Novel length & word count

[katdad]

I copied this message from the old forum because the original thread is GONE...

Okay, I just did a little research work and testing.

I took my novel and first deleted all the 'chapter 23' stuff, all the footers, title page, and everything except the actual novel text itself.

Next I did an MS-Word count, using Word 2000. I tried changing the fonts and styles and page margins and it always gave me the same count: 67,252.

I then saved it as a pure ascii text file, opened that file into an ascii text editing program, saved it again, and re-opened the text file into MS-Word. Count: 67,257.

So apparently the process of converting to pure ascii and re-opening as a Word file added 5 words, probably the result of hyphenated words. (I don't hyphenate words on line break, of course).

Next, on my older computer, I re-opened both these test files in Word 97 and ran a word count. 67,252 & 67,257 respectively. So, MS-Word 2000 and Word 97 agree perfectly.

Finally I re-opened the two files in Corel WordPerfect-8, and the new count was 67,206 and 67,208. So WordPerfect apparently low-counted the document by about 50 words, a disagreement of less than 1%.

And then, I took two random pages from my novel, one with lots of narrative, one with mostly dialog. I counted the actual words and then cut & pasted these pages, performed an MS-Word count on them. The physical manual count and the MS-Word count agreed exactly:

Page 51 = 302 words
Page 149 = 152 words

So here's my somewhat scientific conclusion: MS-Word performs a very accurate word count, accurate to less than 1% error.

As I see it, there is no validity to the idea that counting words or spaces and dividing or whatever is better than using the software. Those methods are only estimates, while MS-Word actually counts the words themselves.

If anyone has information to challenge this, please let me know. In the meantime, I'll stick to MS-Word's methodolgy.

[Azure Skye]

This is good news to me because I hate counting.:eek:

[HConn]

As I see it, there is no validity to the idea that counting words or spaces and dividing or whatever is better than using the software. Those methods are only estimates, while MS-Word actually counts the words themselves.

If anyone has information to challenge this, please let me know. In the meantime, I'll stick to MS-Word's methodolgy.

http://www.sfwa.org/writing/wordcount.htm (http://www.sfwa.org/writing/wordcount.htm)

The words are less important than the number of characters.

But that's not the most important thing about word counts. (http://www.sfwa.org/writing/wordcountredux.htm)

[Maryn]

I did a very similar experiment on my own work a few years ago, with similar results. My mostly-dialogue page had a few items open to interpretation on whether to count them as words or not (words cut off before completion by another character's interruption, i.e. "I might do that, I just mi--"), and I found that Word tended to agree with my call, where WordPerfect seemed not to count words not found in its dictionary.

I figure that nobody really needs to know the literal word count. What they need is a ballpark figure for how many printed pages my manuscript would be in book form, so the differences from one program's word count to the other's don't really matter unless they're quite large.

Maryn

[katdad]

http://www.sfwa.org/writing/wordcount.htm (http://www.sfwa.org/writing/wordcount.htm)

The words are less important than the number of characters.

But that's not the most important thing about word counts. (http://www.sfwa.org/writing/wordcountredux.htm)


Please don't play the tease and require us to skitter off to another site to learn what you're talking about. Say it out front first, then list your references if you wish. Thanks.

Or, as Molly Bloom told Leopold, "Tell us in plain words."

By the way, my agent has never said a single thing about the number of characters. He wanted to know approximate word length. But if character count is important, MS-Word can do that too.

[katdad]

I figure that nobody really needs to know the literal word count.

Agreed. What I do is perform a MS-Word count, then round down to the next lower 50 or 100 words, playing it a bit conservative.

Let's face it -- the final word count of your novel will be performed after all the galley proofs are edited and approved anyway. Anything prior to that is only generalization.

[maestrowork]

Actually publishers would even care less about word count. What they care about is page count. A page full of text or a page full of sparse dialogue mean the same to them -- a printed page.

[HConn]

Or, as Molly Bloom told Leopold, "Tell us in plain words."

Okay. How about this:

The words are less important than the number of characters.

[katdad]

What they care about is page count.
You're right, of course. But the page count is dependent on the typesetting, font size, format, and such. If they have the word count they can convert to an estimate of pages, but that's still only a reasoned guess until the galley proofs are created.

I've had articles of about the same length show up in final print as 2-pages to 5-pages, depending on the spread and layout.

All I'm saying is that if you have a reasonably honest word count, nothing else is relevant until the book begins to be edited. Publishers just don't want to be reading a too-short or too-long novel from a newbie writer, in most cases.

Also, I'm saying that the estimate word count methods are only that: estimates. Using MS-Word to do the count is far more accurate.

