I'm on one beta where a group of about 75 of us get the bits about a week before the rest of the testers get them. It's a last chance for any abnormalities to get found out before wide testing can begin. It is amazing that one or the other of us generally finds a "show stopper" right out of the box! I've even found two or three show stoppers myself and confirmed a lot more when we were pointed to them by another tester.
Another point I'd like to make is that sometimes you, I or Joe Bloe will find something we label as a bug. I'll take myself, for instance. Could be/probably is a bug: I cannot get onto the Internet reliably with Build 10041 update; however, it's marginally better with a clean install of 10041. Release of Build 10049 update is somewhat better, but there's still no prize to be awarded.
Scenario 1) I come here and point out the problems I'm having. Several folks get into the discussion and eventually someone says, "Did you do this?" and so I "do this" and the problem is solved (I Wish!!!). Whether the problem is solved or not, if no one gives feedback Microsoft won't know that a whole bunch of folks are having this problem.
Scenario 2) With Build 10041 I cannot reliably get onto the Internet. Build 10049 is somewhat better, but I still cannot get onto the Internet with any regularity. So, I and everyone else who has the problem decide Windows 10 is not for us, uninstall it and tell all our friends Windows 10 won't let you get onto the Internet. Microsoft never hears of the problem, thus, when it's time for RTM, a buggy Windows 10 shows up.
Of course, Scenario 2) will never happen, because the folks here don't test that way. We're all smarter than your average bear.