Normally my analysis centers around how various titles are doing and what trends seem to be influencing the sales. This time around I'm going to do something a bit different and focus primarily on a single title and get into a bit of the math behind the sales estimates. But don't panic, I'll be doing all of the math.
I mentioned last month that it might be a best case scenario if the current direct market infrastructure isn't meeting the full demand for the products. It would mean there is a larger demand for these products and room for the industry to grow. This month we see some potential proof of that with "Gears of War" #1 which Rich Johnston's Lying in the Gutter column mentioned a few weeks back (http://www.comicbookresources.com/?page=article&id=18852) as having sold over 450,000 units with only 10% having gone through the direct market.
So, why then do I have the sales of "Gears of War" #1 as an estimated 19,715 units?
For the sake of discussion, since Rich the entry marked as green ("the real deal") we'll take the statement at face value. In other words, we'll be working on the belief that "Gears of War" #1 moved the 450,000 units.
For starters, let me answer the question of how I got to the 19,715 units in the first place. I'll try to show my work as much as possible.
We'll have to get into a little math to explain this but it is very easy to understand. As long as you (or your calculator) can do basic multiplication and division and you understand the concepts of percentages and what a variable is, you are completely capable of understanding the math involved. In other words, anybody that didn't completely flunk pre-algebra should be fine. And those that did should still be able to follow along but may have to take the math on faith.
The top 300 comics and top 100 trades lists are algebraic equations with the unstated variable X being the sales of the key item. The index values are the percentage of that X variable that the item sold. In the case of October 2008, the key item was "Batman" #680 which sold 100.00% of X.
As a quick aside, it is interesting to note both covers of "Batman" #680 were included in the 100.00 index value. In some past months, only the regular cover counted toward the 100.00 value with the incentive cover being added on top of that resulting in an index value like the 103.06 value we had for "Batman" #678 back in July 2008. This is an inconsistency in the way Diamond generates the list but doesn't impact the math involved in unlocking the list.
The value of X can be estimated by taking known sales of various items on the list and dividing them by the respective order index values for those items. Obviously this requires data beyond just what Diamond provides to "unlock" the list and convert the index values into estimated units. This as an estimated value for X since the index values only got to two decimal places and this isn't precise enough data to compute the exact value of X. Fortunately we don't need the exact value of X and are able to work with an estimated value.
So how accurate are these estimates? Generally speaking, the margin of error for the estimated sales figures is about half a dozen or a dozen units. Mathematically speaking, the margin of error is plus or minus 0.005% of X (the sales of the key item).
Different people using different data from different publishers routinely end up at index values which match within the granularity of the data provided by Diamond. This serves as a method of cross checking both the results and the source data provided by Diamond. The fact the different people using sales data for different items on the list can unlock the list and arrive at values for X that match within limitations of the precision of the Diamond data means that the process is mathematically sound. The accuracy then boils down to if the data released by Diamond and the source sales data used is accurate. On occasion, errors in the data released by Diamond have been found and the people at Diamond are very good about releasing a corrected data set if errors are found. As such, there is no reason to believe that Diamond is somehow rigging the charts or otherwise releasing misleading or incorrect data.
My number crunching resulted in an estimated index point value for October 2008 of around 1,039.29.
"Gears of War" #1 had an order index of 18.97. Multiply this order index value by the index point value of 1,039.29 and the result is an estimated 19,715.3313 units which rounds down to 19,715 units give or take 5.2 units. What this means is that "Gears of War" #1 sold between 19,710 and 19,721 units to retailers during October within the direct market serve by Diamond. All of this can be done with a rudimentary understanding of algebra.
Remember, for the purposes of this discussion we are working on the belief Rich was entirely correct about how many copies "Gears of War" #1 sold. Obviously 19,715 is a far cry from the 450,000 Rich mentioned in his column. It is this apparent disconnect which some people have a hard time understanding.
The two numbers are measuring two very different things. Rich is talking about the total sales of "Gears of War" #1 which includes both the sales through Diamond and those outside of Diamond. The data in my reports each month is the estimated number of units reported by Diamond as being sold to direct market comic book retailers. Obviously the total sales (the 450,000 number Rich reported on) will be more than the sales through only Diamond (the 19,715 number I arrived at mathematically) since the total sales includes more than just the sales through Diamond.
Anyone thinking this somehow proves that the Diamond data is wrong doesn't understand the definition of the Diamond data set. The data from Diamond covers the sales through Diamond to retailers for the month in question. Any sales through other channels that don't go through Diamond are not included in the data Diamond provides. Clearly those other sales are important in the grand scheme of things but they are outside of the scope of the Diamond data set. Just as you wouldn't expect Burger King to report on the sales of Big Macs at McDonald's, it is unreasonable to expect Diamond to report on comic book sales they had nothing to do with.