There are exceptions, of course. But I'm not trying to write "The Boogaloo Trilogy: Princess of Dragonsfane" either (ha ha). I'm just writing hardboiled American private eye, and most of those are in the 65k to 80k word range.

That's a specific sub-genre that tends to be on the short side of most novel lengths. This is because most PI novels deal with a single case (a murder maybe) and then the fallout from that event. It lends itself to a quick read and a fairly linear story line. The style itself is also contributory to the shorter novel: brisk dialogue, breezy narrative.

[JanaLanier]

Katdad, I'm glad you resurrected this topic.

My two most recent projects both ended up being around 50K after the first draft. I could probably flesh them out some, but any padding I add wouldn't change the essential story. I'm paranoid no one will want to buy a short novel from a first time novelist. Is it hopeless? Should I bag them and start over with a new, longer project?

Uncle Jim mentioned that his outlines are about 75% of the finished project. To me, that's not an outline, that's a book that needs to be fleshed out. So Uncle Jim, what do you add in your second draft that gets you that last 25%?

Thanks all. :)

[Mistook]

I was under the impression that the word count was mainly for the benefit of the typesetters, and that by their definition a "word" is always six characters, so you take the total character count of your MS and divide by six and that's the word count they're worried about.

[Birol]

Yes, Mistook, which I think is the point HConn and those talking about page count are trying to make.

[Euan H.]

I was told recently that word count should be done on the basis of assuming 250 words per page of standard ms. format (12 point Courier New, 1" margins, double-spaced). I can't see how the actual word count would really be important--it's the amount of space you need, right?

[RGame]

What I can't figure out is why publishers wouldn't want to buy a short book by a new author. It seems as if a book in the 60,000 to 75,000 word range would be a better idea for a new author. The page count would be a little lower than the average book, which might make a reader more willing to take a gamble with a new author. Less time invested on something they're not sure about. And fewer pages might bring the price down a little too, also making the book a better prospect.

[Mistook]

Not to mention less publishing cost, RGame.


From what I understand the shorter and sweeter the better, especially for a first time author. If it's good and it's short, you're in a great position. I'd love to come in under 55K when my whole WIP is said and done.

[katdad]

My two most recent projects both ended up being around 50K after the first draft. ... Hopeless?...
Very likely, I'm sad to say.

When I was searching for an agent, I had my two novels in the mystery series "completed" (or so I thought) but I received four separate rejections from legitimate agents who told me the books were too short. (44k and 48k respectively).

I did some more digging and learned that even private detective novels, which tend to be among the briefest of books, are generally a minimum of 60k-65k words.

I sat down and re-wrote the books from the beginning, greatly improving them, and also adding sub-plots and expanding earlier sub-plots. The books then were in the 65k range and that was sufficient. Now I've signed with an agency and my agent is shopping the books to publishers.

Don't despair, and don't think you have to "pad" your books. Explore secondary plot lines and secondary characters that may expand the story lines a bit. You're close to the minimum as it is. So don't give up. Look for places where you may have originally rushed through a section of the novel, and see what happens when you take a bit more time and write more about what's going on at that time.

I'm not saying absolutely that a 50k novel won't find a home, but it's unlikely, especially from a newbie writer. Same for an overly long book.

[katdad]

What I can't figure out is why publishers wouldn't want to buy a short book by a new author.

True, it costs slightly less to print a shorter book, but there is a basic bottom cost that is incumbent on printing and distributing ANY book, regardless of how short. That's overhead and it's basically the same throughout.

Think of this in terms of an imported wine. Imagine how much it would cost for some French or Italian bottler to ship out bottles of simple tap water, corks and labels and all. There would be a basic price floor regardless of the contents and their quality.

Books that are too thin won't be noticed on the shelves, and will be passed over by readers because the reader won't think they are getting a bargain. They will be spending maybe $12.95 for a paperback, $24.95 for a hardback, and if it's a buck cheaper, that won't influence their purchase.

Long books face a similar problem. There's a ceiling for any general reader to buy a paperback or hardback. For paperback it's about 10 bucks, hardcover it's about 30.

This is less of a problem for established writers with a solid readership. But for new writers, there are general upper and lower limits and it's very difficult to get your book sold if it's too short or too long.

You may bring up lots of good reasons to counter this, but it's the way things are for new writers.

[mdin]

The closer you are to 90,000, the better position you're in. But if your work falls outside of the safe zone, I wouldn't give up hope. Granted I've never had a book published by major New York publisher, but I have sold two books to legitimate small presses. The first one was 155,000 words and the second one was 50,000. My current project will hopefully land right at 100,000ish.

[Maryn]

Navigator, you're living proof of the law of averages at work!

And my kids whined that real life doesn't use math. Harrumph!

