Save the Cat

I don’t think I’ve talked about Save the Cat here yet, and how I used it to structure a book. This isn’t a “learn to write so you can make millions like me” blog, so I don’t know how important or useful it is for me to document this. And spoiler alert, the book I wrote using this method did not sell a million copies. But as I’m thinking about book ideas now, I keep coming back to this. So here goes.

Save the Cat! is book by screenwriter Blake Snyder, which describes his method of structuring and outlining a screenplay for maximum impact. It’s essentially a refinement or maybe simplification of the Syd Field “paradigm” or three-act structure, mixed with a healthy dose of the Joseph Campbell hero’s journey/monomyth thing, which has been beaten to death by any number of screenwriting gurus/hacks such as Christopher Vogler, George Lucas, and anyone who has ever made any money for Pixar.

There’s a lot covered in Snyder’s book, but if you’re writing a screenplay using his method, you basically follow these steps:

1) You create a logline. This is an elevator pitch, or a one-sentence explanation of exactly what happens in the movie. What’s important is that you start by writing the logline. You don’t write it after you’ve written the book. If you can’t explain the movie in a sentence, you can’t sell it, and you might not even be able to write it. It’s also important that the logline says what the movie is and not what it’s about, or where it’s set, or how it feels, or anything else. But most importantly, you need to get a logline that works before you do anything else. If it doesn’t work, you need to keep at it until it does.

A good exercise is to sit down and write the loglines for a bunch of existing movies. Three groomsmen go to Las Vegas and lose their about-to-be-married buddy in a blackout drunken bender, and have to retrace their steps to find him. An off-duty NYPD cop goes to LA to visit his estranged wife and her office building is taken over by terrorists, which he must stop. A captain is sent up the river in Vietnam to assassinate a colonel who has gone crazy, or is the war what’s really crazy? A rich guy meets a prostitute with a heart of gold and falls in love. Whatever. 

2) After you get the logline, you come up with a title. Maybe the title changes later, but you do this first. It’s part of the refinement process, making a logline that summaries everything and making a title that explains it. So if, for example, if you pick a stupid title like The Journal of the Whills, and everyone you pitch it to thinks it’s stupid, you might want to keep hacking at it until you come up with something better, like Star Wars.

3) Snyder says there are ten different plots, and everything falls into one of those ten buckets. Anyone can argue it’s really 20 or 12 or 2, but he has ten. He has a sequel to the first book that goes through a ton of Hollywood movies and says which of the ten it falls into. Like the logline exercise, a good practice item is to learn the list of ten, and then go through existing movies and determine which plot they use. (There’s an entire message board where people argue about this.)

I’m not going to explain all of plots, but the ten include stuff like Monster in the House, Dude with a Problem, and Superhero. The categorization isn’t always obvious, and it’s not strictly by genre. The movie Jaws is a Monster in the House even thought it isn’t in a “house” per se. Alien is also a Monster in the House, but the house is a spaceship. You have a monster, you put it in the house, you put people in the house, you somehow piss off the monster with a Sin — something monetary or greed-based is always good — and then the people have to either get the hell out of there or somehow stop the monster.

4) One of the core tenets (and points of criticism, but I’ll get to that later) is that Snyder has a really specific 15-step outline that every screenplay should use. And each step takes up a specific number of pages. The fifteen steps form a three-act structure with the first act taking up 25% of the script, the second act 50%, and the third act 25%. I’m not going to dump his fifteen steps here; if you’re curious, google “Blake Snyder Beat Sheet” and you’ll find them. If you follow the book, your plot should not only hit each of the marks in the list, but it should spend the specific amount of time on each step. If it doesn’t, it means (according to him) that something’s wrong with your plot, and you need to brainstorm it a bit more. 

A quick example is how he beats out Act One for a script. The first beats in his outline are Opening Image, Theme Stated, Setup, Catalyst, Debate, and Break into Two. Basically, you’ve got some guy in an office/kid in a space desert/private dick hired for a job. You open with some first-impression image of their dull office park/a monstrous castle in the distance/a dreary factory/a beat-up frathouse. You spend about ten pages describing the “before” and their everyday drag. Somewhere in there you state the theme, like in Office Space, the theme of “every day is worse than the last.” And then on page 12, some catalyst appears, like the droid your uncle bought shows a hologram of a princess asking for help. Or Captain Willard is given a mission. (Every military movie has someone being given a mission on page 12.) But you don’t take the mission right away; you burn the next dozen pages in conflict, because your uncle wants you to work on the dirt farm and you’ve got shit to do. Or you’re not sure you’re supposed to use your superpowers for good, because you’re just a kid in high school. At the Break into Two moment, the protagonist basically choses that he’s got to get off his ass and launch into Act Two. Luke’s Aunt and Uncle get turned into charcoal and he tells Obi-Wan they need to sell the landspeeder, find a dodgy pilot, and find this princess. Peter doesn’t go into work on Saturday and do his TPS reports. John Connor has to bust his mom out of the loony bin and stop the bad Terminator. The monster enters the lair. The protagonist’s life suddenly turns upside down.

