These limits are largely mechanical. Shorts are limited by their need to be shown in advance of a feature or screened in program of five or so other shorts. As their name might suggest, shorts are too short to be shown by themselves. Even worse, longer short can only be realistically screened as part of a program of other, shorter shorts... since what audience will sit through a 40-minute short as the lead in to a standard 90-minute film.

Lastly, programmers tend not to allow a short to dominate any shorts program. This leave little room in the typical festival schedule for longer, featurette-length shorts (25-40 minutes). For instance, this year, Sundance programmed only four shorts longer than 25 minutes ("Buyo," "History of America," "Kids + Money," and "La Corona") out of a total program of 85 shorts.

For feature-length films, filmgoers expect at least 75 minutes of film to justify the price of admission, so shorter features tend to be underrepresented at festivals. Overly long films tend to be challenging for programmers since the slotting of long films over 130 minutes typically means the use of two standard festival time slots to show one film, which is another way of saying that longer films displace other films that could have been shown in the second time slot. These long films are hard on audiences as well. As we can all attest, an audience tends to become restless when a movie continues beyond the two-hour mark.

There are exceptions to these rules, such as Lav Diaz's epic 540 minute "Death in the Land of Encantos," which showed in 2007 at the Toronto Film Festival; but they are just that—exceptions. In a sample of over 4,500 films showing in 33 major festival in 2007, only 7% of short shown were between 25-40 minutes; only 5% of features were longer than 130 minutes; and only 11% of features shown were shorter than 75 minutes.

What is interesting for the shorts category is that longer shorts tend, paradoxically, to be more award-winning than shorter ones. Though these films are harder to place at festivals, once they get screened their substantive nature, in part the result of their length, tends to make them more attractive as winners. Their quality and impact also tend to be high in order to have overcome the normal programming resistance to such odd festival length. The tipping point seems to be at around the 8-minute mark, and the winning effect of longer shorts grows as the short become longer. Films less than 8 minutes long win little if anything, although the majority of shorts on the festival circuit run at those lengths.

This phenomenon was very true of the most recent documentary film "Freeheld" that I produced, which in 2007 surprised us by getting into Sundance at 38 minutes and went on to take the Special Jury Prize. It subsequently went on post-Sundance to win 14 awards and was nominated for an Academy Award--which it actually won (Best Documentary Short Subject)!

Longer films are harder to place at festivals, but if they can get play on the festival circuit they tend to be award-winners.

The balance between optimal slotting and award-winning potential seems to be around 15-18 minutes. That length is not too long to be unscreenable while still maximizing the chances of taking home an award. The optimal length for features to win an award seems to be around 85 minutes, which suggests that festival juries and audiences tend to reward tighter editing more than epic storytelling. Unlike the case of shorts, that award-winning length happens to also be a good length for getting on a festival schedule.

Cutting a film optimally to play on the festival circuit does have its downside, since what works at a festival does not always work for the distribution outside the festival circuit. For instance, Oscar-nominated short documentaries tend to be at least 35 minutes, which makes them hard to place at festivals, and the Academy of Motion Picture Arts will disqualify your film for consideration if you have shorter or longer festival version.

Television distributors tend to like films that can be played in television's hour-long slots with time for commercials, which favors a problematic approximately 50-minute runtime. Cable outlets such as HBO and International TV networks can be more flexible, but still most television deals are at lengths not well suited for festival play. It is therefore wise to consider the needs of the film's full lifetime when deciding on a final runtime, and not let only the needs of festivals dominate your creative decision-making.

I will end this article with some words of caution. Runtimes are not the only determinants of success in getting films into festivals and doing well there. You need to also consider the film's craft, content, filmmakers, cast, and previous festival track record. That said, runtimes are an easily controlled variable to ensure the best festival run possible for your work.

The competition for festival screening is great, especially for the top tier of festivals where a couple of hundred films are screened each year and thousands of filmmakers apply each year for the honor. In this hyper-competitive environment, anything a filmmaker can do to improve the odds should be welcomed.

To the creative types who might resist this advice, I would like to note that the rigor of cutting to a marketing-defined target length tends to improve many independent films, which too often are cut too long. Just my opinion.