Every scout/GM I've ever heard talk about his process has said that the vast majority of their evaluation is based on tape. Rarely, you'll see someone do something freakish at the combine that will change the perception of his potential (guys like Walter Jones, Jason Smith, JPP, Dontari Poe, and Bruce Irvin are some examples). And sometimes you'll see a player kill his stock with a bad combine (Vontaze Burfict, Tom Brady, Chris Polk (allegedly interviewed poorly)). But those are the exceptions, not the rule. For the most part, a prospect's status isn't impacted much from the Combine.
I love the combine though. It's kind of like the preseason for the draft.
And I will also say that the combine definitely has it's uses. I'll list some examples:
#1: Field speed is often difficult to judge on tape. Some players are "sneaky fast." I thought Bobby Wagner was solidly fast, but he wasn't explosive fast. Then he runs a low 4.4's. That's really handy for me as a blogger because I know that Pete prefers fast linebackers, but a lot of times it's hard to tell the difference between 4.4 and 4.6 on film unless the guy is really explosive. Having that list of 40 times helped me highlight Korey Toomer on the blog before he became our 5th round pick. Sure, some players are track fast and not field fast, but if a guy runs a really great time, you can always go back and double check to see if there was something you weren't seeing before. Often there is.
#2: Basic measurables such as height, weight, arm length, and hand size. These are really important numbers to know, and schools are notorious for giving out wildly inaccurate numbers, off by as much as 2 inches of height or 20 pounds of weight. Hand size is considered fairly critical to QB success and can't be scientifically judged from film. Arm length is crucial for offensive and defensive lineman, and can be at times difficult to judge from film (I thought Ezekiel Ansah looked like he had stubby arms on video, and then he measured 34"). It's good to have all those measurements taken at one place so you know the measuring was done correctly and officially.
#3: Speed drills help separate top prospects. Maybe you want to know who the better edge rusher is between Dion Jordan and Damontre Moore? That's hard to tell since they don't play side to side. But events like the 40 yard dash (10 yard split), 3 cone, and short shuttle help illuminate which is the more nimble player. That doesn't mean you should draft a crappy player because he's 5% better in gym shoes, but in such an extremely competitive field, something as small as a 5% advantage matters more than you might think.
#4: It provides a centralized area to conduct massive amounts of interviews while drawing relatively minimal attention. JS interviewed Russell Wilson at the Senior Bowl- and it was reported. His interviews at the combine were much more numerous and weren't fully reported on (to my knowledge). Interviews are extremely important and good players have been taken off draft boards for giving bad ones. Teams view draft picks like business decisions, and a player interview really is in effect a job interview. The interview process is a huge reason for Seattle's success- because it helps them find those highly competitive, chip on the shoulder types (Sherman, Baldwin, Lane), and sort out the ones who are not.
Again, the idea that the combine makes or breaks players is essentially a myth, propagated by a handful of such stories every year out of hundreds of participants. That said, it certainly has a ton of use, and I do not think PC/JS's legacy would be as great today without it. It provides very useful information for those who know how to properly use it.