One important thing about this formula is you have to hit each of those five parts, in that order, with those page lengths. If you cold open the movie with Luke and Han racing toward Alderaan, you miss all the foreplay of building Luke into this boy-turned-hero. If you don’t have the period after the Catalyst where Luke isn’t sure what to do, it’s not as exciting when he does decide to do it. There’s similar structure defined for all fifteen points in his outline.

5) You divide a board into four strips, one for each quarter of the movie (act 1, act 2 part 1, act 2 part 2, act 3) and you get 40 index cards, one per scene. You outline each scene on the cards. There’s some junk about putting the emotional change and the conflict of each scene on each card. The basic goal though is that each card has a purpose, contribututes to the rise, has its own conflict. None of the cards are “spend five minutes showing cool stuff for no reason/” When you lay out the cards, you pace yourself and avoid overloaded acts and black holes. A lot of writers have an Act 3 problem, where a ton of stuff happens in Act 2, and then Act 3 has a giant “and stuff happens” black hole between the turning point and the resolution. So you’re supposed to use this board with index cards to identify the cards clumping together and the empty spaces with no cards and adjust accordingly.

6) Once you have the 40 cards and the number of pages from the 15-step outline, you start typing. I used Scapple to make my virtual cards, then imported them into Scrivener, and was able to use that to create all the blank documents I then filled in with actual writing.

The book also has a bunch of sloganized rules on writing that might be helpful, but read the book if you want to get into that. One example is the title of the book: Save the Cat. You want your protagonist to do something in the beginning to make everyone want them to win. Another one is Double Mumbo Jumbo, which is the argument that you can get the audience to believe one bit of magic, but it’s hard to get them to believe two. You can have zombies, and you can have hobbits, but if you put both together, people won’t buy it. But he states this example, and then gives several counter-examples that have made billions of dollars. Like Spider-Man has the kid getting bit by an atomic spider and turning into a wall crawler. But at the same time, it has the Green Goblin dicking around with chemicals that spill and turn him into a monster. By his rules, this is too much suspension of disbelief. But every superhero movie is going to have Double Mumbo Jumbo, so… whatever. 

There are a lot more rules, many having to do with developing your good guy or your bad guy. One that I found useful was Six Things That Need Fixing. You give your hero a laundry list of problems, which sets them up so there’s payoff when the things happen. He’s stuck in a small town, his parents are assholes, he can’t get laid, his friends are losers, his job is stupid, he wants to go to college and can’t afford it. Then when the catalyst comes, you have these various goals adding to the conflict, and when the journey starts, he can start ticking off boxes from this list. Lots of other little tricks like that exist, some that work, some that don’t. The important thing though is the logline, the genre, and the 15 steps. 

* * *

OK, so why did this interest me? I don’t write formulaic fiction, and I definitely don’t write movies. I write a lot of nonlinear fiction, plotless fiction, gonzo fiction. Unlike every book reviewer on Goodreads, I don’t think there is a problem with plotless fiction. I believe anything experimental is important, and I think a lot of the tools mainstream writers use daily evolved out of people pushing the form in experimental writing. Telling writers they have to adhere to plot is like telling painters they have to paint pictures that look like they popped out of a Polaroid camera. The fact that there isn’t more plotless fiction is honestly a travesty, but that’s probably another post.

It bugs the shit out of me that people dismiss my writing because I often don’t use plot or follow formula. After Atmospheres came out in early 2014, I fell into a deep depression because it was my favorite book I’d ever written, and it didn’t sell, and the only real feedback I got were from people who weren’t the target audience for the book immediately dismissing it with the word “plotless” and that was it. And that made me really want to write something that was so insanely plotted, there was no way somebody could say that it wasn’t. I wanted to write a book with a bulletproof plot, just out of spite. So I studied plot, and I read dozens of books, and I ended up getting hung up on the Snyder book.

Save the Cat isn’t really meant for fiction. Books aren’t film, and there’s a lot more room for more complex narrative, things that couldn’t be shot, things that can develop in a reader’s head. That said, very formulaic fiction is totally like film, so StC can easily be used for writing this.

