Crazy, no. Since it's a search request, it's not weighted as heavily as a panda request. So you'll get away with .5sec, but it's going to kill your workflow elsewhere (ie, you'll get PREs trying to actually accept). But unless you have a HF-based autoaccept, there's no way your manual reaction time is going to be good enough to justify refreshing a scraper every half second. tl;dr Use those page request resources elsewhere.