Last week Matt G tweeted about a quick blog post I wrote about iteratively applying models using some of Hadley’s packages, to which Hadley replied:
@mammykins_ @m_a_upson the only thing it's missing is that you should store related data frames and models as list columns of s data frame
— Hadley Wickham (@hadleywickham) April 2, 2016
Now this blog is pretty new, and we haven’t made much of an effort to promote it. In the time since I transferred from the old domain name (more on this later) pretty much no-one has been looking at it (apart from my mum – thanks mum).
So what happens when Hadley reads your blog post and mentions it on twitter?
Quantifying the Hadley effect
Well, taking the true Hadley Effect to be a single day event ocurring 2nd April (we term the tail of this spike: residual Hadley Effect): it’s around a 7000% increase in unique daily visits.
A bit more data
So this isn’t the whole story. I’ve been blogging since the end of 2014 (although I only signed up to Google Analytics in April 2015), and used to host this all using github pages (actually we are still hosted by github pages: https://machinegurning.github.io, just we bought a err more sensible domain). So I’ve actually got a bit more data with which to evaluate the Hadley effect.
From a quick glance, it seems that not only is a mention from Hadley worth about 3 times as many reads as my most popular article (seems to be this one); but his briefest of mentions convinces people to read more of the blog for longer! – Thanks Hadley
