Mathematical Representation of a Good First Chapter

Posted by Unrepentant Escapist

February 13, 2012 -- 2:28 p.m.

I ditched homework and went to LTUE this weekend. It might have been a mistake for my GPA (I completely failed to put together proper spider charts for my linear programming class) but it was still fun. Mostly, I listened to authors reading excerpts from their work because I thought the opportunity to network with other authors in a less-crowded format would be a better career investment than hearing more from panelists that may or may not be specific to me as an individual.

The highlight of the reading for me was, surprisingly, hearing Natalie Whipple's excerpt from her upcoming YA book, Transparent. Very good stuff. I liked the other excerpts I heard, but for some reason, this one just hit all my buttons. Maybe it was the mention of having chocolate-smelling breath as a super-powered mutation :)

I sat down afterward and tried to figure out why it engaged my interest. Usually, when I read the opening of a book I like, I've had it in front of me so I can point to specific paragraphs I like. But Transparent isn't due out until 2013, so I couldn't do that. But in a way, that makes things better, because certain things stuck out in my memory, making it VERY clear what hooked me.

Since I was also focused on my upcoming data analysis (statistics) midterm, I automatically created an interest equation that represents my own personal hook regression. In other words, this is the mathematical representation of whether the first chapter of a book hooks me or not. Pretend that I know how to make blogger do Greek letters coefficients:

B1symapthiccharacter+B2UpFrontConflict+B3CoolSettingDetails + u = Y

Where Y is a dependent variable (outcome) which equals my own personal hooked-ness. If you want, you can record it as a binary with a 1 of hooked and 0 of non-hooked, or you can just consider it a continuous number with the higher outcomes equaling greater hookedness.

OR, in less math-speak: having a high Y is good for a book. And the higher the factors inside the equation, the higher the Y.

Let's break down the independent variables one by one.

1) The sympathetic character variable. It's an index made up (roughly) of the following components. Everyone's personal sympathetic character index will assign different weights to each part of the index, but I think, in general, each person will have a significant number of the following factors involved in their unconscious enjoyment of a character:

a. Competence at something cool (in this case, robbing stuff while being invisible). Note: if the character is incompetent/unintelligent, this will end up being the only negative number in the index, since all other variables are strictly positive/zero. And what is one negative number multiplied by x positive numbers? A negative number! In non-math speak, it doesn't matter how awesome your character is in all other qualities, if she's stupid or incapable, I will have a low Y. In fact, I will end up hating the book MORE the higher the other numbers. An author will actually be worse off if an otherwise awesome character is incompetent than if a mediocre character is mediocre.

b. Character's "outsider" quotient. This includes the character's is emotional/unusual physical pain and their relationship to society and its authority figures. So, characters who don't fit in well will generate more sympathy.

c. Character's sex. While I first thought about this as a binary o=male, 1=female variable, I realized that, in reality, it's going to be an index of feminity/masculinity with characters that fall into the masculine stereotype being harder for me to sympathize with. This indicator may be reversed for male readers, esp. if they're into action novels. I strongly suspect that my own personal utility equation involves a positive interactive variable of female*competence, which means the effect of female competence is drastically higher than the components of either femaleness or competency on their own terms.

Other components of the index that weren't as present in Transparency's case (or at least not in as great quantities) but would definitely add to my enjoyment:

d. Character's sense of humor (This was minimally present in the reference of chocolate breath as a super-power).

f. Character's "sameness" w/ reader's self (i.e., how much is this character like me? Is she a bookworm? A gamer? A math geek? I guess there was a little of this going on, but it mostly is absorbed by the "outsider" variable.)

g. Some mystery about character or their history. (Again, I don't think there was much of that here, except that I wondered about the scientific rationale for turning invisible.)

So, those are the factors that go into the sympathetic character index. The variable UpfrontConflict is less complicated. In this case, the conflict was twofold:

1) The first chapter involved a robbery, which included conflict with the people being robbed.

2) The chapter hinted at future conflict w/ the character's father--a promise fulfilled later in the excerpt when the father uses his own superpower to manipulate the protagonist into doing something she didn't want to do.

Probably, the variables should be separated out into internal and external conflicts, but I don't necessarily think the coefficients would have any statistically significant difference from one other, and most of the time, a good author will layer both into the same situation. This would make it hard to detect whether one or the other would be sufficient on its own, because in books where you only have one or the other, you are also likely to have a confounder present (the dreaded variable BADWRITING). Hmm, could you strip that out with some sort of instrumental variable? I'll have to think about that.

The third and final factor going into the overall equation is Cool Setting Details. This is pretty self-explanatory. This is what draws me to genre work. Maybe other people read for the characters, but I'm all: "Give me more about the evolutionary history of that man-eating shrub you mentioned on page 41, BITCHES."

I also think this variable is what distinguishes the medium-level-novice from the expert-level-novice, both of whom are in the pre-published, larval state. People who are past the beginner-novice stage can usually create characters who tick off some of the traits on the character interest index in their first chapter, as well as produce some kind of conflict. But very frequently, they forget the setting details that make me hunger to read more. In other worlds, the intermediate novice will begin their book with a fight scene using an outsider character that has a sympathy-inducing DARK and TRAGIC past. But the focus of the scene will be on the sword-fight, which will be generic enough that it could appear in any book from Conan to Game of Thrones, with a little tweaking.

So, in other words, great sympathetic characters + conflict + good setting details=great hook.

...wow...

DO YOU SEE WHAT GRAD SCHOOL/DATA ANALYSIS HAS DONE TO MY BRAIN, PEOPLE?????