Humble Bundle payments: Just because you use open software doesn’t mean you’re cheap

Humble Bundle V payment breakdown stats

Humble Bundle V payment breakdown stats

Just purchased the Humble Bundle V indie game pack, which is notable for being one of their best deals yet, with such star games as Psychonauts (Double Fine) and Sword and Sworcery (Superbrothers) among the collection.

Even better is that all the games are cross-platform, which is great news for Linux and Mac users including myself. I think this has been the case(?) for past bundles, but this is the first one I’ve actually bought.

At the bottom of the downloads page is a chart showing the revenue breakdowns by platform. Bundle V has surpassed $3.2M in payments. As expected, Windows users account for more purchases than Linux and Mac users combined. But check out the average payment by platform; I’ve added the percentage over/under the average for reference’s sake:

Average (+/- average)
Overall $7.87 0.0%
Windows $7.27 -7.6%
Mac $9.46 20.2%
Linux $12.24 55.5%

Looks like the Linux users are using the money saved from free software for a good cause…and the Mac users, I guess we just like to spend extra 😉

[insert pithy statement about how a spirit of generosity and sharing begets more generosity that I’m too busy playing these games to write]

* Edit: As an astute HN user points out, the Linux average is likely skewed by Notch, who apparently chipped in $9999.99 and presumably did so from a Linux system. His outlier effect is further exaggerated given that the Linux userbase is relatively small.

What the Windows userbase has going for it is that it presumably has more game developers, who you think would be more sympathetic to Humble Bundle’s pro-developer ideals than the average gamer. However, it’s unlikely they make up enough of the overall Windows userbase to have a large effect on the average.

(as for the Mac users, they probably really are more affluent or are more used to paying a premium for most software – see the anchor effect.)

This is yet another reminder why you should use the median if your other option is a simple average.

I'm a programmer journalist, currently teaching computational journalism at Stanford University. I'm trying to do my new blogging at