Thursday, September 08, 2011

Tell 'em What You're Gonna Tell 'em


A few days ago I sent out a link to friends and students to a story about a studythat said readers found stories more enjoyable if they knew the ending first.  This seemed obvious to me, but I was informed by a few people that this finding was counter-intuitive.

For me this study was nothing master storytellers haven’t known for millennia.  The first rule of story construction is “Tell ‘em what you are going to tell ’em.”

To explain to my friends and students why this finding was no surprise to me I sent them the following response. They thought it would make a good post, and so here you are:

A friend once asked me to read a novel he was writing, and I tried, but could not get through it. I told him that it bored me and I couldn't finish because I didn't know why I was reading it—he needed to tell people why right away. 

He didn't know what I meant, so I asked him if I ever told him about the time I almost got shot. He said that I hadn't, and he leaned in to hear the story. Then I told him that it had never happened, but it was clear he wanted to hear that story.

Starting with the ending is often how stories are told in the natural world of conversation because it engages people—it hooks them.

“I was fired today.”

“My wife’s been cheating on me.”

“My house was robbed.”

Those are the endings of the stories. Then the teller goes back in the timeline to flesh out the details.

I remember bad reviews of TITANIC (a mediocre film in my opinion, but with some very strong story elements) where reviewers panned the film because they couldn’t understand why people would sit through a story where they know the ending—the ship is going to sink.



But that knowledge is the engine that drives the story. People sat through the film because they knew the boat was going to sink. And they knew not just because of the history, but because Cameron put that information right up front.

In SUNSET BOULEVARD Billy Wilder starts the story with the body of William Holden floating face down in a pool and the rest of the film is his voice-over telling the story of how he got there. But we know right off he ends up dead, so we watch to find out how.

Wilder did basically the same thing earlier in DOUBLE INDEMNITY where you see that the character Walter Neff is hurt—shot. And he then tells the story of how he came to get shot into a tape recorder because he wants to get the story out before he dies.

Old stories use this structure: This is the story of how the brave knight defeated the dragon.

Rudyard Kipling's Just So Stories: How the Camel Got His Hump.

Sometimes the serial nature of story implies the ending. We know that Batman will defeat The Joker. And that Superman will win against Lex Luthor and that Sherlock Holmes and Miss Marple will solve the mystery: Yet this does not stop us from enjoying the journey.



The classic long-running television detective show COLUMBO is often called a mystery show, but it was not a mystery at all. Each show started with the murderer committing the crime. The audience saw every aspect of how and why the crime was committed before the detective was ever introduced.


Of course the viewers knew that Lt. Colombo would solve the crime, but they did not know how. The fun was in seeing just how the detective solved the murder—or how he trapped the killer into admitting their crime. It was engaging and suspenseful. So much so that the show “ran,” in one form or another, from 1971—2003.

Hitchcock said that he never created suspense by keeping information from people, but by giving it to them.

Because, I believe, we are listening to stories for survival information we are want to know just how William Holden died so that we can avoid it. My friend with the novel wanted to know just how I got away without being shot.

Since human conversation is the parent form of all storytelling, it contains the seeds of drama and literature. If you listen to stories in “the wild” when people are talking and may be unaware that they are even telling stories, they will often start with the ending.

Imagine someone from a non-profit speaking to an audience of potential donors. She shows a slide of a distinguished businessman on a suit in a fancy office and says that this man was once homeless. People will listen.

But what if she started with a slide of the man while he was homeless? The lead-up to the ending it would be far less interesting. It may be a surprise ending, but it's a far less interesting story.

Knowing the ending does not answer all the questions—it poses more questions. Plus people understand why they are they are listening, or reading, or watching.

35 comments:

Clint said...

Excellent post; stellar examples. You're right on about how real people tell stories. That said, it does matter to me what kind of story is being told for revealing the ending first not to distract or disappoint. Sunset Boulevard is an absorbing film, but I was both distracted and disappointed by knowing Holden was going to die. I would still have been disappointed to find he dies if it hadn't have been given away, but I wouldn't have been thinking about it so much during the film. I think having a hunch about how a story will end--say, Star Wars, Raiders, Jaws--can be just as powerful, without making the ending too anticlimactic. I'm not saying that such endings necessarily reveal all the interesting details, only that even knowing the general conclusion can, depending on the kind of story being told, be too much for sustaining the same level of excited interest as you would otherwise enjoy. Coincidentally, I've been weighing this very matter in my current script. I'm considering threading the needle by using a series of flashbacks from the build-up to the climax, somewhat like Usual Suspects. That way the audience gets the idea of the nature of the conclusive action, but doesn't find out how it actually resolves without watching to the end. Not sold on it yet, just because I'm partial to the impact of linear story-telling in this type of story, and am often put off by too much contrivance in how a story is related. I don't like having my attention too drawn to the nuts and bolts. But it does help with some aspects, and could be compelling if I can pull it off well, so we'll see.

