Rob

Learn your subject and you will get a good feel for what is available. Trying to construct a one size fits all rarity scale cannot work because you are combining similar data in theory, but differently skewed by the sampling methods employed and no rational way to weight them relatively. i.e. A variable standard.

The trouble is that it is booking x5 or more compared to the rest. 1932 & 36 have only two and a half to three times as many, but sell for considerably cheaper than they should based on this factor. I notice that this year's Spink addressed the discrepancy to some degree.

An investor won't care about relative rarity, only whether the price paid is likely to generate a return.

There is also another skewed parameter to muddy the water. Collectors tend to put off the purchase of the more expensive items until later so that excessive amounts of capital aren't tied up. By extension, it means that the expensive coins often return to market a little earlier than one might otherwise expect. This improves availability for the more desirable pieces, giving the impression that they are more common in absolute numbers than a proper census would suggest..

Yes - I understand this point. I think Rob is conflating 'rarity' and 'availability', which is of course a perfectly justified and meaningful position. But then there is 'absolute rarity' which - as you say - includes all specimens held by collectors whether or not they appear on the market from time to time. Coming as we all do from a collector's viewpoint, I take rare to be a term used with regard to the number of pieces potentially available to collectors. This would include all those privately held whether long term or not, but not those in museums which are unlikely ever to be in the market again. So, something where there is only a very small number available would have to be considered rare - take an arbitrary figure of up to a dozen for example. What I do not consider rare is something like the 1934 crown. Sure they only made 932 of them, but virtually every other sale has at least one in it. They are available even if not in the quantity the market could fully absorb and can't in my view be classed as rare. There must be a couple dozen going through UK auction houses every year with more being sold privately to dealers. Conditional rarity is a different matter as exemplified by the more than 4000 1925 halfcrowns sold at Noble a few years ago. Of 11 lots, only one coin warranted a lot to itself reflecting the difficulty in obtaining a high grade example, but it isn't that they aren't available.

Much of the nit picking has concerned the terminology 'excessively rare'. Whichever system you use be it Rayner, Freeman, Peck, or whoever's, excessively rare is the one below unique. The reason I used excessively rare for the groat was the existence of a second example in the BM. The fact that only one is available to collectors does not make it unique. I prefer the approach of Cope and Rayner with variable grade/rarities shown. If you want to assemble a date run with anything that is available in any grade regardless, with the exception of a handful of difficult dates, I think you would complete most gaps within a few days by trawling the internet. The net has made a lot of coins more common than they used to be as you have a greater number of dealers at your disposal. A further consideration with regular currency pieces is the reluctance by many to overspend on a coin because of the published prices in the guides. This enhances the rumours of rarity, whereas in essence it is really reflecting the willingness to pay. If you posted a message on a board saying you would pay £50 for a £10 coin, you would soon get a reply. A fairer expansion of the rare definition would be the difficulties encountered in finding a coin in an exact grade or the slab with the highest number, with a maximum number of marks, with a certain colour tone, at a certain price relative to the books and all with eye appeal. The last sentence is probably a reasonable summary of the main impediments to obtaining a coin. As always, it is each to their own interpretation.

Someone tell him he's a ****. Why quote Marsh on the second item, when you can't be bothered to check the images of the 1817. Numerals are arabic 1s, missing serifs are blocked dies or broken punches. Simple. I cannot understand why people get so obsessed with infinitessimally small incremental die changes due to being used for their original purpose. I would like to claim the well deserved excessive rarity of my unmodified, struck as intended, virtually impossible to obtain normal type coin.

If people want rarities they need to specialise and dig deep. Things might be difficult to buy because there isn't enough to satisfy demand, but that's not the same as saying they are rare. What it probably means in all bar a few cases is that they aren't prepared to spend more money than the opposition.

There are very few coins for normal run of the mill collectors that are virtually unobtainable. Just taking milled pieces and leaving aside the unrecorded varieties which once announced appear from every nook and cranny, none of the circulation crowns would fit that description. Halfcrowns the hardest to get would be 1667/4 (3 known, best is fine) and elephant below. Even the 1823 first issue is commoner than these two. Shillings, nothing springs to mind apart from possibly the odd transposed shields variety. 1850s come up every year or two. etc etc. We have the question of some of the small denomination dates, but here we are not even certain they exist.

