View Full Version : AuthorEarnings.com: a disrupstor

02-12-2014, 12:37 AM
Hi, folks. It's been a while since I've dropped in here.

Thought you all might be interested in this: http://authorearnings.com

Because the site is currently overloaded, I've also included a link to the entire current text of AuthorEarnings.com, mirrored with commentary on Joe Konrath's blog: http://jakonrath.blogspot.com/2014/02/me-hugh-howey-and-legacy-john-on.html

Enjoy reading!

02-12-2014, 02:23 AM
And yet I still want to trade publish and I still don't want trade publishers to undervalue my work by pushing down prices to nothing.

This is already spreading across my facebook as vindication that the 'indies' were right and trade is a big bad money grabbing pool of mud where poor fool authors are taken advantage of on a daily basis.

02-12-2014, 02:32 AM
Hi Libbie, since you've not been around for a while you may not know that these issues have been the subject of quite a few threads here lately.

Rather than just posting the link(s), would you mind commenting on them also? Give us some sense of why you think it's worth our looking at them, and some sense of what your own opinions are.

02-12-2014, 02:45 AM
I started skimming, and the bad math and poor understanding of statistics was already making me *headdesk* so much that I stopped.

Y'all know how I feel about bad math.

I admit to not RTFA, but it saddens me when an article appears to be trying to fit data to the conclusions they want to happen. It hurts me in my little scientist heart. And it makes me want to disassociate myself from whatever cause* the authors espouse, even if there are other very good and valid things about the topic under discussion. (I'm SPing, for anyone who doesn't know.)

* Tbh, though, I'm totally stumped as to why SP has become a "cause" in the first place, rather than a business decision as to what's best for that author and that book.

02-12-2014, 02:48 AM
Slhuang, would you mind explaining why the math/stats are bad? They felt funny to me (like ignoring the possibility that trade pub readers and self-pub readers are often distinct groups with different standards), but I don't quite have the background to identify what's wrong with the rest.

Liosse de Velishaf
02-12-2014, 02:50 AM
Ooh... A literal straw man. I bet JAK and HH enjoyed knocking him down. Perhaps they plan to burn him in a big bonfire later?

02-12-2014, 02:50 AM
I agree. I got to the first statistic (average review score and average price comparison) and went Whaaaaa? Maybe I missed it, because I literally lost interest and stopped reading, but it didn't look normalized at all. Frankly, I'd much rather have 7,000 reviews that averaged out to 3.5 stars than my mom and my 5 star reviews.

J. Tanner
02-12-2014, 03:06 AM
I'm also interested in reading further about math/statistical issues.

There are certain obvious issues with the generalizations. Like calculating yearly earnings based on two days of data. How many books sell the same a year from now as they do today?

The data seems incomplete compared to the charts. Where is the info about format that is referred to? (Their print percentages listed shocked me, so I tried to find the supporting data.) Why isn't genre listed? Is it assumed that every book has a different author, or are they collated before being anonymized. Etc.

Their isn't enough info on methodology. From checking the data, it seems they make a rank=sales assumption with little supporting data. Why does 1=7000? My understanding is this fluctuates dramatically from day to day. I could see Hugh having enough contacts through the spectrum of the top 7000 SF/R/M genre to get a sense on a particular day but if that's how it was done, I'd like it to be mentioned. It's just a column that appears created from whole cloth that feeds into pretty much every chart.

Regardless of my reservations about this initial report, I can see how improved data collection over time along these lines from the largest online retailer will tell a very interesting story as compared to the Bookscan data most commentary has been centered on in the past. It's the first attempt I've seen at all that tries to come at the issue from the point of data analysis rather than self-selected survey and that's commendable even if some of the charts and conclusions from that data are suspect. But the data is being made available, which means others can crunch it too.

02-12-2014, 03:32 AM
Hi folks,

Since the OP didn't stick around to offer her reasons for starting the thread, and since everyone seems to find big problems with the math and handling of the statistics referred to, and since as I posted upthread we've had numerous discussions here lately on this issue, I am locking this one for now.

If people would like to open a thread to discuss how best to collect and offer useful statistics about sales and publishing methods, please feel free. If you begin the thread with a link to another article, please include also some comment as to why you're sharing it here, and what your own opinion is.