Maryn

[katdad]

Okay. How about this: The words are less important than the number of characters.

So, if your agent asks you how long your book is, you say "375,000 characters"?

I have never heard an agent or publisher ask about character count. Oh, well.

[Kasey Mackenzie]

I don't think that's what HConn means--though I could be wrong. I think this refers to the fact that publishers care more about the actual SPACE that the characters/words are going to take up when setting them than they actually care about how many of them there are.

But it wouldn't be the first time I was wrong if that's not what HConn meant. =)

[jules]

My understanding of the situation is that the most important thing is the amount of space the writing takes up on the page. This is usually counted using an estimated word count, which can be calculated by counting the number of characters you can fit on a page in your format (with no spaces or blank lines) and dividing by 6, then multiplying by the number of pages you have. This is the figure that usually comes out around 250 words per page (the number of actual words on each page is almost always less than this). I hear that this figure is the most useful one for estimating the number of pages in the eventual book, because it is unaffected by writing style (e.g. average word length, paragraph length, ratio of dialogue to narrative, etc.), while an accurate word count from your word processor is (a 70,000 word book with a lot of short paragraphs and dialogue is going to be a lot thicker than a 70,000 word book with long involved narrative paragraphs).

[Jamesaritchie]

I copied this message from the old forum because the original thread is GONE...

Okay, I just did a little research work and testing.

I took my novel and first deleted all the 'chapter 23' stuff, all the footers, title page, and everything except the actual novel text itself.

Next I did an MS-Word count, using Word 2000. I tried changing the fonts and styles and page margins and it always gave me the same count: 67,252.

I then saved it as a pure ascii text file, opened that file into an ascii text editing program, saved it again, and re-opened the text file into MS-Word. Count: 67,257.

So apparently the process of converting to pure ascii and re-opening as a Word file added 5 words, probably the result of hyphenated words. (I don't hyphenate words on line break, of course).

Next, on my older computer, I re-opened both these test files in Word 97 and ran a word count. 67,252 & 67,257 respectively. So, MS-Word 2000 and Word 97 agree perfectly.

Finally I re-opened the two files in Corel WordPerfect-8, and the new count was 67,206 and 67,208. So WordPerfect apparently low-counted the document by about 50 words, a disagreement of less than 1%.

And then, I took two random pages from my novel, one with lots of narrative, one with mostly dialog. I counted the actual words and then cut & pasted these pages, performed an MS-Word count on them. The physical manual count and the MS-Word count agreed exactly:

Page 51 = 302 words
Page 149 = 152 words

So here's my somewhat scientific conclusion: MS-Word performs a very accurate word count, accurate to less than 1% error.

As I see it, there is no validity to the idea that counting words or spaces and dividing or whatever is better than using the software. Those methods are only estimates, while MS-Word actually counts the words themselves.

If anyone has information to challenge this, please let me know. In the meantime, I'll stick to MS-Word's methodolgy.

Ms Word does count words very accurately, but word count isn't what publishers are looking for. Ms Word counts "I" as a single word, and also counts "encyclopedia" as a single word. Trouble is, "encyclopedia" uses more paper than "I."

Paper count is what publishers are concerned with, and no matter how accurately a word processor counts words, it still can't give a paper count.

And look at a page at the end of a chapter. It may have only two or three sentences on it, but it still takes a full sheet of paper to print that page. This means, by a publisher's count, that page still has 250 words on it, even if it has only thirty actual words.

There was never any doubt that any word processor can count words accurately, and if actual words mattered, that's how the count would be done. But what matters is how much paper the publisher will have to use to print the book, and the easiet was to know this is to use Courier 12, one inch margins, 25 lines per page, and count every page as having 250 words, no matter what the word processor tells you.

[katdad]

I do understand completely how pages are used in printing and estimation. I've done ad copy layout as well as brochure & flier creation, and understand why page count is significant to publishers.

But at no time have I ever been asked by a prospective agent (I'm now signed with an agency), or by a newspaper, magazine, or book publisher "How many pages is it?"

I've always been asked "How may words is it?" and I've always been told things like "Give me 2500 words" by my editor. Never "Write a 3-page article."

The only exception was when I was writing copy directly for a newspaper, and was asked for a "half-column" or "three columns", etc.

That being said, maybe my experience in the past 20-30 years is skewed. I can only speak for myself. But in the meantime, I'll put things like "68,500 words" on my synopses and let the publisher calculate the page length. They know the precise layout for their print venue and can do it far more accurately than I.

By the way, in a recent posting in this forum, there was a breakdown of story length, by the word, to indicate what a short story, novella, novel, etc. were. Nobody listed page count and nobody complained that the length criteria were given in word length.