One thing I'll say is that reviews aren't really that big of a traffic draw and really the money is made in interesting op-eds, guides, exclusive features etc. Because most of the time review scores get pulled into aggregates or posted in short-form with a score to places like GAF, you actually don't draw a lot of CLICKS out of reviews unless you're lucky enough to be one of the top-three that charts on google when somebody punches in "[game name] review". Anyway, my point is - if you think reviews are making or breaking the revenue of gaming sites, you couldn't be more wrong, basically.
Aside from that... A good wide review spread and a good critic average can, I think, drive word of mouth and kick-start it before launch, however, and Prey probably needed it. Streaming and YouTube is great, but it's also worth pointing out it can't have that immediate 'at a glance' impact on a person's opinion, and so because of that I do think traditional reviews have 'a thing' they can do that streaming etc can't. To get something out of a stream it takes 5 minutes at least, probably more. To get something out of a slew of reviews it takes typing "Prey Reviews" into google and seeing, right at the top of the page, an 80% MC average presented in big bold text (with metacritic proper the first and second result on the proper 'results' page). That takes 30 seconds.