As the co-founding CEO of probably the world’s fastest WordPress hosting company, I have seen some crazy crazy stuff over the years: pages generating 500+ requests (Amazon.com is around 200, by the way), 200+ (honestly!) plugins on a WP site, updates not done on a site for many years and being constantly hacked and 10+ ad networks all serving ads at the same time to a page, making it crawl like in wet cement.
You get the idea.
When it comes to websites and website speed, one inescapable truth emerges:
"the most powerful computer in the world will struggle to shift a horrible website"
My favorite testing method to quickly resolving this – especially with WPX customers – is to install an empty default WP site into a test subdomain as the hosting account where the problematically slow site also lives: same server, same specs, same overall server load and same hosting account.
For example, if I want to compare how the configuration of my personal blog at terrykyle.com is impacting page loading speed, I need a BENCHMARK of what a hosting server is capable of with an empty, almost plugin-free site first.
Then I can COMPARE and instantly rule in/out whether there is a hosting server issue (lame host probably) or something that I did as the webmaster that has screwed with the optimum performance of the site.
Though there are big flaws in these tools and the way the results are interpreted – I wrote about it here and here – these are the results for terrykyle.com on Pingdom Tools (Washington DC) and GTMetrix (Dallas):