Reading, Writing, and Arithmetic. Part 3: Arithmetic

The Statistics of Getting Published

13221 titles were published in the US in 1996 (the latest information I could find that broke out fiction from non-fiction).  That covers all of literature, which seems to cover re-issues, children's picture books, graphic novels, collections of poetry and short stories -- everything, in short, that's not non-fiction.  It seems as though it may count each edition of a given work separately.  It was hard enough to find data that broke the numbers down even this much.  I could not find what I really wanted:  how many new novels are published each year in the US.  

Let us call this (unfortunately) unknown number X.  What will it take for a writer's work to become one single part of that X?

I have a couple of anecdotal reports from separate literary agencies that give some consistent data to work with.  Consistency is encouraging, even if the data is not.

The first is from Agent Kristin at Pub Rants.  Kristin's agency has 2 agents and 25 clients.  Each agent gets 50-75 queries every day.  They request partials from just over 1% of the queries.

Literary agent Nathan Bransford gets about 50 queries per day (including weekends).  Again, just over 1% of queries get a request for a partial.

In round numbers, let's say each agent gets 50 queries per day, and requests partials from 1-2% of them.  I'd assume the number of partials that get a request for a full is fairly low -- say 10%.   I'd also assume the number of fulls that result in an offer of representation is about 25%.  These numbers work out to about 4.5 books per agent per year from new-to-them authors (they also have their existing clients providing them with manuscripts, but we're not considering them).

That says that on the high side we're looking at 1 out of every 4000 queries gets all the way to acceptance by an agent.  Seems high to me, but that's what the numbers say.

I'm ignoring the fact that not every book that gets agented gets published.  I think the percentage that gets left on the agent's desk is fairly small (less than 20% -- meaning 80% get to store shelves).  Let's be positive and pretend that doesn't change anything, and that if an agent accepts your manuscript it will get published.

Of course you can't just send your query out 4000 times and get a book deal.  This is not a random process.  The vast majority of queries are rejected because they're unprofessional nonsense.  Another part of them are not in the proper genre for the agent they're sent to.  You won't have those problems, right?  Your query's chances just went from 1:100 to, say, 1:2.  Not a bad payoff for doing your homework.

Next, we'll assume you learned how to write somewhere along the way.  Well.  And you can tell a story.  And it's not memoir (which can be a difficult sell).  So instead of 10% on your partial, you can expect more like 30% acceptance, and 50% for your full.  That gives you 1:4.  Some agent's info will be out-of-date, or they'll already have what they're looking for in your area, and you'll be rejected for reasons you couldn't find out beforehand.  Using a 30% "just because" rejection rate, gives about 1:26.  

An average of 26 queries to get yourself published.  Now that's not bad at all.  That's a lot better than the shotgun 1 in 4000.  Do your homework.

The Statistics of Travel Writing

We've all heard it said that writing can take you places:  flights of fancy, treks through the wasteland of abandoned plots, book tours, fame, fortune.  Aside from literary road-trips, though, it takes your fingers on a journey.  Lets see how far they might go.

I calculated the finger throw distance for each key on a conventional keyboard.  That's key travel (4mm for an average keyboard) for a key in the home position where your fingers rest on the keyboard, including the space bar.  Then I added the extra strokes needed to be off home position (like w or m), and here I assumed the required key was only one key away from a home position, which added another 20mm of travel by my measurement.  Of course this ignores the extra reach for punctuation up above the number keys:  in my manuscript, and probably yours too, these are pretty rare.  Capitalized characters require a separate 20+4mm depression of the shift key, but I didn't account for punctuation that requires the shift key.  Call me lazy if you wish.

Then I calculated the frequency of each character in my 96000 word fantasy manuscript, the round-trip distance the finger(s) moves in striking the key and returning to home position, and the total distance in mm that key required of my fingers over the course of the manuscript.

                       total
   key    freq  mm    travel
   ---    ----  --    ------
   '/'       4  48       192
   'Q'       9  96       864
   '('      29  48      1392
   ')'      29  48      1392
  '\t'      32  48      1536
   'K'      33  56      1848
   'U'      42  96      4032
   '!'      63  48      3024
   'P'      70  96      6720
   'V'      78  96      7488
   ':'      92  56      5152
   'E'     107  96     10272
   ';'     108   8       864
   'R'     130  96     12480
   'M'     151  96     14496
   'G'     160  96     15360
   '#'     163  48      7824       (a single '#' marks my section breaks)
   'z'     195  48      9360
   'L'     196  56     10976
   'O'     247  96     23712
   'q'     249  48     11952
   'J'     261  56     14616
   'C'     270  96     25920
   'B'     331  96     31776
   'j'     391   8      3128
   'Y'     450  96     43200
   'x'     461  48     22128
   'F'     515  56     28840
   'A'     606  56     33936
   'D'     798  56     44688
   'N'     832  96     79872
   'H'     872  96     83712
   'S'     975  56     54600
   'W'    1108  96    106368
   '-'    1235  48     59280
   '?'    1298  48     62304
   'Z'    1463  96    140448
   'T'    1526  96    146496
   'I'    1686  96    161856
   'v'    3001  48    144048
   "'"    3181  48    152688
   'k'    4406   8     35248
  '\n'    4721  48    226608    ('\n' is a newline)
   'p'    5572  48    267456
   'b'    6005  48    288240
   ','    6019  48    288912
   'c'    6645  48    318960
   'f'    7035   8     56280
   'm'    7064  48    339072
   'y'    7696  48    369408
   '"'    8154  48    391392
   'g'    8429  48    404592
   '.'    8542  48    410016
   'w'    9162  48    439776
   'u'   10486  48    503328
   'l'   13561   8    108488
   'd'   19042   8    152336
   'r'   20501  48    984048
   'i'   21609  48   1037232
   's'   22709   8    181672
   'n'   24140  48   1158720
   'h'   25119  48   1205712
   'a'   29857   8    238856
   'o'   31202  48   1497696
   't'   35941  48   1725168
   'e'   50591  48   2428368
   ' '   92836   8    742688 

The total is 17.4km, or for the metrically impaired, 10.8 miles.  That's how far my fingertips moved in typing my manuscript.  It adds-up, doesn't it....

Of course, I didn't just type in my manuscript -- I made mistakes and retyped, backed up and retyped again, etc.  I think it's a conservative estimate that the real number is 3 times the straight-through calculation; say 32 miles.

But it gets worse.

I use a special keyboard (the Datahand) because I have arthritis in my hands.  It is a low-force / low-motion keyboard, and I thought when I did the above calculations that I'd have saved some distance.  Not so.  Instead of the 4mm home-row travel, this keyboard has a consistent 7mm travel for every key, with an extra 7mm for shifts.  Essentially every key is a home-row key, but it's got a travel nearly twice a conventional keyboard.   For this keyboard I get 24.2km, or 15 miles, which gives a total 3x travel of 45 miles.

Writing will take you places, even if it's only 4mm at a time.  So remember to be kind to your fingers.

One more thing -- this, the third part of a series posts, the one on Arithmetic, falls on pi-day (3-14).  I didn't plan it that way.  I'm not that much of a geek.