Thanks for another well-distilled, thought-provoking entry.

Brian McD said...

Hey Clint,


Always happy to know that people still want to read what I have to say.

Thanks for the kind words.

As for structuring your piece, the best thing to do is to write the story the same way you would tell it verbally. It may even help to tell a friend. You have to learn to listen to the needs of your story. Just do what the story needs and don’t embellish and you’ll be fine.

I hope that you don’t mind another piece of unsolicited advice – assume that Billy Wilder is smarter than you are. He has a list of bona fide classics as long as your arm:

The Major and the Minor, Double Indemnity, The Lost Weekend, Sunset Blvd., Ace in the Hole, Stalag 17, Sabrina, The Seven Year Itch, Love in the Afternoon, Witness for the Prosecution, Some Like It Hot, The Apartment, The Fortune Cookie

He has four films on the AFI’s list of the 100 best films ever made. These do not include the classics he wrote before he became a director: Midnight, Ninotchka, Ball of Fire, Bluebeard's Eighth Wife.

I mention this because few people in the history of film have his track record and the best way to learn from him is to assume he is right and you are wrong. Sure, there are times when he made mistakes, but he made fewer than most. When you think he’s done something wrong you’d do better to question yourself before you question him.

That’s what I do.

Paul Chadwick said...

This speaks to peoples' interest in how and why.

Or, as you might put it, survival information.

I remember Robert Altman's lifetime achievement Oscar acceptance speech. He said he wasn't interested in stories, particularly, but in human behavior.

That might explain the shaggy construction of some of his films, but it also expresses a truth: we consume stories to identify with people going through troubles, and being amused, moved, or instructed by how they act.

It's not "will he get out of this?" but "how's he gonna get out of this!?"

That's a thought that occurs looking at that N by NW still on the post below.

nordst said...
This comment has been removed by the author.
nordst said...

Brian,

I have been reading every blog you have posted for over a year now, and I must say that although I have enjoyed all of them this is the best.

Your advice in this post is tactical, and yet very straightforward.
You really understand to bring the tacit knowledge that all humans inherently have, and make it explicit by use of great examples as well as great storytelling-structure.

Already looking forward to your next post!

Joon Kim said...

This blog entry reminds me of a story I sometimes tell to people. It begins with:

"Did I ever tell you about the time I got arrested?"

It has always been effective in generating interest in the story. Too bad the actual incident isn't as interesting as one might hope from that opening statement. Hahahahah

But my story definitely has some 'survival' information.

PAY YOUR SPEEDING TICKETS ON TIME!

imyjimmy said...

Thanks for sharing, Brian.

I saw you tweet about this the other day, and I was wondering that a bit myself. Then I saw the film Gandhi, and I noticed how they used the end to hook the audience into watching the rest of the film. It was an epic film by the way, very important subject matter. I don't think I've ever cried watching someone request orange juice and bread.

Anyways, my question to you is more about the scientific process. As you remember your tweet was more along the lines of (and I'm paraphrasing here), "look how long it took scientists to figure this out, storytellers have known this forever, link." So I took it to mean that sometimes the scientific process isn't the best way to discover something.

Scientists, by definition, are bound by the scientific process which is: pose a question, set up an environment where as many variables as possible are controlled, tweak one variable, and watch/record the result.

Scientists have been doing this for thousands of years and it has led to many innovations. But recently scientists have the power to look inside people's brains, and it has led to some pretty interesting/weird studies, in fields that are traditionally for english / drama / philosophy / religion.

There are scientific studies trying to figure out what the funniest joke in the world is. I scratch my head at that one.

Then there is the study on mirror neurons which you actually use in your story class.

And then there's this study on story endings. But as you pointed out master storytellers have been using this information all the time.

This has left me wondering, to what extent is the scientific process valid? Maybe the real problem is, scientists are too confined to a test tube environment. They should also observe real life. Go into the wild as you like to say.

