What's The Deal With Google & Its Changing Results Counts?
Ever wonder why Google sometimes gives you more results when you do a query that should bring back less, such as when you "subtract" a term? Here's the answer, courtesy of Twitter exchanges.
another reason why i've long argued Google should have result counts that add up. if you don't, it's conspiracy time! read.bi/90inNR- I've defended Google plenty over the years, as well as criticized them, but fair to say -- I wouldn't be counted as an enemy of Google, by your logic.
So I did a search for ["danny sullivan"] and got 193,000
results. Then I did ["danny sullivan" -gerbils] and got 1620,000
results. That shouldn't be the case. I shouldn't get MORE results when I'm searching a previous set of results and asking for a subset without a particular word.
That would debunk your conspiracy, I'd say. It doesn't debunk that Google is screwed up with its results counts. Do a search for "25 Things I Hate About Google." this was number one on my list, in an article from 2006. So not only isn't this a conspiracy -- it's a long-standing bug that they ought to fix - turns out, even would-be "friends" of Google are subject to the "query sabotage" conspiracy read.bi/b2tSut
- Hi Eric, the answer to the headline's question "Is Google Guilty Of Deliberate Query Sabotage?" is no. We've talked about the fact that results estimates are just estimates for years, see e.g. http://video.google.com/videoplay?docid=-4814548594071648913#
As to why the query [A B -C] can return more estimated results than [A B], that's easy to explain. The query [A B -C] causes us to go deeper through our posting lists looking for matches, which can lead to more accurate (and larger) results estimates. Other things can cause us to go deeper in finding matches, such as clicking deeper in search results.
Results estimates can also vary based on which data centers or indices your query hits, as well as what language you're searching in. It certainly has nothing to do with whether you're a "possible enemy of Google," as you put it.
We try to be very clear that our results estimates are just that--estimates. In theory we could spend cycles on
that aspect of our system, but in practice we have a lot of other things to work on, and more accurate results estimates is lower on the list than lots of other things. 
I took 5 minutes out of my day to nip a potential conspiracy theory in the bud: goo.gl/UMnZ @fromedome @dannysullivan- @mattcutts now if you could just fix the counts or remove them. to a layperson, it makes no sense that subtracting terms increases counts
- @dannysullivan people have talked about removing results estimates altogether. I'm not a fan of that. Still useful, even if noisy.
- @mattcutts add a "about this number" link next to counts, link to explanation, or make the counts a link. problem largely solved.
- @dannysullivan Not worth the pixel real estate on serps and annoying every user on earth. Better to debunk the yearly conspiracy theory.
@mattcutts if the results estimates aren't accurate to an order of magnitude, why show 3 SF? Why not show an error estimate eg 1,000-5,000- @kevinmarks 3SF is tradition that's just stayed around for the last several years. Really depends on how deep we go to find answers.
@mattcutts, can you clue me in on the overall purpose behind showing the counts?
@mattcutts accurate URL count or trend in web tools wuld be awesome. Especially when trying 2 clean up a large site with lots of issues- @mattcutts i know its tough. but if the counts are effectively meanless, don't cite them. geez, even supreme court has used them. scary.
- @dannysullivan or maybe it's just that result count estimates are a really poor thing to base theories on?
- @dannysullivan they're not meaningless. [A B -C] has been known for longtemps, but estimates are stateless across queries; not worth trouble
- @kevinmarks agreed. and I almost said that, too. but they ought to fix the bug. that won't stop conspiracies, though.
@mattcutts @dannysullivan results estimates are crucial, they help prove that google is truly indexing most pages on the web- @mattcutts @dannysullivan cont... And that google are choosing from a large pool of possibilities. It's important for search accountability
@mattcutts Agree, but relevance ? remains: If u know a subset of serps is always <= the master set, then u know subset count is always less- .@alleyinsider, you should also have mentioned Eric Clemons' conflict of interest: goo.gl/myDr *cough* *cough* ;)
@mattcutts Dude! your tweet about Eric Clemons caused me to just spend half my day reading an article on your site from a year ago
