Accurate Coin Populations

Look closely at Heritage sales some time - you'll normally see specific coins on the block 4 or 5 times over the years, skewing the apparent total. Other examples never appear at all, so any estimate is just that - an estimate. I believe people like Doug who devote their professional lives to certain specialties before I believe TPG information.

You are never going to get accurate #s - multiple submissions and cross-overs hoping to upgrade skew the counts.

In the 3rd edition published in 2008, Winter estimates total number of 1846-C half eagles as 100 to 125.

He then estimates the number grading VF, XF, AU and UNC.

The 30 to 40 was his estimate of those grading XF.

-MV

One thing about publicly stating something is rare or only X # are known, it will almost always bring more out of the woodwork. Good example can be seen in the large cents. In 1949 the 1796 NC-4 and 1803 NC-1 had three of less pieces known in collectors hands, an R- 8- (Definition of a NC) Today they are R-5- with close to 75 pieces known of each. The 1825 N-5 was described by Andrews in 1883. Newcomb listed it because Andews had, but not a single piece had even been seen including the coin Andrews took his description from and it was de-listed in later references. In 2010 or 2012 I forget which, Tom Deck discovered one in a ebay lot. Today there are seven of them known. For ever 130 years years no one saw one and then seven found in less than five years. Why? Because no one was looking for it. It had been called and "error listing" and that the variety didn't exist. But once one was found and announced people started looking specifically for it and they started being found.