I read up on and visit a lot of tips and resource sites. Last year (2013) I did see broken links talked about fairly often. Many talked about how having broken links on your website can have a negative impact on your rankings.
I am not a broken link expert, and frankly I can’t think of a way to actually test them for negative ranking. There are just too many variables that come into play that could be a contributing factor.
I recently came across an older standard html (non-PHP/database) type of site that did seem to rank fairly well. I did not spend a lot of time analyzing what it ranked for compared to the competition because I don’t have time to do that for the fun of it. This site had a fairly significant number of links to other sites that were dead, non-existent, or in other words broken. These were on inner pages. I didn’t count them all, but there were plenty. I would estimate in the 25% to 50% range. Two of the inner pages that I examined had a Google Pagerank of 2 and 3, which is fairly decent.
Would these pages rank better if the broken links were fixed or replaced? Possibly, but who knows. Google could change something at the same time and you really wouldn’t know for sure.
I think any site that has been around for a while and has a decent number of articles, posts, topics, etc. is bound to have some broken links. The bigger the site the more they will likely have.
I am more concerned about visitor experience when it comes to broken links than I am with losing rankings. I mean if you write a great article and tell your visitors to visit some link for additional information and it happens to be dead that’s a poor user experience. I know we can’t catch them all, but fixing or replacing them with a suitable working alternative is ideal. I just hate it when I have been searching long and hard for something unique and when I go to click on a link…BOOM! Nothing but a dead non-existent page or website.
Since I hate it when that happens I do try to fix and update any broken links that I have on my website several times per year. I use http://www.brokenlinkcheck.com/. It is free to check up to 3000 pages on a top level domain (main domain ie: yoursite.com, but not yoursite.com/blog/).
What really amazes me about the broken link checker is just how many it finds related to comment links. By that I mean the area where someone leaves a comment on your site/blog and they enter their site in the name and/or CommentLuv field. I swear I find more non-existent sites and CommentLuv links than I would like to deal with or fix sometimes. Why these sites or links disappear is a good question that I don’t have an answer for. I guess maybe they gave up or quit. Maybe they thought they were going to get rich quick, or found out that running a website was more work than they thought. It could be anything really.
I am not as concerned about broken comment/CommentLuv links as I am with in article or post links. I still consider them to be a bad user experience, but in my opinion they are not quite as important to fix as the ones in an actual post. I do fix them just not as often. I usually let them pile up for three to four months before I get around to them.
Maybe you would see a drop in rankings if your site had broken links plastered all over every post or page in large quantities. This isn’t something I have an interest in testing out to see what happens.
My site has been around for quite a while now and at any given time there will be at least some broken links. I haven’t noticed a drop in rankings that could be directly attributed to them.