Some of the best scientists had an innate intuition about things. They didn't do the test tube thing, necessarily. And some of the best scientists are also artists. I would categorize Da Vinci as a scientist / artist. His writings and drawings can be technical, or artistic. And most of what he did was observe. He didn't necessarily use the lab environment. And then you have guys like Aristotle who made an important impact on every field imaginable. How did he do that?

I guess I answered my own question. It's just that right now, scientists can't proclaim a fact unless it's been lab-tested. I wonder if they would let the apple fall on their head, so to speak.

Clint said...

Hey Brian. Your advice to think about how I would tell a friend my story is more spot-on than you might think. A couple of years ago, when I began working on this particular script again after a break, I told myself, "Okay, you're overthinking this. This doesn't have to be so complicated. What's the basic story? I'm telling somone this story—how do I start?" I started a Word file to write my answer, which began, "Once upon a time, there was a pilot named Jay...." You're right: There's a reason that form has lasted through time. It concentrates the mind wonderfully. (And now you know why I bought your book after reading the first page of the first chapter. Incidentally, your 7-step version was more helpful than the four steps I came up with. It prompts the mind better and fits the basic three-act paradigm better.)

As for Billy Wilder, I didn't mean to suggest, if it came off that way, that he was somehow wrong to use that start-with-the-end structure. I was only explaining that for me personally, while it does heighten the curiosity, that construction dilutes the impact of the end for me, as well as creates a distraction, however subconscious, while I'm watching the film. "Oh, man, I like this guy. I don't want him to end up dead." That sort of thing.

But I certainly don't begrudge him the method or question his genius for storytelling. On the contrary, he's one of my favorite filmmakers. I know his catalog well. The Apartment? Don't get me started. One of my all-time favorite films. But one of the reasons I enjoyed it more than Sunset Boulevard is because Wilder chose not to use that structure in it. He went linear for that one. So it's just a personal preference. Knowing the end at the outset with any precision doesn't work as well for me. I'm just glad Wilder knew how to use all the tools in his tool box equally well. He makes it look so easy.

Brian McD said...

Nordst,

Thank you so much for letting me know that the blog has helped you.

Glad to know that some of the things that I say help other storytellers.

- Brian

Brian McD said...

Hello Paul,

Thanks for the Altman story. Most drama now does not deal with real human behavior. It's really a shame.

Brian McD said...

Joon,

Thanks for the story about paying your tickets.

Brian McD said...

Hey Jimmy,

At its core science is about observation, isn’t it? Jane Goodall is a scientist and all she really does is watch.

One of the ways I learned what I know comes from going to the movies as a kid and paying attention to audience response. Or I watch how people tell stories to each other. I have done this for years.

People tell stories all the time so the world is a lab. And master storytellers have done a lot of work in that lab.

I used to wonder why scientist didn’t talk magicians about human perception. Magicians after all, have practical experience when it comes to this. In the last few years scientists have been doing what I thought they should have been doing:

http://www.youtube.com/watch?v=J5x14AwElOk&feature=player_embedded#!

http://www.pbs.org/wgbh/nova/body/psychology-magic.html

I guess I just felt like there were experts in the craft of story that science could have taken advantage of.

That’s all I was saying.

Brian McD said...

Hey Clint,

Glad that the advice spoke to you. Good luck with the story.

As for Wilder’s work happy to hear you’re a fan. He was pretty great.

Ian said...

Wow, this post gives me a lot to think about! I really understand what it means that "all good storys end the same way it begins." I think it is kind of the other way around, you tell them what you are going to tell them, then at the end, you prove your point.

When you said your opinion of the film titanic, I was wondering what you thought of Princess mononoke? (I know you hate questions like that though, I just cant help but ask) I was reading on wikipedia that Princess Mononoke was one of the best selling movies in japan, only to be surpassed fairly quickly by titanic.

David said...

I'm a big fan of the kind of suspense examples you show, and agree with all the examples you give, but I'd nonetheless like to argue that knowing what happens at the end actually can harm our full enjoyment of a story.

I remember watching Se7en in the cinema, and was absolutely riveted at the climax - I had that giddy feeling of being utterly at the mercy of good drama. As Brad Pitt held his gun to the killer's head, the characters and the audience were in complete alliance in our knowledge of the situation. None of us knew whether the villain's plot would be completed, or if the moral pleas of the colleague would prevail, but we cared desperately about it. If I had known the ending of the film, I would have been denied that 'what happens next?' feeling which I crave as an audience member.

It's not for nothing that it's considered impolite to reveal the ending of Citizen Kane before someone has had the opportunity to discover it for themselves.

