Ebay Stamp Sales Totals Last 90 Days

Don I used the available top level ebay categories;




For example United States includes all the sub categories;





Of course there are singles, sets, box lots and collections included in my numbers. Looking back I think "listings" would have been better terminology than items.

Ikeypikey I agree with you, when I first looked at numbers I was seeing incredibly low median prices often less than $3, and that didn't fit with my experience. I find my auction sales to be quite robust. This further review shows that there are a huge number of items selling for $5 or less but the rest of the market shows the biggest $ sales being in the $25-100 range depending on category. There is also an interesting bump in sales between $1000 & $2000 which surprised me.

If anyone would like to see histograms for any other categories such as Europe, Worldwide, Asia just let me know.

This is brilliant KRelyea, thanks for doing this.

Now I have a yardstick that can be used to compare for example ebay to HipStamp as a selling venue for stamps. HipStamp recently posted that the site achieved $195,000 for the month. Now we can compare this to ebay's av. $11.5m. HipStamp has a lot of work to do.

Given that ebay is the dominant force in stamp sales, if we had a summary, say, every 3 months, we would know whether global stamp sales are generally trending up or down, a barometer for the general health of the hobby!

A few years ago, I wrote an article about the Israel stamp market on E-bay. I tracked all listing, and put them in a spreadsheet, then after a few months did some analysis. I was interviewed about it for Nancy Clark's radio show.

Here is my data:
Israel Stamps on E-bay
A look at what sold in the last 137 days

I have been collecting Israel stamps for about 20 years, and increasingly have turned to E-bay as a source of new material. Why? On any given day, I can log on to stamps.ebay.com and enter "Israel" in the search box and see hundreds of items for sale, with an average of 164 new lots listed each day. When I attend the local shows here, I am lucky to find a dealer with a handful of Israel items, and most of them are the same items they brought to previous shows.
Since I look at e-bay nearly every day, I decided to start tracking what was being sold, and what was not selling. E-bay does not make this easy to do, since there free search tools only allow you to view two weeks of completed items, and I didn't think it was worth it to pay for there merchant tools that allow access to 90 days of data. So, I decided to do what any computer engineer would do, and develop my own parsing tool to analyze the data. This tool allows me to take the output of the e-bay search and convert it into something that is useable in a spreadsheet tool. Now, I simply take the output of the daily search, run it through my parser, and add it to the previous day's data. As a result, I now have 137 days of sales data.
The data I track is simple. I look for any auction that completed in the stamp category with the word Israel in the title. Some items may slip through since the seller tagged them as Palestine or Judaica rather than Israel, but I catch the majority of the lots I am interested in this way. Once I have the data, I take each lot and place it in to one of 30 categories I have defined. These 30 categories and the lot break down are shown in the table below.
To date, I have data covering December 14, 2006 through May 1, 2007, a total of 137 days. During this period, there were a total of 22,451 lots listed and 10,203 of these sold (45.45%) for a grand total of $365,318.30. Therefore, an average lot sold for $35.80, while the median price (the price at which half the lots sold above and half below) was $9.95. On average, each lot sold had 3.47 bids, with one lot amassing 62 bids before bidding ended. On average, 163.88 lots were listed per day, with 74.47 selling for an average total of $2,666.56.
There were a total of 32 lots that managed to close over $1000 each, totaling over $55,000. What this means is that .03% of the lots sold account for 15% of the sales dollars. This leads to the large disparity between average price and median price. Of those 32 lots that closed over $1000, 20 were large collections, 5 were individual stamps (3 sets of 7-9 tabs, 1 set of J1-5 tabs, and 1 ministers sheet with 1-9 tabs), 2 were plate blocks (a set of 7-9 plate blocks and a complete collection of Zodiac definitive plate blocks), the remaining five were a souvenir leaf (#45c), a Palestine overprinted French stamp, a full sheet of the Petah Tiqwa stamp (#27), a cover with interim period stamps used as postage dues, and a set of sheets for #33-4. The lot that fetched the highest price was a large interim period collection, closing at $3739.89 after garnering 62 bids.
Some rather interesting trends became apparent while looking at the data. Wednesday has the lowest average dollars sold, but Friday has the lowest number of average lots and the highest average sales price per lot. Sundays have the most lots, and highest dollar sales. February had the highest success rate of the 4 and a half months I have tracked, while April had the highest average sales and most lots closing per day. January 25th was the day with the highest single day total, at just over $16,000, while December 27th had the lowest single day total, at just $468. Early February yielded both the day with the highest average price per lot (the 9th, at $145.08) and the lowest price per lot (the 6th, at just $8.56). March came in like a lion, with 73% of lots on the 9th successfully selling, but went out like a lamb with just over 21% of lots selling on the 28th, both of these are the single day leaders in these categories. January had both the day with the most lots (the 3rd, with 632) and the least lots (the 26th with 41).

