The quicksand data come from Version 28 of Crypto's List, published in December 2007. (Numbers for the 00's decade were prorated to account for the missing years 2008 and 2009; only full-length movies were included.) Film industry statistics are taken from the MPAA for each year after 1945, with the remaining information filled in using estimates of production from various histories of American cinema.

Several caveats are worth considering. Crypto's List does include foreign films, which aren't reflected in the overall industry numbers. As a result, the percentages on the Y-axis will be skewed upward by a small amount. That said, if film production around the world were correlated with production in Hollywood, the shape of the graph should remain more or less the same. In any case, it's very unlikely that more accurate, global industry numbers would affect the most general finding presented in the graph—i.e., a sudden spike in quicksand cinema during the 1960s, followed by a more gradual decline.

It's also possible that the spike reflects a selection bias on the part of the quicksand community: Since so many of the people who made the list grew up during the 1960s, they might be more inclined to include films from that era. But the same bias seems just as likely to work in the opposite direction: If there were indeed more quicksand films in the 1960s, that might explain why so many quicksand fans are now in their late 40s and early 50s. Also, you might expect that more recent quicksand scenes would be easier to identify and catalog, given how much material has now been digitized and posted online. If that's the case, the numbers at the far right of the graph—in the 1990s and 2000s—would be artificially high.

The histogram below shows the same pattern using different data. Feature films tagged with the keyword quicksand in the Internet Movie Database are plotted by decade. It may be that these keywords come from the same quicksand fans who contribute to Crypto's List, in which case the same selection bias would apply to both datasets. But if that's true, it's not clear why the quicksand fans would have stopped tagging in IMDB after just 72 films.