Endings in a proper story are of course foreshadowed, but my favourite definition of a good ending is one that is Unexpected Yet Inevitable. I'd hate to be denied either of those two things.

Brian McD said...

Unknown,

Happy to know that the post got you thinking.

Sorry, but I cannot get into Princess Mononoke because

My real life is too full of people asking me if I like this or that film and why or why not and I don’t want this blog to become that.

I do The Movies I Like posts to give folks an idea of what I think works and why.

Sorry. Thanks for understanding.

Brian McD said...

Hello David,

Thanks so much for writing.

After many years of teaching I have noticed that people are often using the same words and speaking a different language. We may mean different things when we say “knowing the ending”. It is still possible to surprise people with your ending. I talk a little bit about this kind of thing in my previous post about cliché.

Also no one is a bigger fan than I am of Rod Serling and the original Twilight Zone that show is all about surprising endings. But the construction of those shows is so solid that the endings are, as you say, “Unexpected Yet Inevitable.” You can read more about that at an older post: http://invisibleinkblog.blogspot.com/search?q=Shyamalan+is+Not+Serling+

Even Citizen Kane, which you mentioned, starts with an overview of Kane’s life and his death and yet there are surprises. The entire story is of his life is told in that prologue:

http://www.youtube.com/watch?v=qY-eqnw_DXE

I understand wanting to be surprise, but these two ideas are not mutually exclusive. And even if you do not agree with me, how do you account for the findings of the study?

Ian said...

I can understand why you did not answer my question. I also really appreciate all of your "Movies I Like" posts, they are all very helpful and I have read them several times.

However, I still have one more question to ask.

What the process you go through to tell if a film is well written? I want study the best writing I can, so it is important for me to be able to separate good writing from the bad. With so many bad movies out there, I find it difficult to figure out which ones are actually good.

Unknown said...

After reading that email I was wondering how long it'd take for you to bring up that study on your blog... : ) nice post Brian. You always highlight such great universal Truths... this is one of my favorite! Not only do we tell stories this way, but we were taught to communicate like this in elementary school - the 5 paragraph essay right? Though maybe you've made that example before and that's why it popped in my head...

Brian McD said...

Hey Ian,

You will have to find for yourself what you like and why. For me it was about studying the classics – that’s how I learned. I read interviews with old directors and writers to learn what they knew.

Be open to learning and be open to changing your mind as you grow. A film you think is great today may lose its luster as you learn more and you have to be open to that. Most people are not and their development stalls as a result and they start making excuses for substandard work.

For me a film that works is one that communicates clearly and keeps the same idea from the beginning to the end.

Santiago said...

This post about endings reminded me of why I hated the movie Open Water. In a way, by expecting a happy ending (that the stranded divers would be saved by the end of the movie) I gave the annoying characters a chance.

When neither of them survived I realized that I had suffered through 80 minutes for two characters that had no redeeming qualities and whose deaths I actually welcomed.

Sometimes playing with the audience's expectations can be a boon, but unfortunately most of the times this does not happen in a good way.

David said...

Brian thankyou for your thoughtful reply.

I agree that they are not mutually exclusive, and that we're probably discussing two different things; this is intended as an 'yes and' not a 'no but' to your essay. I guess that since you seemed surprised that people were nonplussed by the study, I thought I'd explain why I had had that reaction, and why you might find the opinion vigorously defended.

Knowing the general direction of events, the broad sweep of a story is what you've identified as useful (John McClane thwarts the terrorists, the two musicians survive as witnesses of the St Valentine's Day Massacre, the bridge over the Kwai is destroyed). I'd agree that these generalised endings tend to be implicit in the genre of the film, so once audiences are a few minutes into a film they're aware at some level that this is where the story is headed. The problematic thing for me is knowing specifics of the ending (The terrorist leader dies falling from the top of the building as McClane is reunited with his wife, the two musicians escape to the rich guy's yacht with the girl after the mobsters kill one another, the British Colonel accidentally detonates the bridge after battling to save it).

I believe the films were constructed such that they worked best when these kind of details were unknown at the start. Even in films such as Sunset Boulevard where the storyteller uses the ending as a hook to draw in the audience, we're left in the dark about details of what happened - that's to be discovered in the telling. I think there's a fundamental difference between a storyteller who has chosen to tease the end of his tale to draw us in, and a fellow audience member who tells us events that are only revealed in the final act. It undercuts the choices of the storyteller, who hopefully has built the story artfully and is revealing information only when it's best. And I'd suggest it's not merely twist endings that matter, it's all the tiny details that go in to building the totality of the tale. I trust Billy Wilder to tell me what I need to know when I need to know it, and want no-one spoiling that experience of sitting at his knee as he entertains me. It's these kind of spoilers that I react against, because of the difference of the first time viewing.

