Had an interesting email exchange with Tim Bray, who apparently believes that extrapolating from his weblog’s referrer logs is a reasonable way to estimate search engine market share (i.e. percentage of total search queries handled).
Discuss.
(Special prizes for the first readers to identify any of five ways in which this estimate could diverge from reality.)
Here’s a study with a larger, more diverse sample, though admittedly it lacks the transparency of Tim’s work. (Tim was kind enough to actually post his log analysis script.)
Tim Converse 
1.) different algorithms scrub link quality differently. his site is no doubt well linked, with lots of high quality links. due in no small part to adsense there are billions of crap links out there, and currently I think Google is a bit better at scrubbing link quality than Y! or MSN is.
2.) different algorithms place different ratios of relevancy on off page vs on page factors.
3.) different engines tie in / appeal to different userbases. young geek guy will be more inclined to search for information about search is more likely to use Google than other engines.
4.) ad placement in the search results… Yahoo! and MSN place ads more prominently than Google, but of course nowhere near as out of hand as Ask Jeeves was for a long time
5.) whatever I write about and promote my users are more likely to use, since I am promoting it. my bias is easily read in my words.
6.) search engine parnerships. people who use Adobe search Yahoo! a bit more (since there is that partnership). people who download WinZip get a google toolbar bonus. people who read Washington Times frequently got Google, now they get Yahoo.
7.) browser interface. Firefox uses Google as the default. IE will soon use MSN Search in every nook and craney. I think default IE errors send you to MSN
please send lots of prizes…