kudos distribution

asherdashery replied to your post: mooeena reblogged your ph…

Oh my god how do kudoses fall out please I need to know

Anonymous asked curlicuecal:

How does Dave ship fic kudos fall out?


I know, it’s shocking.  *ahem.*  Anyway, these are just the graphs for the overall distribution of kudoses among Homestuck works on AO3.

Graph number one, regular scale:

Yes, that blue line tracking the axes really is trying to tell you something about kudoses.  It’s ridiculous, I know. 

Basically:  Along the horizontal axis we have works ranked into percentiles by number of kudoses received. (So the top percentile of works would be from 0 to 1, and the bottom percentile would be from 99 to 100.)  Key take away message:  Works in the top 1% of fics are receiving a shit-ton more kudoses.  (Yes that is a scientific measurement.)  ((Actually it’s like half a percent receiving 500 or more kudoses.))

But that top half a percentile is really swamping out a lot of the picture, so let’s put this on a log scale:

Okay, way prettier.  So from this graph I’d like to point out that somewhere between a quarter to a third of homestuck fics receive 10 or less kudoses.

Also in general, you can see pretty standard logarithmic growth (that’s the straight part of the line in the middle) which means that for every kudos you get you’re a set amount more likely to get more kudos….except for at the very lowest and highest rankings.  There’s a little pile up of fics with no kudoses in the bottom percentiles and then a HUGE jump up in the top percentiles.  (There are probably specific fancy terms for these effects but damnit Jim I am a scientist not a mathematician.)

Anyway, this indicates a tremendous feedback effect where getting enough kudoses enables you to get even more kudoses and so forth to a way way way greater degree than at any of the lower rankings.  I call this the “being listed in the first few pages of a tag” effect.  (The top 1% of homestuck fics take up 12 pages.)

If you wanted to be really ambitious you could calculate the area under the curve of that first graph and get an idea of how total kudoses are distributed.  Just eyeballing it I’d say that more than half of all kudoses given to homestuck works are given to less than 5% of the works.  (It’s probably more like 1-2%?  WE ARE THE 98%.)

But fortunately kudoses aren’t really a limited resource or anything, SO.  I guess we don’t need to start a revolution or anything.

you guys type fast

I got enough responses to my previous post about typing speeds that I decided to make a graph.  (thus surprising no one.)

Major finding: tumblr users are speedy typers!

(The bottom graph is modified for comparative purposes from this study.)

There were 61 respondents, 1 was excluded from the study because I refuse to believe you survive on the internet with a 1 wpm typing speed.  Also, I’m calling this a study now because SCIENCE. 

100% of people in this study type faster than the population average. 27% of tumblr respondents type 100 wpm or faster, putting them in the top 2/10ths of a percentile of the general population (cite).

The average typing speed for tumblr respondents was 90wpm.  This was both the mean and median and compares to a population average of about 30 wpm or about 50-70 wpm for people who type regularly in their occupation. 75 wpm is a common cut-off score for people applying for typist positions.

One potential bias in this study is self-reporting, which could skew the scores higher because of things like pride and greater interest and I guess it could be viewed as less costly for faster typers to type out responses to requests for typing scores.

Thanks to everyone who took the time to take the typing test and report back, and thanks especially to Roach for the signal boost.