In my understanding there are two different ways to experience any given movie - for the first time, and then every subsequent time. They hold different pleasures for the audience. Repetition will of course be satisfying, but don't underestimate the rather different enjoyment of discovering it piece by piece, as the teller intended.

Working as an editor I've actually noticed what you might call an institutional bias amongst filmmakers against the experience of the first viewing. When I'm cutting a film, the director and I are constantly fighting the fact that we know the story inside and out. Creating the 'Illusion of the First Time' when watching it back is incredibly hard. We do all sorts of things to combat it, but even with fresh eyes coming in and test viewings, it's only natural that films tend to be optimised for the repeat viewer. After all that is who the filmmakers are - everyone from the scriptwriter to the sound mixer are making the vast majority of their choices whilst knowing what happens next. The vast majority of audiences however will ever see the work for the first time.

Without reading the study (there's quite a high fee to access it online) it's hard to comment; we don't know if the additional information was of the general kind or of the specific kind, revealing the what or the how. I would guess it was of the general kind, and the increased reader's pleasure was due to suspense.

Thanks for writing your blog, and for reading our responses.

imyjimmy said...

"Imagine someone from a non-profit speaking to an audience of potential donors. She shows a slide of a distinguished businessman on a suit in a fancy office and says that this man was once homeless. People will listen."

That's a very cool observation and it speaks to how you can rearrange events in a story to get what you want.

That said, I just saw a movie that shows the main guy starting off as a loser figure and then transforming into a success: The Verdict. Lumet and Mamet told the story straight up, but then again they were not asking for donations for down and out lawyers. That was not their point so the end was not their hook.

What I wonder about The Verdict is, what keeps people from saying, "Oh he's a loser alcoholic, let's switch channels." People love to ignore losers. When's the last time anyone you know stopped to talk to a hobo?

It's a little off-topic from your post but I couldn't help and wonder.

Phil Laaveg said...

Great post, Brian. And, it's backed up by science:

http://articles.boston.com/2011-09-19/bostonglobe/30176405_1_stories-spoilers-reading

Michel Maas said...

Hi Brian,

It's not the first time I understand you think Titanic is a mediocre film. Can you tell me what it is? Maybe in a next post?

Thanks!

Brian McD said...

Hey Santiago,

I have noticed that when things work people tend not to notice, so that when I bring up a concept the only thing that comes to mind for most people is when someone has done it poorly.

It might help you to look for where this has worked, you'll be surprised at what you find, I think.

Brian McD said...

Hey Michel,

I think that Tatanic does some things really well, but for me it is not up to what James Cameron can really do. Aliens and T2 are, in my opinion, better constructed stories.

I think that is nothing short of a stroke of genius that Sarah Conner, who was terrified of the killing machine becomes one herself. She even tries to kill someone for something they haven't even done yet -- she becomes a terminator. Genius. Really.

That guy can do better work than he has done for a while. That's all I'll say about that. I try not to put films down on this blog.

imyjimmy said...

Brian,

I just saw Road to Perdition and it had the same thing you were talking about. The movie starts with the kid, having already gone through most of what they will show in the movie, talking about his father: was he a bad guy? And then it backtracks. By the end of the movie we find out that sons shouldn't judge their fathers. Amazing.

Road to Perdition is a good movie in my book.

Enrique said...

Hi Brian,

I have just read the original treatment of Terminator by James Cameron and it reminded me constantly of your theories. If you read it, he keeps repeating thematically again and again that the machines are just a part of what we are as humans, and not a bad thing for itself. I thought you may want to use it as a good example in your classes because it's shorter to read than the actual screenplay. Moreover, maybe because James Cameron was younger and more unexperienced, how it is done is a lot more obvious.
Here is the link to the treatment. Hope it helps and thanks for the great articles:

http://www.scifiscripts.com/scripts/Terminator_Treatment.txt

Kaki Flynn said...

Brian,

Have you ever considered doing a v-log of you teaching Invisible Ink or The Golden Theme?

Not a lot of people have the ability to teach creative topics in a straight-forward way.

(I get the impression from talking to someone that took his classes Walt Stanchfield nailed this to perfection).