20 pieces is relatively common. I think you need to get away from the 'isn't available today' approach. Most coins where there are only a few known don't get on the radar of the average collector who is usually concentrating on the more mundane varieties. Price is a factor here because many would steer clear of something which is going to cost them multiples of the common type. In these instances it is a case of waiting patiently for what goes around to come around. Anything with 20 pieces will probably come around once a year to a year and a half on average. With say 4 or 5 pieces they might come around every 6 to 8 years. I would call that extremely rare but not excessively so which I would take to imply virtually unobtainable. Availability in the case of seriously rare pieces is effectively determined by the average time taken to form a collection, as once it is complete the pieces come on the market again if sold. Obviously it can't apply if the collection is set aside for generations or passes to the next generation as an ongoing concern, but most collections are not like this. Most serious collectors don't really go for decent collections including major rarities until they are middle aged with a bit of money in the bank, so a generation is probably the time limit for most people to collect. That is the basic criteria I used in arriving at the above figures. In the case of the coin above, it was available in the market to buy in 1919, 1956, 1979, 1995 and 2011, so available every 20-25 years on average, or one generation. Then it only depends on the depth of your pockets.

I would suggest a case where there is one available to collectors or possibly 2 only. There is an example of this in the BM but I don't know what it's like. Oxford 1644 groat (Morrieson F-2) with the Rawlins signed bust and first issue style line declaration reverse as opposed to the later cartouche. Not in Brooker, not in Morrieson who referred only to the BM's piece in his 1922 BNJ article, not in the most comprehensive Charles I collection I know (and that has more varieties than Brooker's silver) and not in Montagu either who would certainly not have passed up the chance to get one. Worth noting that despite being double struck and only around the VF mark, it was illustrated in Hamilton-Smith's 1919 sale which would not be normal for a commoner piece, or even some rarities.

Oops. Sorry, misprunt. My brain was following the title which says 1662-1972.

There's always going to be a few anomalies, but the basic principle is sound. Tongue in cheek, maybe Seaby had a load of 1923 halfcrowns to shift.

I can see the rationale for a logical numerical progression of absolute numbers, but a scale such as this merely tells you how many are out there, not the numbers typically available which defines rarity for the collector. 1000 collectors and 500 pieces known is extremely rare in terms of availability because most will be held in collections and so unavailable to others. 500 collectors of a piece where 1000 are available will always be for sale in a tray somewhere. There was a book written by Cope and Rayner in 1972 called English Milled Coinage where an estimate of the rarity for milled coins in a given grade was made. This is far more useful than a number representing the pieces available which can not anticipate supply vs. demand and hence to a certain extent, price.

The reverse is a bit better in the centre, but it is weak at the periphery. There is a lot of fine detail missing from smoothing. Here's another one.

Fine Spink 2791

You want a proper touchpiece, not the milled ones. A bit of 'ammered gold with an 'ole

Rob, That was a nice coin, looked even better in the hand. I was hoping to get away with a sneaky cheap bid on that one, but is was sought after. Picked up several bits and underpaid with all of them. Regards Mark I was the underbidder

Sharp E1/1 im. Bell.

No I would have bid on line. The only other thing I was remotely interested in was the 3a1 halfcrown. mm. bell which appeared to be over a portcullis. Presumably this was a residue of the previous incarnation of the die as a type 2 halfcrown as there is no portcullis 3a1 to my knowledge. An interesting oddity, but not a must have coin.

I was contemplating that one too, but in the end decided not to bother as I only need a basic type. And this and the next lot were a bit too bright for my liking with no toning. I like my toning.

I have angels both pierced and intact but sadly, none for sale at the moment.

Get yourself one. They are still affordable. Get a holed one used as a touchpiece and the price drops by a half or more. I like them.

Banking as a whole would benefit from transparent pricing. If people paid for the services they used instead of having the cross subsidies as now which offers free banking at random, the systemic risk should be easier to manage with fewer exceptionals on the horizon. There would be less scope for profit and therefore bonuses (which will satisfy the left), and no need to employ unnecessary staff on account of automation (which will satisfy the city). Profits would be dictated by the efficiency of the system used in relation to a capped charge.

At the risk of opening a very large can of worms, here we go again on the variabilities of slabbing the same coin on different occasions. Many coins, maybe even a majority would come back a different grade following resubmission. I have had several coins that were graded differently. Someone posted a line of 5 resubmissions of the same coin which came back 67-69. Just accept and move on. Considering the coin alone and paying little or no attention to the number will stand you in good stead.