As I was studying StC and thinking about a possible idea for this next book, I watched a bunch of movies and carefully outlined and summarized them as I wrote them, trying to find the StC plot points. I also logged the times in the movie when these events happened. This completely validated Snyder’s formulas. I did this with three movies: The Maltese Falcon, The Big Sleep, and Blade Runner. All of them hit the exact points in Snyder’s 15-step beat sheet within a few minutes of accuracy. (All three of these were what Snyder calls a “whydunit,” which is basically a whodunit except you already know who, and you want to know why. Every crime noir is a whydunit. Every whydunit has a protagonist get knocked unconscious by a hitman at exactly the 90 minute mark, denoting the start of Act Three. It’s uncanny.)

I then went back and read Falcon and some other Dashiell Hammett hard-boiled fiction, and it more or less followed the same outline. The only issue with fiction is you have to fiddle with the page numbering. A script is 110 pages; a detective novel is about 200. So your first act is going to end at about page 50; you reach your All is Lost moment at page 150, and so on. And obviously writing fiction is more verbose than screenwriting; you’re going to end up with more words on the page in prose form, rather than the fancy indenting and whitespace you get out of Final Draft.

(I actually just looked this up again, and in 2018, a YA author wrote an official franchised book on using StC for novels. I haven’t read it, and this was released years after I did this. From the Amazon reviews, it sounds like it’s a rehash of the first StC book, but for novelists. So, I guess some people are doing this.)

* * *

There are many criticisms of Snyder’s book. One is that Snyder is a hack, in the “those who cannot do, teach” way, because he only wrote two released movies that were not exactly masterpieces, and a few loose episodes of a kids’ show. (He also died at age 51, so maybe with more time, he would have had his Citizen Kane. Or maybe he would have just churned out a StC sequel book every year.) 

The main criticism of the method is his strict adherence to specific page numbers for each transition in the movie. Your script must be 110 pages. The catalyst must happen on page 12. The finale must start on page 85. Because of this, the adherence to the ten genres, and the same basic tools for problem-solving means that, according to some critics, all StC scripts are basically the same. I agree with this assumption, and it’s a problem.

There are a lot of devout followers to Snyder’s rules, and this is pretty obvious in Hollywood. I know I will get a lot of shit about this, but I personally feel like every Marvel or Pixar movie follows this strict structure religiously, and that’s turned every summer blockbuster into a Mad Libs-like script where the only things that change are what’s filled in the blanks. Yes, every one of the 167 Spider-Man reboots drastically changes something about his powers or his origin story or how hot his aunt is, but go back to what I said about loglines a while ago — you’re changing the how or the where. You can change Bruce Wayne to be more edgy or more campy or more cartoony or more 21st-century or a metaphor for why we shouldn’t be in Iraq, but you’re still following the same outline. His parents will always get killed on page 25. And if you wrote a script for Marvel that didn’t have ten pages of origin story right after the theme was stated, comic book fans from around the country would flock to your house and beat you to death with collectible figurines and drag your corpse through the streets like you were the deposed leader of a third-world country. It Absolutely Must Happen according to template.

There’s a vicious cycle with this, because when producers and yes-men are trained to recognize this structure, and see this form making money, they will only green-light movies that match the formula exactly, and then we only see movies with this outline, which means in the future, the only movies that get financed… well, you get the drift. If you’re tasked to write this year’s Batman reboot and you turn in a 450-page script that burns 87 pages pondering Bruce’s childhood before even talking about his parents getting killed, you’re going to get a ton of red pen on your pages, and see very little movement in your bank account. Stick to the formula. And if you want to write some Richard Linklater Slacker movie that doesn’t follow the curve in exactly 110 pages, you can fuck off to indie-land, deliver pizzas to make the nut on your film stock, and release direct to video somewhere. 

This is an unpopular opinion, but I have the same feeling about best-selling kindle books. Writers structure page-turners in a very specific format, and readers are placated when they hit the same plot points at the same marks, and are pissed off when the Act 3 collapses too quickly or whatever. Books that meet this exactly are reviewed higher, which pumps the Amazon algorithm and spurn higher rankings. And then the sequels have the same structure to promote more sales. This is a race to the bottom, and it’s not art. It’s how people sell vitamins and energy drinks. I know, sour grapes, my writing sucks, and I’m a shithead for saying Marvel movies are formulaic. But something is getting lost by people feeling they need to match this formula. Every book is quickly becoming the same.