I bet you people would pay for them.

Even if you posted them for free, it would draw more people to your books, which is a good thing (more good stories will get told; less miserable screenwriters).

Like this:
http://www.lynda.com/Documentaries-training-tutorials/99-0.html#Creative+Inspirations

Kaki

Brian McD said...

Hey Kaki,

I have thought about that. Thanks for the reminder. And thanks for the compliment about my teaching.

- Brian

Scott Lincoln said...

Hey Brian,

I stopped by here after reading your Book "Invisible Ink" which I found to be both excellent and objective. Your breakdown of masculine/feminine writing hit the nail on the head for me. Being a cartoonist I am obsessed with contrast and balance visually and I know this concept will help me keep me my writing contrasted and balanced as well.

This post caught my attention because of the mention of Columbo and serial stories. about a month ago (before I found your book) my wife and I were watching episodes on Netflix when we stumbled onto that precept; This isn't a mystery because they tell you who and how. That and "Sunset Blvd" really reenforced the "start with the ending" idea.

I am, however, noting in this post the idea of the serial story having an implied ending. Is it implied because we've seen it happen before so many times or because of our perception of heroes in general? I'm wondering if I'll need to compensate in my story since my hero is an anti-villian?

Thanks for a truely great read, Scott Lincoln

Brian McD said...

Hello Scott,

Thank you so much for reading the book. Happy to know that you got so much for the masculine/feminine stuff. Yeah, that contrast and balance stuff is useful, isn’t it. I apply so much of what I have learned over the years from my art friends to my understanding of story and filmmaking. That’s why I watch this Paul Rand video from time to time:
http://adland.tv/commercials/one-show-paul-rand-tribute-film-2007-400-usa

About starting with your ending – first off it helps to have a sense of your ending because when you know where you are headed it makes your construction smoother. You will tend to create things that matter to the overall story rather than a random series of events.

But starting with you ending is about hooking your audience and that can be done in a few ways. If you can get the audience to ask questions then they’re hooked. They want to stick around to get the answers. How did this guy die? As in Sunset Blvd.

These things have little to do with what kind of characters you have in your story. Structure has little to do with the types of people that populate one’s story world.

I hope that helps you with your piece.

Thanks again for the complements about the blog and book.

-- Brian

anne said...

I like to explore the idea that there's an important distinction between knowing or expecting the ending and predicting the ending. I think it has to do with the writer's control over information and thus the story. If the writer controls the information, I think this works well to create suspense for the viewer. But I've also watched many movies where I've seen the ending coming like a freight train, and it just kills the whole thing for me. I suspect in these cases the writer was trying to keep the ending a surprise. Thus the enjoyment of the movie is diminished by being able to predict the ending.

This theory seems to suggest that it is to the writer's advantage to show all their cards, especially if the ending is a weak one. Tell the viewer where they are headed. Then you diffuse any potential disappointment about the ending and allow the viewer to focus on the journey and the telling of the story itself.

Brian McD said...

Hello Anne,

Thanks for writing. I'm not sure I meant to say what you got from the post. Yo said, "This theory seems to suggest that it is to the writer's advantage to show all their cards, especially if the ending is a weak one. Tell the viewer where they are headed. Then you diffuse any potential disappointment about the ending and allow the viewer to focus on the journey and the telling of the story itself."

Firstly, I am not exactly sure what you mean when you say "weak ending". And ending may be dissatisfying for a few reasons. It may have an ending that seems tacked on because it was not set-up properly. Or the main character may do something out of character. Or maybe the ending just feels contrived or convenient. All of these issues have to do with how the story was set-up.

If one's story is a puzzle than I guess it is better to keep the ending a secret from the audience. Though I don't think a puzzle is the best use of the dramatic form because it is intellectual rather than emotional, but I suspect you disagree.

Whatever one's ending is it had better make sense within the context of the piece. This is what makes any ending weak. There are no tricks one can use to make a weak ending strong.

It's just that if your ending is the only reason you are telling the story, than the story itself is weak, not the ending. A weak ending is always a symptom of bigger problem.

If your story needs a surprise ending then by all means use it. but more often than not one gets more mileage from suspense.

Yes, it can suck to see an ending coming, but this almost always means that there is nothing of value happening within the body of the story and all they really have is their ending.

Let me put is this way, M. Night Shyamalan is a proponent of the surprise ending and has become a virtual laughing stock. Hitchcock, on the other hand, was in favor of suspense and had a thriving 50 year career.