THE TRUTH BEHIND FILM FESTIVALS
This is the first of three articles about Film Festivals. In part 2 I will publish a survey I have conducted with film festival directors around the world, showing the data on their point of view. Part 3 will give a voice to the human stories of the festival directors by sharing stories and advice for filmmakers. This article focuses on the data behind film festivals around the world.
At the bottom of this article I have gone into detail on the methodology I used to uncover these results. I also list numerous ways in which I regard this study to be lacking, meaning that numbers should be assumed to be around 90%-95% accurate. The short answer is that I found 9,706 unique festivals which have run at least once the past 15 years, with data from twelve different sources. The main results are:
- There are around 3,000 film festivals currently active (i.e. ran in the past two years)
- 9,706 film festivals have run at least once in the last 15 years
- 75% of all film festivals were created in the last ten years
- 2009 was the peak year for new festivals
- 2012 had the lowest number of new festivals launched for 14 years
- 39% of film festivals only ever run once
- 71% of film festivals screen short films and 52% screen feature films
- Half of all film festivals run for less than 7 days
- North America hosts 70% of the world’s film festivals
- October is the busiest month, with five times as many festivals as December
I’ve collected data on almost 10,000 film festivals throughout the world, but many of these festivals have either closed or are taking a time-out from running events. I would say that to be fairly regarded as ‘active’, a festival will need to have run events in the past two years. By this standard, there are 2,954 active films festivals.
While it can be debated whether or not film is a ‘quintessentially American’ art-form, it is certainly true that film is an American business as nearly three quarters of all film festivals are based in North America.
The data shows that film festivals are often run once, and then never again. A third of film festivals only lasted a single year, with under a quarter making it past six years old.
Half of all the film festivals I could find were created since 2007, with the peak in 2009. Since then there have been far fewer new festivals launched. 2012 was the worst year for new festivals since 1999.
The uptick in film festivals in October might be due to a focus on horror festivals around Halloween. Sadly I was not able to find enough data about festival genres to test this theory.
I wasn’t surprised to learn that nearly 4 out of every 5 film festivals screen short films, but I wasn’t expecting the result that only half have any form of competition. It’s actually nice to see that the image of filmmakers entering festival purely to win prizes is not backed up by the data.
There was a huge spread in the number of days film festivals held events. Part of the reason for the large number of festivals claiming to run “over 30 days” might be due to year-long celebrations and ‘always on-going’ festivals.
I used a number of data sources for this research including publicly available information on sites like
Withoutabox (by far and away the largest credible collection of film festivals online),
Short Film Depot,
Festhome,
Reelport,
Festival Focus,
Short Film Central,
BestInFest,
FilmFestivals.com,
Wikipedia,
IMDb, the
British Council and
good ‘ol Google.
The data used to calculate the ‘Months of the Year’ came from a smaller sample of just over 1,500 festivals which I could personally verify. This is because the original data showed a disproportionate number of ‘December’ opening nights which did not seem credible. It seemed much more likely that the list was polluted by the quirks of some of my data sources, such as requiring year-round festivals to put a specific date which led many people to put the 31st December. I also suspect that some sites use that date as a default, further skewing the results.
I have been unable to provide a completely accurate and definitive figure on the total number of active film festivals in the world today. My data combines a number of data sources (see ‘Methodology’ for the list) and no one place was perfect. Below are the claims made by various sites. I found them to be reasonably accurate, although all had a certain amount of duplication and erroneous data.
Withoutabox say they have “more than 5,000 festivals” although they don’t qualify this by saying how active these festivals are. The number of festivals you can apply to via Withoutabox is either 850 or 900, depending on which page of their site you read. I ask them for clarification but they have not got back to me in the time it’s taken me to do this survey.- FilmFestivals.com use different figures on their site, including 4,000, 5,000 and 6,000 festivals
- Festival Focus list 2,303 festivals
- BestInFest list 1,832 festivals
- British Council list 1,582 festivals
- Yahoo list 957 festivals
- Wikipedia list 437 festivals
I believe this to be a very comprehensive study by the standards of the world of filmmaking, but doubt I will be receiving any honorary doctorates in Asymptotic Statistical Analysis because of it. There are a number of areas I would have liked to have had more, or cleaner, data. For example…
- I’m sure I missed some festivals and that others could not be counted (being false, not true film festivals, etc). That said, I don’t feel that the margin for error would be high, so the overall trends should be fairly reliable.
- There is no way of verifying if a festival actually took place, so it’s conceivable that some of my ‘first year’ festivals didn’t actually happen.
- I speak English and so will naturally have a bias away from festivals which only provide information in a non-English language.
- My data gathering was largely online, thereby ignoring festivals whose main communication is via offline communities (such as schools, churches and local groups).
- I am assuming that the information presented to the public is factually accurate. There have been a few times where I have manually excluded data that is demonstrably false (festivals claiming to have been going for almost 100 years despite film only being 125 old).
- It’s possible some festivals may have been duplicated, although I have cross-referenced names, contact details and websites. It’s conceivable that a festival changed its name and principle points of contact, in which case it would be counted as two festivals here.
