Some stats on the first 50 posts

We have just started the 18th month for this blog, and this post is to celebrate this anniversary and the passing of a milestone: this is the 51st post! It is also to share some statistics about the blog, partially because I wish other bloggers would share theirs (as benchmarks or aspirations for people like me that enjoy metrics too much) and so that I can quickly refer to them later.

When I first launched this blog, I had the ambitious goal of having two posts a week, but in the back of my mind kept the more realistic target of creating one post a week. Unfortunately, even the modest goal has not been met with an average of ~2.9 posts a month. Most importantly, it has been hard to keep a regular schedule with large post-less gaps like 2011/10/13 – 2012/01/10, 2012/03/29 – 2012/05/13, 2012/07/23 – 2012/10/11, and the most recent 2012/12/04 – 2013/01/23. How do you keep yourself on a regular blogging schedule? How do you balance blogging with things “that pay the bills” such as school work and research?

Monthly viewership statistics for TheEGG.

Monthly viewership statistics for TheEGG.

Surprisingly enough, it was during one of these gaps that I realized that this blog can actually reach people. In the first 9 months of activity, the blog received a total of 1806 views. In June and July, I started cross-listing more heavily on researchblogging and the viewership jumped up to 996 (June) and 1108 (July). On August 23rd, somebody shared on Reddit my July 23rd Programming Playground post. The share resulted in a flood of 2146 views on that day alone, and brought August up to 4628 views. This is a significant fraction of the 16751 views that the blog received so far.

After this realization, I started promoting the blog on Reddit, although never as successfully as this first share. The self-share that generated the most traction and insightful comments was my critique of Chaitin’s “Proving Darwin”. This was the one post I noticed circulating through twitter, G+, and tumblr. However, most of the interest seemed to stem from the novelty of Chaitin’s book rather than my critique.

Overall, in terms of sites driving traffic here, Reddit leads by a long shot with 5146 views, followed by search engines with 2707, and then a more tightly spaced list: researchblogging (251), Facebook (240), Twitter (245), and scoop.it (137). That being said, I don’t really understand how WordPress collects these views statistics. For instance, if I was to trust researchblogging’s view statistics then they say that they drove 5514 views to TheEGG, with some posts having more clicks from researchblogging than total number of views (from all sources) recorded for those posts by wordpress. Unless 19/20ths of the people get lost between clicking on a link on researchblogging and arriving at WordPress then there is some big discrepancy with how the two sites record views. Can anybody more familiar on web analytics fill me in?

In terms of WordPress viewership, the top 5 posts are:

  1. Programming playground: A whole-cell computational model (2863)
  2. Is Chaitin proving Darwin with metabiology? (2195)
  3. How would Alan Turing develop biology? (1319)
  4. Marcel’s Generating random k-regular graphs (1012), and
  5. Marcel’s Spatial Structure (559)

On the other hand, the researchblogging stats tend to have less spread and correlate more closely with age:

  1. Tom’s Fewer Friends, More Cooperation (485)
  2. How would Alan Turing develop biology? (418)
  3. Can we expand our moral circle towards an empathic civilization? (397)
  4. Bifurcation of cooperation and inviscid ethnocentrism (385)
  5. Is Chaitin proving Darwin with metabiology? (331)

The primary goal of this blog remains as a way to foster collaboration, and as a companion to the EGT reading group that I host. The reading group is resuming next week after a hiatus. If you have suggestions for what to read, please email them to me or leave them as comments. In terms of collaboration, I am happy to say that 22% of the posts on the blog thus far have come from my colleagues and co-authors: Julian Z. Xue (6 posts), Marcel Montrey (3 posts), and Thomas R. Shultz (2 posts). I am very thankful for their contribution and encouragement, and I hope they will continue to participate in this blog. In the coming months, I am also trying to recruit some new faces with four potential new contributors expressing interest. If you would like to write a guest post about evolutionary game theory, mathematical or computational approaches to evolution, or agent-based modeling in general then let me know!

I look forward to the adventure of the next 50 posts with you.

About Artem Kaznatcheev
From the Department of Computer Science at Oxford University and Department of Translational Hematology & Oncology Research at Cleveland Clinic, I marvel at the world through algorithmic lenses. My mind is drawn to evolutionary dynamics, theoretical computer science, mathematical oncology, computational learning theory, and philosophy of science. Previously I was at the Department of Integrated Mathematical Oncology at Moffitt Cancer Center, and the School of Computer Science and Department of Psychology at McGill University. In a past life, I worried about quantum queries at the Institute for Quantum Computing and Department of Combinatorics & Optimization at University of Waterloo and as a visitor to the Centre for Quantum Technologies at National University of Singapore. Meander with me on Google+ and Twitter.

8 Responses to Some stats on the first 50 posts

  1. Terrific T says:

    Thanks for sharing your blog stat. For a new blogger like me this is really helpful. While I am not yet brave enough to share my blog stat, hopefully I will be able to do so in another 15 months.

  2. Charles Goorsky says:

    Hey keep up the good work, here’s to 50 more!

  3. neuroecology says:

    Relevant: http://studiotendra.com/2013/02/05/33-observations-on-the-year-2012/

    I’m curious: do you have a twitter account that you use to promote the blog? And do you post your own blog posts to reddit (or is that considered spam)?

    • I used to have a twitter account which synthesized all my online presence (stackexchange, blog, G+, and link sharing) and it was very useful for meeting other researchers (I was surprised, actually) and engaging with authors. It was a great experience, but I closed that account for personal reasons, and so don’t use it anymore. However, I can’t say that the account was for promoting the blog, since most of my tweets were just about papers and articles that were not on this blog. For some reference, in November I posted on G+ thoughts of my first 1000 tweets.

      I do post some of my posts to reddit, and it increases traffic significantly and sometimes generates interesting discussions. If you only post links to your own blog, then it is usually looked down upon. So it is a good idea to occasionally post random cool links you find, as well. I usually post to interesting blogposts I read, such as yours. However, I tend to have too many links to my own posts and too few to others, so I would probably be a self-promoting spammer in most redditors’ books. This is mostly because I don’t have the time to hang out on reddit.

  4. Pingback: Stats 101: an update on readership | Theory, Evolution, and Games Group

  5. Pingback: An update | Theory, Evolution, and Games Group

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.