Sure they can. If you look not too far back in time, you'll find a lot of dotcoms that claimed one "best year" after the next, right up until their liquidation in bankruptcy court. They found all kinds of ways to measure performance (number of free user accounts, number of ad click-throughs, total dollar value of products sold at a loss as a "temporary" effort to build market share, number of underpants stolen by the gnomes, etc.) that, in truth, had little bearing on the ultimate success of their business model. In some cases, the "better" they did, the more money they were losing.
Someone quotes Charles as saying there are now (according to their polling studies) 4.6 million active D&D players, the highest in all the time they've done their polling. On that basis alone, he would be justified in calling it the best year ever for D&D. In the odd chance that someone called him on it in a lawsuit (very unlikely, given the lack of any specific numbers and the round-off-error level of significance that D&D sales have in the big picture of Hasbro finances -- they don't even get broken out, as far as I've seen, in the 10Qs and 10Ks on file with the SEC), he's covered. After all, his audience for these remarks is not an investor conference, but a congregation of fans -- and to the fan the important assurance is that there are people to play the game with, so this measure of "best" is entirely appropriate, regardless of whatever stats and figures may lurk behind the corporate curtain.
Investing in Hasbro because D&D has had its "best year ever" would be like investing in my company because I got a bargain price on my plane flight to Gen Con this summer. Nice and all, but not really material to the bottom line.
Having said that, I know no specifics about worldwide sales of D&D, so I don't know how broadly Charles' claims should be interpreted...it just seems that they could be interpreted in multiple ways, without accusing him of being deceptive or dishonest.
---
As an aside: Some stores that send data to C&GR are bookstores, and some are online retailers. This doesn't prove that the sample is representative of the industry at large or any segment thereof. C&GR gets cited simply because it's numerical data that is available in public to be seen and discussed.
There are other sources of data, but anything that comes from the private realm couldn't be independently verified even if someone felt comfortable in sharing it.