent subjective. Elsen is of course a serious and reliable dealer. I have a few EF denarii of Trajan in my collection - will show one later.
As I prefer to be pessimistic will say gVF , usually this dealer don’t overgrade coins he’s very reliable but XF is a bit optimistic or subjective and all the VF I saw in auctions were really worst compare to this one. But how to tell a well known dealer long time established that it is not for you a XF you you are not willing to pay for a XF. I don’t bargain when I am in a famous dealer shop ,it is not like in a fair. Ok I pay sometime a bit too much at its shop but he make me also very good deal on some coin that overall it is a balance. How do you deal with a famous dealer in his shop to tell him it is overgraded and you are agree to buy only at a lower grade price ? A good question interested to have your opinion as pricing is related to price that a dealer ask for.
"Fair" has a specific meaning for US coins. See http://coinauctionshelp.com/how_to_grade_draped_bust_large_cent.html . You will confuse everyone if you have a grade "Fair" that is better than "Fair" for a large cent. Your system needs to make sense both to people coming from classic coins and people wanting to predict the grade NGC Ancients will give them if they slab. Stay as close as you can to those two things. In addition to Doug's pages take a look at Robert Kokotailo's pages http://www.calgarycoin.com/reference/grading/grading.htm Sometimes we see Islamic coins that are 100% Arabic text, and the text is 100% there, but the coin is only Fine because the text has lost all depth. Roman coins usually have a face on one side, but your grading explanation will be better if you challenge yourself to write sufficient descriptions to explain the grading of "text and a wreath" and not just faces and horses.