http://www.dakkadakka.com/dakkaforum/posts/list/7.page JForum - http://www.jforum.net
Spinning this off from discussion on the AoS thread before we get Red Text Shouted At. And as per the title, it’s on why we can’t treat ICV2 as data, but as an anecdote.

Ostensibly, ICV2 is a snapshot into what’s hot and what’s not in the gaming world. And it offers a top ten ranking of the best selling games, RPGs etc. In the past, it’s been used by folk to declare the demise of GW etc, particularly when it reported X-Wing as outselling 40K.

Except…the report itself has serious flaws. First, the reports and data it relies upon are entirely voluntary, and only cover FLGS in North America. Secondly, it doesn’t appear to have any way of verifying data provided to it. Thirdly? GW are one of the relatively few wargames companies that sells direct to the public.

Combined? We can ably demonstrate that at most, it’s only accounting for a minority of GW’s overall sales, and we don’t know what percentage of North American FLGS actually provide data.

Let’s look at GW own financials, which are publicly published, and independently verified. For 2024/2025, it had a core income (so not including licensing) of £565m. And its Trade (so FLGS and Distributors) account for £351.7m of that - or 61%.

Unfortunately the report no longer offers a breakdown by territory, so the next bit is more illustrative/speculative. But we do know that North America is GW’s largest single market. And because GW doesn’t have the same store saturation, a higher percentage of its North American Sales are going to fall under Trade.

For arguments sake, let’s assume that of GW’s total global Trade sales, North America accounts for 60%. That would mean, if every trade account reported to ICV2? At best, it’s reporting 36% of GW’s total sales. Which is of course a minority.

From there, we dig deeper. Its rankings don’t seem to show the underlying percentages. So when it comes to first and second places? We don’t know the margin of victory. Could be just a couple of bucks, could be tens or hundreds of thousands, maybe millions at the extreme. Nor do we know what percentage of FLGS contributed data, or whether they all carry the same lines. Both of which can skew the data.

It’s also likely, given not every company selling miniature wargames sell direct? Any data provided to ICV2 represents a naturally higher percentage of sales than those for GW, as their own stores and online don’t partake.

And if we wanted to be Proper Suspicious? We can’t even rule out that some data contributed could be entirely made up. Doesn’t seem much point in doing that though. Other than “people do weird things for weird reasons sometimes”. Which in a thread criticising data analysis I’m going to suggest we don’t consider further, because any old claim can be made on that same basis, for the same reason.

Silly last point aside? We’re essentially left with useless data set with no real context to it. It’s not exactly “trust me bro”, as there’s at least a wide range being reported. But it’s still a report based on inherently limited sales data, and data which isn’t itself verified by those compiling it. As such, it’s no more use than an anecdote, and can’t be used to come to any sort of informed conclusion.]]> http://www.dakkadakka.com/dakkaforum/posts/preList/818635/11811694.page http://www.dakkadakka.com/dakkaforum/posts/preList/818635/11811694.page GMT