most recent movie sales are the most predictive and then other factors get put on top like cultural events, weather, whether or not advertising took place (and was efective). So, on that topic, I'm not in the older female demographic but would be interested to know how much impact advertising for the movie might have had on that demographic. Is there a publicly available online resource for measuring how effective different kinds of advertising strategies may have been with different demographics? I would be interested to factor that in too. Stuff like social media mentions I could probably measure myself or with analytics tools/websites I'm aware of. The other strategies... umm... they do sound kind of specious and I'm surprised to hear people would give them any credibility lol. Looking at parking lot car count might sound like a good idea but I'd say it's specious to conclude anything about a particular movie from that. Sampling a demographic like the guy at the theater sounds a little more informed but you'd probably need to do a lot more sampling than just at one theater or one time of day, I dunno.. I like the MIT approach because I surmise they based it off a feature set (e.g. previous director OW sales, weather, etc) they thought would have good predictive ability. Thanks for the info, good discussion points.