INTRODUCTION:

The database below is assembling and offering results of my own „cinemetric” measurements of films and television episodes. I achieved those measurings by counting film shots by hand-clicker, by processing the number of shots in special software or by making a shot-by-shot analysis. To calculate the average shot length (i. e. ASL, a concept created by Barry Salt), I measure the length of the film in seconds (24fps in case of sound films) and then divide it by the number of counted shots.

In my database, it is possible to find more than 2950 measurings of more or less contemporary films from many countries. Besides the fact that you can find here hundreds of ASL measurings of movies from the last decades, the database is fairly unique from at least three other reasons. Firstly, it includes an almost entire number of preserved films of Czech silent cinema and at least 33 % of Czech sound cinema production from each year till (so far) 1933. Secondly, it covers a large number of films from Eastern European countries. Thirdly, it consists of a lot of entirely rare silent films from the 1910s all around the world. Since I am updating the database a few times a month, it is going to continue growing larger.

It is needed to point out that very ASL values themselves are not the final results of any research. Those values are meant to be used as tools for further analysis. For what kinds of research may ASL be used? In the case of a particular film, the very ASL value will help us to get a rudimentary awareness of its editing pattern, but it will not tell much more. When analyzing a particular movie, it is therefore much more useful to observe the ASL’s variations during its course (for example, the ASL of individual scenes, sequences, or acts). On the other hand, by the force of ASL values or shot counts, we can gain a lot of various and inspiring suggestions for stylistic research in case of groups of films. An acquaintance of ASL values can facilitate the juxtaposition of various film or television works of a period, region, genre, company’s or director’s production. The mere fact that certain films of a particular director (or period, region, company, genre, etc.) have identical or similar ASL values may not be a priori remarkable. That is because identical or similar ASL value can be caused just by a significantly different configuration of longer or shorter shots. However, it is worthwhile to consider the similarities in the case of rather high and rather low ASL values – assessed against the background of the relevant norms. Nevertheless, more remarkable and predictable is when we find out that certain films of a particular director (or period, region, company, genre, etc.) have significantly different ASL than the others. These different ASL values imply a notable variability of creative choices that filmmakers made. Since that we can put forward many research questions. I do believe that my database offers extensive as well as reliable background for such questions.

Update (July 2020): I have added two entirely new parameters to the database, which will offer slightly different quantitative perspectives and comparative options than the existing ones. The first new parameter is the average number of cuts per minute of film (CPM), which may be easier for some users to imagine than the average shot length in seconds. The second, in my opinion, a much more important new parameter is the number of cuts per hour of film (CPH), which allows the detection of long-term trends and massive comparability. I work with this parameter in two forms in my database: optimally (a) in the form of the exact number of cuts per the first hour of the measured film; and (b) in case I do not know this number, in the proportional number of cuts per hour of the film. The exact number of cuts per hour is noted in the database with precision per unit of shots and is based on my own measurement (at a projection speed of 24fps). The proportional number of cuts per hour is rounded down to „fifties“ and „hundreds“. I’ve done hundreds of control comparisons of this calculation on the movies I also knew the exact number of cuts per hour, and it’s relatively accurate in terms of broader trends. For the time being, I have decided to state the number of cuts per hour only for sound films, because we can speak with certainty about the standardization of the projection speed in their case.

Radomír Douglas Kokeš
Department of Film Studies and Audiovisual Culture
Masaryk University
Brno, Czech Republic

MY PARALLEL PROJECT:
DATABASE OF MOBILE FRAMING STATISTICS (1929-40) / www.douglaskokes.cz/pdz/mfs/

Douglas‘ Film Database of Shot Counts and Average Lengths (ASL)
Douglasova filmová databáze počtů a průměrných délek záběrů (PDZ)

28th September 2021
Number of items: 2975

The Last Measurings: Shoes (1916), Sur un air de Charleston (1927), Daïnah la métisse (1931), Yeyo pravo (1931), Život je pes (1934), Death at Broadcasting House (1934), Gjest Baardsen (1939), Tarzan’s Magic Fountain (1949), To Hell and Back (1955), Old Shatterhand (1964), Rękopis znaleziony w Saragossie (1965), Sól ziemi czarnej (1969), The Poseidon Adventure (1972), The Mechanic (1972), Papillon (1973), Superman (1978), Moonraker (1979), Le Marginal (1983), Dune (1984), Red Dawn (1984), Robin Hood: Prince of Thieves (1991), Terminator 2: Judgment Day (1991), The Last of the Mohicans (1992), Rising Sun (1993), Signs (2002), Just Like Heaven (2005), 2021 (2009), The Two Popes (2019), Last and First Men (2020), Zátopek (2021), The Tomorrow War (2021), Blood Red Sky (2021), Old (2021), Kate (2021), The Film Journey of Professor Burnett (2021), Black Widow (2021), Shang-Chi and the Legend of the Ten Rings (2021) + (regularly) Columbo (1969-2003).