I try to mention as clearly and often as possible that the data here only covers the sales through Diamond. While I would love to expand the coverage of my number crunching and analysis, doing so requires additional data which currently isn't available.
To know the full extent of the sales either the publishers need to release that data or other data sets covering the sales outside of Diamond need to be available. Neither is the case.
DC Comics has a policy preventing the release of sales figures. Since this is data about the core business for the publisher it is completely understandable why they don't make this information publicly available.
Ingram, a magazine distributor which provides comic books to many mass market outlets, also has a policy preventing them from releasing sales figures.
Another option is the BookScan data. Unfortunately it is too expensive for the limited financial resources I devote to tracking comic book sales. Even if cost weren't an issue, it only covers around 75% of the mass market book sales and there are limitations on reporting on the BookScan data. BookScan is a third party data collection and reporting agency which sells the data. Business would dry up fast if anybody who bought the data could just give it away.
One of the non-Diamond data sources available is the annual Statement of Ownership printed in comics with mail order subscriptions sent via second class mail. Marvel still prints these but DC switch methods of delivery for subscriptions ages ago and hasn't been required to print a Statement of Ownership in over twenty years. Although the percentage varies from title to title and year to year, the sales through Diamond average well over 75% or more, sometimes as high as 90%, of the overall paid circulation for ongoing monthly Marvel titles. Obviously different publishers will have different sales trends across channels. It is impossible to do more than a ballpark estimate of the percentage the direct market sales account for of the overall sales since the totality of comic book sales across all channels for all publishers isn't known.
When someone says the Diamond numbers are "wrong" all they are doing is illustrating a lack of understanding of what the Diamond data is. These claims of faulty data stem from the incorrect belief that the Diamond data is meant to measure all of the sales of comic books. Diamond makes no claims to this effect nor do any of us reporting on the Diamond data.
Why then report on the Diamond sales estimates?
While the data isn't all-inclusive, it allows for the identification of sales trends at the retailer level. These trends are a reflection of the sales trends at the reader level. The information available indicates the majority of comic book sales occur through Diamond. It is also safe to believe that the majority of the core audience of habitual comic book readers is covered by the Diamond data. So, by looking at this data we can see the sales trends influencing which titles survive and which get cancelled.
Without analyzing this data, the only people who would know putting a variant zombie cover on a Marvel title would result in 5,000 to 7,000 additional sales would be the people at Marvel. The only people who would know "Nightwing" is ending because of editorial decisions while "Family Dynamic" got cut short because of low sales would be the people at DC. By analyzing the data and reporting on it, things that might seem like puzzling actions by the publishers like the recent variant monkey covers at Marvel and the constant sequence of event titles at Marvel and DC become much more understandable.
The question remains as to how and why "Gears of War" #1 sold a total of 450,000 units but had less than 5% of those units reported by Diamond down in rank #134 on the top 300 comics list for October 2008. That is a really good question and one worth discussing further. Clearly those other sales happened outside of the Diamond sales channel. More to the point, they happened outside of the direct market comic book shops and therefore outside of the scope of the Diamond data.
"Gears of War" #1 was anything but a typical comic book. It was released around the time of the release of the second "Gears of War" video game. That game sold over 2 million units during the first weekend it was available. The game was promoted with a strong advertising campaign including television commercials. The first six issue story arc bridges the gap between the first and second video games providing an incentive for hardcore fans of the video game to get the comic book.
"Gears of War" is the sort of title one would expect to do abnormally well outside of the comic book direct market, particularly if it were available in stores selling the video game. After all, it is written specifically for the fans of the video game.
What Rich didn't mention in his column, presumably because he didn't know, was copies of the "Gears of War 2" video game pre-ordered through Blockbuster, GameStop and other select retailers included a free exclusive variant cover version of the "Gears of War" #1 comic book. As such, to say that the "Gears of War" #1 copy sold 450,000 is a little misleading as it implies people were purchasing the comic as a independent item when they got it as a bonus when they pre-ordered the game whether they wanted the comic or not. But, from the publisher's perspective, those units were still paid for so they count a sale. In essence, this is similar to copies sitting unread on store shelves. Since the retailer bought those copies from the publisher, from the publisher's perspective they count as sales.
Regardless of how many people bought the game and got the comic or how many specifically pre-ordered the game in order to get the comic, "Gears of War" #1 was a huge success for DC and Wildstorm in terms of reaching out to a potentially new audience.
Just for comparison, 450,000 units exceed the combined estimated sales through Diamond of "Final Crisis" #1, "Secret Invasion" #1 and the Director's Cut of each. If only a single percent of the people who read the copy of "Gears of War #1 go to the comic book stores and buy the second issue, the sales would easily double even if those were the only sales of the second issue.
The direct market sales data for "Gears of War" over the coming months will give us an idea how many of the people that got a copy of "Gears of War" #1 along with the video game they pre-ordered at the video game store liked the comic enough to go into a comic book store looking for the following issues.