Monday, July 11, 2011

Looking Back at Previous Harry Potter Films

Putting together a chart below. Please let me know if you would like see comparisons on any other data that we can slice.

The Month Opened inNovemberNovemberJuneNovemberJulyJulyNovember
The Day of the Week Opened onFridayFridayFridayFridayWednesdayWednesdayFriday
Theater Count3,6723,6823,8553,8584,2854,2754,125
Screen Count8,1198,5158,900?9,000+near 10,0009,400
Midnight GrossN/AN/A~56.31222.224
Opening Day w/o Midnight32.3329.6333.2733.8232.2335.9837.68
Opening Day32.3329.6338.2740.1244.2358.1861.68
Fri-to-Sat Change+3.6%+15.5%-17.7%-8.5%+9.6%+9.0%-38.0%
Fri-to-Sat Change w/o Midnight+3.6%+15.5%-5.3%+8.6%+9.6%+9.0%+1.4%
Sat-to-Sun Change-27.0%-28.4%-24.0%-29.6%-18.9%-25.5%-34.3%
Internal Multiplier2.792.982.452.562.982.902.03
Internal Multiplier w/o Midnight2.792.982.672.852.982.902.68
Multiplier at the End of 4th Weekend2.652.422.262.382.572.492.06
Overall Multiplier3.522.972.662.822.982.852.36
Overall Gross317.58261.99249.54290.01292.00301.96295.00

Several points immediately coming to mind when looking at the comparison:

1. The screen count had been trending up until the last release. With this being the last film of the franchise, expect it to beat the 9,400 figure, but probably won't exceed 10,000 considering several other films still command good number of screens. The market Half-Blood Prince faced was emptier.

2. The midnight gross has exploded, and this one should certainly continue the trend and most likely exceed The Twilight Saga: Eclipse' current record of just over $30m.

3. Despite the growth of midnight showings, the opening day without midnight has in fact remained relatively stable, so don't expect that portion to grow proportionally as the midnight will. In the last several Harry Potter films, pretty much ever since Prisoner of Azkaban, we have seen the tendency of jumping the gun on the optimistic side just because the midnight or opening day got bigger.

4. Another indication of the Harry Potter fanbase having not grown much at all over time is that the Sunday number has been stable since the first film. It means once the upfront demand burned off, each film basically returned to a similar trajectory, varied only by the date of the year, not the reception of the film itself. The end result is that none of the latter films exceeded the first one in raw total, let alone in admissions; instead, the franchise has simply become more frontloaded toward the first couple of days of the release. It should change with Harry Potter 7.2 due to the finale factor and some help from 3D/IMAX, but we should keep this in mind when following all the reports this week. Nonetheless, it is an exciting event to track. :)


  1. Great chart and analysis.

    I don't know if this is a typo or not but shouldn't the Fri-Sat change for HP5 and HP6 be -36.0% and -49.8% respectively?

  2. They opened on Wednesday, so that'd be Wed-to-Thurs change. I know it could be a little confusing, but decided to use the actual Fr-to-Sat change for them as well to keep the chart consistent.

  3. And for those two, all the "... w/o midnight" numbers outside the opening day are the same as the ones with because the midnight happened a couple of days before the weekend started.

  4. Great chart, Xia. The consistency of the films' internal weekend performances just astounds me. This is one of those franchises we'll be studying for years to come, I think. There's basically nothing quite like it.

  5. Interesting. A number that you don't have in that table is the overall multiplier without midnight. It's only really different from the overall multiplier for DH1, since OotP and HBP opened on Wednesdays, so their Wednesdays weren't part of their opening weekends anyway, and midnights were small enough for PoA and GoF that it doesn't change much. This number appears to be close to the internal multiplier without midnights for all of the movies except the first. But in the case of DH1, midnights represented nearly 20% of the opening weekend, so removing them from the calculation of the overall multiplier brings that number up significantly, to about 2.68, and closer to being in line with the rest of the movies.

    It also reveals the interesting pattern that the overall multiplier without midnights has been either close to or above the internal multiplier without midnights for all the movies. Only HBP is really lower and even that would round to the same number if we looked at only one place after the decimal point.

    Using the $43.5M number for DH2's midnights (assuming that didn't decrease when the estimate for opening day decreased), DH2's internal multiplier without midnights would be 2.64. So, if its overall multiplier without midnights is close to that, then we're looking at a finish of about $375M.

  6. (Left a sentence in that previous comment that I meant to take out. Please ignore the next to last sentence of the first paragraph.)

  7. Hey Xia, do you know remember you got that POA midnight figure from?

  8. Think it's from the WOKJ forum, but I'll need to confirm.