If you watched CNN or MSNBC last Monday, or visited any number of Web sites that carry box-office data, you probably already know that The Chronicles of Narnia finally managed to edge past King Kong over the New Year's weekend, $32.8 million to $31.6 million.
But as Mickey Kaus recently observed, sites like Variety.com and BoxOfficeMojo.com, as well as wire services like AP and Reuters, seem to be engaged in some crystal-ball-gazing when they announce "estimates" for weekend grosses before the weekend is over. What are these numbers, and how are they obtained?
These box-office "results" released over the weekend are simply a studio's own estimate of its movie's weekend performance. Distribution executives arise at dawn on Sunday mornings to crunch their numbers and report them to the media. Until recently, I was the Daily Variety reporter who collected box-office grosses; at 8 a.m., my phone started ringing with executives handing over the numbers, along with a healthy dose of spin. Making a weekend projection on a Sunday morning is quite similar to how the media call political elections when they have the results of only a handful of precincts: You compare the numbers you have against some past results to make an educated guess.
The numbers the studios have, but don't release to the public, come from one of two box-office tracking firms, Rentrak and Nielsen EDI's Flash service. These services get sales results directly from theaters and place them on a secure Web site; only a select number of studio executives have access to this enormous information database. Here, for instance, they can compare King Kong's matinee grosses at AMC's Times Square theater with what The Lord of the Rings: The Return of the King did there its first afternoon.
Every studio has its own intricate formulas for divining weekend estimates. Here's a simplified, hypothetical example: Let's say you're 20th Century Fox and you're trying to figure out the early estimate for the opening weekend for Star Wars: Episode III—Revenge of the Sith. The last two Star Wars movies were released on the same weekends in previous years—the week before Memorial Day. On the first film, The Phantom Menace, Sunday grosses were down only 10 percent from Saturday. On the second, Attack of the Clones, the Saturday-to-Sunday drop was 22 percent. If you're coming up with a Sith estimate, you would by Sunday morning have a good handle on Saturday grosses, so all you have to do is guess how much Sunday will drop. If you split the difference on the last two films, you end up with 16 percent, which is basically how Sith performed—Fox issued a Sunday estimate of $158.4 million for Sith's first four days.
Once the studios see the box-office grosses from all the Sunday shows, they issue revised, more informed estimates to the media on Monday, which are known as the "actuals." Fox's "actual" box-office report for Revenge of the Sith was not far off-base: $158.5 million.
Since Rentrak and EDI are collecting actual box-office numbers from the theaters, why go through the trouble of making estimates at all? Why not just release the real data? The problem, as the studios see it, is that the private numbers only reflect grosses from approximately 90 percent of the theaters in North America (in Hollywood, North America is only the United States and Canada). The "actual" estimates the studios issue include a guess at how much business they're doing in those other theaters. This "missing factor" affords studios some wiggle room to secure a box-office record, or to say that the sequel beat the opening of the original. Generally such tweaking remains a Hollywood secret, though; executives tempted to complain know too well that one day they may want to do the same thing.
As was the case with Revenge of the Sith, studios' weekend estimates are typically closely aligned with actuals released on Monday. But there have been situations where weekend projections are wildly off course. Maybe a bad storm hurt business on Sunday night. Or the studio underestimated the millions watching a big World Series game or the Oscars. In 2004, for example, Christmas Eve, typically a low-grossing day at theaters, fell on a Friday, which is ordinarily a high-grossing day. This wreaked havoc on the studio estimates. Universal's estimate of $44.7 million for Meet the Fockers was $2.5 million too low, while Fox's $12.7 million estimate for Fat Albert was $2.7 million too high. While Fockers maintained its first place box-office position, the optimistic estimate for Fat Albert meant it slipped into third place behind Lemony Snicket's A Series of Unfortunate Events.
Though rare, sometimes studios jockey for position on the box-office chart. The last great Sunday-morning battle was in the summer of 2002 when 20th Century Fox's Minority Report opened against Walt Disney's Lilo & Stitch. Looking at the same data, both studios were sure they had cinched the top spot. In the end, Fox eventually claimed the day, when their actuals showed that Minority Report's $35.7 million opening had squeezed past Lilo's $35.3 million.
Although there is always room for chicanery, there are two factors keeping studios honest. First, some Hollywood players now have movie contracts that tie payments to how well their film performs at the box office. Frequently, the contract specifies that the benchmark will be the studio estimate printed in Variety, so it's in the studios' best interest to not overreport.