Does this compensate for lots that are listed in two different categories? Big sellers like NobleSpirit and nystamps usually list country collections (and sometimes singles and sets) in both the home country category as well as in Worldwide->collections, lots. Those two sellers alone could seriously distort the data if their sales are counted twice.

I recognized the 2nd category problem (I always use 2 categories on my listings) but didn't have a good idea how to correct for it, for this reason my numbers are inflated to some degree.

Thanks ADCAPLAN for that study , interesting to see your breakdown .

Thank you for the very informative analysis and discussions.

What an insight! Thank you Ken and being a buyer on ebay sometimes, this data is so insightful..

I have reviewed the data with regards to items listed in multiple categories, it's a bit confusing but here's what I found.

As mentioned nystamps and Noblespirit often list items in 2 categories. There are some random usages by smaller sellers but these 2 sellers are by far the biggest users of 2 categories. Nystamps uses 2 categories when selling country collections and lists them in Worldwide and the specific country. Individual stamps or sets use only 1 category. Nysamps sold approximately 2000 items in the Worldwide category and almost all of these used 2 categories.

Noblespirit uses 2 categories for almost all their listing, however when it's a United States item they list it in 2 US categories and the ebay data I used does not duplicate the numbers. They sold 4000 items in the Worldwide category these were essentially all duplicate. So the bottom line is that the Worldwide category is inflated by at least 6000 items with some additional duplicated scattered among smaller collectors.

I am one of the smaller sellers that add to the problem because like Noblespirit I used 2 categories for all 500 my listings. I think the extra exposure is well worth the additional Insertion fees. I'm sure there are others but they are buried in the data and I don't have the time to dig them out.

I backed out the 6000 Nystamps and Noblespirit sales from my data and the revised my data as shown in the charts below;








I want to thank everyone for the kind words and I will update these charts in 3 months.

There is one other problem with the data that I noticed. In the US category it look like a LOT of the higher priced sales are from the cartel and I'm not sure these items actually sell, but I don't know how to adjust for that.

Thanks Ken for spending time on this; I enjoy 'numbers'.




In additional to 'second category' listings and other fraud/misidentified adjustments there is also the matter of 'returns'. Depending upon who you listen to, ebay returns run anywhere from 0% to as much as 5%. There is some overlap between the return rate and the fraud/misidentified group. I am not sure how you might parse these. Typically in brick and mortar store 'returns' are still counted in any sales numbers but this might be changing with online sales due to the more substantial issue it has become (buying things sight unseen).

Another issue is that many buyers have 'built in' the cost of shipping by offering free shipping in the listing and raising the listing starting price. Other seller do not do this but rather charge it separately. When you calculated your numbers did you include shipping numbers?

The key is to understand exactly how the numbers are gathered and compiled; these details are often buried in the compiled metrics.

As an aside, it is so funny to watch the current pollsters and media completely miss the mark on virtually everything they gather metrics on. I think they are finally picking up on the fact that phone polls are no longer accurate and they will have to change they way they gather data. (Everyone now has caller ID and large groups of folks are now not answering the call if they do not want to reply to a pollster.)
Don

What I find most striking is that the median stamp sale (site-wide) is only ~$5. What an embarrassing number!

Why is a median price of $5 embarrassing? The vast majority of stamps are worth less than $5, and since most dealers don't carry them, since they aren't worth the time and effort, why shouldn't collectors find these on E-bay?

So what are we supposed to do with this data? Is anyone going to change their buying habbits on ebay because of it? I doubt it. Is anyone going to change their selling habbits on ebay bcause or it? Perhaps, but I wonder what the logic will be.

Is the distribution shaped the way it is becaquse that is what buyers are demanding? Or is it strongly influenced by what and how sellers are offering their singles, sets and collections on ebay? Certainly what is being offered on ebay is not identical to what one would find at a stamp show or in a conventional auction, so the sellers are definitely doing some of the shaping of the distribution.

Very interesting stuff. I suspect that a huge chunk of the Asia category are China stamps. "Europe" can be misleading as the category includes all the colonies and territories which in many cases are much hotter than the mother country (e.g. French colonies vs. France).

ebay is still such a crap-shoot. As a collector of strictly MNH material I am still amazed after 18 years on ebay how sellers optimistically morph all kinds of conditions into "MNH." To what degree does this affect philatelic sales on ebay? One can but wonder.

I wish ebay would standardize quality notations the way they do with other categories. For example, a MNH stamp can also be called NH, Never Hinged, NHM, UM, UMM, Unmounted, No Hinge, Not Hinged, etc. Would be amazing for buyers and sellers alike for ebay to pick one term, make a box to click in, and save space in the title for other info.

It would be for sellers to possibly change their selling habits.