* * *

Despite the arguments against it, I tried the StC method, and I wrote a book using it. (This was six years ago. I won’t even mention which book, but you can figure it out.) There were some good things to the process. One is that I often don’t title my books until the end, and my book descriptions are almost an afterthought. Starting with those made me much more confident about the direction I was going. And the 40-card process made me figure out a few dead ends before I started writing. I have a bad habit of coming up with a great idea, writing a ton, and then the whole thing falls apart when I get into Act 2. With this, I knew exactly what would happen before I even started writing. That made the writing happen much faster, and I was much more confident about what was hapening. It was easier to keep on track, and figure out exactly what I had to do on each page.

One misconception with any of these Lego-like writing systems is that they don’t do all of the work for you. There’s a lot involved in figuring out exactly what the logline should be, who the characters are, and how it should all go together. You can’t take an idea like “guys selling drugs” and plug it into a mad lib template and have Pulp Fiction pop out of it. Mining and working ideas is hard; this system only really defines the pacing of how they work out.

I went into the process with a basic setting, an idea of a main character, and an idea. The beat sheet gave me a transformation or an application of that idea, how the protagonist struggled with the idea, and it forced me to use a certain number of characters to move the protagonist through the outline. It helped me develop my protagonist, and differentiate the other characters, not only to make them more interesting, but to make them more integral to the movement of the plot.

Another big thing this helped me with is the dynamics of the plot, the movement. Snyder has this saying, “Turn Turn Turn,” which is that a plot doesn’t just have to move, it has to intensify at each step. And this helped me a lot in my Act 2 to Act 3, which is what I always screw up in a book. I was able to raise the stakes through the plot in exactly the right proportions, but it also made it so my chases were more than just moving from point A to B really fast; it gave meaning to the chase, which brought the reader through the outline.

* * *

I really enjoyed writing the book, and I liked the structure of it. It developed well, and the experiment was a success in that way. But short story long, it did not sell. My faithful readers thought it was way too off-brand. “Serious” science fiction readers didn’t get it, and nitpicked the plot. (There are some other factors involved, and maybe I’ll write about that someday.) I proved to myself I could do it, but that I didn’t need to. I went back to writing weird non-linear stuff that doesn’t sell, and I guess that’s my lot in life. I sometimes think if I had the perfect idea, I’d do this again, but I think a lot of dumb things.

Anyway, this is the most I’ve ever written about plot, so I better get back to writing without it, before someone takes me seriously.



    • I just “got back” from a one-week “vacation” I had to take. I couldn’t go anywhere, and I couldn’t sleep in, because Sarah’s office is the bedroom. So I had plenty of spare time to watch the news, which was just a bad idea.
    • A week ago, as I was closing in on one of those deadline-of-the-year projects, my crown fell out. It was the same one I lost back in 2015 while on vacation in Indiana. I got ahold of my dentist, but it took a week to schedule a repair on it, because of COVID stuff and reduced schedules. I went on a Wednesday afternoon, and had the shortest trip across the Bay Bridge I’ve ever seen, ever.
    • My dentist is at a mall (Tanforan) and I did a quick lap (outside) before I went in. One of the anchors is Target, which was business as usual, mostly. There’s a JC Penney, which was closed, and a Sears, which died earlier this year. So there was a certain creepiness, but there’s also a BART station there, and people always illegally park at the mall, so it wasn’t that vacant.
    • There were a million various protocols: take a giant quiz the day before on where I’ve traveled and how I feel, etc; wait in the car and call them when I get there; come in with a mask on; temperature check; the same quiz, but sign and date it; wash your hands (timed); rinse your mouth with peroxide (timed); then mask off. The room was lined with plastic like I was about to get executed by the mafia. The dentist and assistant were in full protective gear, masks and shields and gowns. The crown was fine, and the re-glue was a two minute job.
    • Speaking of abandoned malls, I went to Hilltop last week, in Richmond. Couldn’t get in, although the Walmart was open, with people lined up outside. I did a lap outside the mall, which was eerie. The most ironic thing is that the JCP and Sears parking lots are now being used for storage for a ton of those Amazon sprintster delivery vans.
    • Also went to Sunvalley out in Pleasant Hill. They had Jersey barriers at every entrance, closing off the entire parking lot; every door and window was boarded over with plywood. Took a long walk around the outer ring of the lot, and it all looked a bit too surreal.
    • JC Penney put out their list of store closures for their bankruptcy. The Elkhart store at Concord Mall is on the list. I could probably write a part 3 on the death of that mall, but I’m too lazy. University Park isn’t losing theirs, and none of the Bay Area ones are slated for closure. I mean, they all will close in the future. But those aren’t in the first round.
    • Spending a lot of time walking NAS Alameda, trying not to fall down the k-hole of researching what the base used to be. Luckily, it’s so poorly-documented I don’t have much to go on.
    • There’s more, but nothing I want to discuss here. The world is a crazy place right now.
Posted in Uncategorized