Former Leaguer King Suckerman, (aka Alex Segura, whose debut detective novel, Silent City, you can buy here) recently tipped us off to an incredibly interesting infographic on spec script coverage.
About a week and a half ago, a Reddit user named profound_whatever analyzed and compiled comparative data on 300 screenplays s/he had covered as a reader for five different film companies. (Readers are the first point of entry to getting your script repped or sold. They are the people - sometimes in-house, like interns, which I did for two companies during college; sometimes hired externally and paid on a script by script basis - who read most if not all incoming material for a production company, agency, or studio. They are then tasked with writing coverage, which consists of a brief synopsis of the script, followed by an evaluation of its strengths and weaknesses. They are then asked whether to recommend, consider, or pass on a script - and let me tell you, they are encouraged not to recommend, as each recommendation could lead to many staggering expenses, including buying a script and making a movie, which could lose the company a lot of money. The whole point is to get the muck out of the way, so that the higher ups can be spared the trash and focus solely on promising material. In short, you want a reader to like your script.)
The reader on Reddit assembled stats on the 300 screenplays to provide a broad, and sometimes quite detailed, snapshot of the kind of material assigned to him/her. Of the 300 scripts, they averaged 107.22 pages (which should give you an idea of where your script should land), but ranged from 79 to 147. As a former reader, I can tell you that the shorter script will always be read first, though too short a script is an immediate red flag. The most popular genre was horror/slasher, with 49 script. More than 1 out of 7 scripts (43) was set in New York City. The scripts were male dominated (270 male writers and 137 scripts that featured both a male protagonist and antagonist).
Perhaps most importantly, of the 300 script, the reader recommended only 8. Yes, 8 out of 300, or just under 3% of script were considered worthy of recommending outright. The reader "considered" (or considered with reservations, which means s/he didn't want to be the one to say no, when it might be worth looking at a bit further) a further 89 scripts. That's a good stat, but most considers get a second read by someone else, and then a pass. A total of 203 (more than 66%) of scripts received a straight up pass. Someone higher up at the company will read coverage on a pass and might read the script if it sounds interesting enough, but generally, a pass is where a script's life ends at that company. (You should also be aware that coverage is archived, so if you rewrite your script and resubmit it, the company will go through the files, see that they have a script of the same name by the same writer that they already passed on, and they won't read you again. A pass, friends, is a pass for good.)