IN A separate browser Window, surf to the url saved in step 1, without clicking on the link in the search results. Observe the page and its url this brings up.
IN the Brave Browser window with the search results, click on the item for which the url was saved in step 1.
NOte that a different page comes up from the one corresponding to the url saved in step 1.
With pooling, instead of closing a connection after it is used, we can keep it idle while it waits to handle another request. This saves the resources required to create a new connection.
I still don’t understand why a redirect away from the target page was occurring only when navigating to it by clicking on a link in Brave search. It must have been caused by something in brave browser intercepting the click event and redirecting to a different url. I suppose it is possible the target page could redirect to a different page only if the referrer field in the http header contained a reference to brave search, but I think the redirect was taking the browser away from the target berore the browser reached the target. I could have checked that, but now the phenomenon is no longer occurring.