I’m endlessly fascinated with the psychology of moviegoers, and what it is that makes their minds tick. Why do certain bad movies resonate, while some very good movies do not? There’s not an enormous amount of academic literature devoted to the subject (at least to my knowledge), so we’re mostly left with assumptions formed on the basis of box-office results, and on critical and audience aggregation services like Metafilter, RottenTomatoes, and IMDB. There is no Moneyball for movies: There are formulas, obviously, but for everyone that works, another formulaic movie fails for the exact same reason: Because people are tired of the same old thing. Unless they aren’t. What makes moviegoers embrace formula in some instances and reject it in others? Why do dark unsatisfying endings in films like The Dark Knight resonate, while they are rejected in movies like The Grey?

There’s not a lot of openly available data to subject — we are not privy to the results of focus groups, nor do we know if changes suggested by focus groups hurt a film’s success as much as they help it. Film studios are businesses, and businesses value profit over creativity, so film studios tend to rely on what works. They make choices based on past performance; they greenlight movies in familiar genres, with familiar storylines, with familiar actors because they don’t know any better. If one comic book movies succeeds, they will make 100, or at least as many as it takes to realize that audiences have grown weary of them. Successful anomalies beget new trends, which are pulverized until a new anomaly surfaces. The success of Iron Man beget comic-book movies based on lesser known characters; Liam Neeson’s Taken gave studios the bright idea that decent action flicks might work in January; and Tim Burton’s Alice in Wonderland spawned the fairy-tale movie trend that we’re currently in the midst of.

What we don’t understand is why these anomalies work or don’t work. Psychologically, we don’t understand why certain movies — like Office Space or Donnie Darko or Swingers or Napoleon Dynamite — do not resonate with audiences until months or even years after their initial release, but then — almost overnight — become part of our cultural lexicon.

It’s a funny business.

Beyond box-office numbers and critical scores, and the assumptions we can derive from them, I am particularly fascinated with Cinemascore, a box-office evaluation service that started in Las Vegas in 1999. The company surveys 400 to 500 people in 25 cities around the country on the night a movie is released and makes predictions about box-office based upon letter grades assigned by moviegoers. Those predictions are often very accurate. For instance, Cinemascore predicted the box office success of The Hangover (nearly to the exact million), as well as the failure of Land of Lost. Movies are given a grade score, and the vast majority of films fall within the curve: B+ to A-. A pluses are rare (there have been less than 20, including The Help, Tangled, King’s Speech, The Blind Side, Remember the Titans, A Few Good Men, Titanic, Antoine Fisher, Toy Story 2 and Driving Miss Daisy), but not as rare as an F (The Box, Solaris, Darkness, Bug and recently, The Devil Inside). Grades are not always indicative of opening weekend (see The Devil Inside’s $30 million), but they are very good in predicting box-office multipliers (see The Devil Inside’s $52 million overall gross, far less than double its opening weekend).

Anything below a B is considered a bad score. A C is the kiss of death, box-office wise.

However, Cinemascore’s Grades are very hard to come by. They’re easier to find with more recent films, but it took a hours of Google Searches (and the Wayback Machine) to find scores for several older films. I have managed to collect the grades, however, for 15 good to great films that were nevertheless received poorly by audiences, and in most cases, the box office results reflected the poor Cinemascores.

The question is: Why did audiences reject these movies? What can we extrapolate from this information? And it has to be something more than simply: Audiences are dumb because “dumb” audiences are nearly as likely to embrace a well-reviewed film as they are a poorly reviewed one. What is it about these critically embraced films that audiences disliked?

Drive — Cinemascore (C-), RottenTomatoes (93 percent), Box Office ($35 million).

Haywire — Cinemascore (D+), RottenTomatoes (80 percent), Box Office ($16 million, after three weeks)

Let Me In — Cinemascore (C+), RottenTomatoes (89 percent), Box Office ($12 million)

Hanna — Cinemascore (C+), RottenTomatoes (71 percent), Box Office ($40 million)

Shutter Island — Cinemascore (C+), RottenTomatoes (69 percent), Box Office ($128 million)

Children of Men — Cinemascore (B-), RottenTomatoes (93 percent), Box Office ($35 million)

High Fidelity — Cinemascore (C-), RottenTomatoes (92 percent), Box Office ($27 million)

Wonder Boys — Cinemascore (C+), RottenTomatoes (81 percent), Box Office ($19 million)

The Royal Tenenbaums — Cinemascore (C-), RottenTomatoes (80 percent), Box Office ($52 million)

Confessions of a Dangerous Mind — Cinemascore (C), RottenTomatoes (79 pecent), Box Office ($16 million)

Contagion — Cinemascore (B-), RottenTomatoes (84 percent), Box Office ($75 million)

Splice — Cinemascore (D), Rottentomatoes (74 percent), Box Office ($17 million)

The Grey — Cinemascoe (B-), RottenTomatoes (77 percent), Box Office ($35 million after two weeks)

The American — Cinemascore (D-), Rottentomatoes (66 percent), Box office ($35 million)

Boogie Nights — Cinemascore (C), Rottentomatoes (92 percent), Box Office ($26 million)

——

What ties all of these poorly graded, highly reviewed movies together? Ambiguous, unhappy, or unsatisfying conclusions. What implication can we draw? That, formula or no formula, audiences overall really do love a Hollywood ending (exemplified again in this weekend’s Chronicle, which deserved much better than the mediocre B Cinemascore it received). It does not, however, explain the relative box-office success of Shutter Island and Contagion, both of which had bleak endings, nor the high Cinemascores of movies like The Dark Knight (A-) or Inception (B+), which also had dark or ambiguous endings.

What is at play here? Maybe it’s simply that behavioral patterns are difficult to nail down. Maybe that’s why there’s been so little academic study done on the mind of the moviegoer: We’re too goddamn unpredictable to tick off in a box.

Dustin is the founder and co-owner of Pajiba. You may email him here, follow him on Twitter, or listen to his weekly TV podcast, Podjiba.