You can watch the games AND check the stats. No one is required to do both, enjoy the team any way you like. However, don't accuse someone of not watching the games simply because they take the extra step of checking the stats. It is unhelpful and insulting. There is no need for that.
I think people who take their time to post on HF and follow the team do watch the game. However the expression shouldn't be "watch the game", it should be "watch and understand the game". I think many people watch the game, but don't truly understand the intricacies involved. That's where their views on players go wrong. Stats don't lie, but to gain understanding of them they need to be used with the proper perspective and understanding. It's like reading Shakespeare--many people can read Shakespeare, fewer people can gain an understanding of what's going on, and even fewer are able to absorb all of the information and actually apply it.
I think that some people rely on the "eye test" too much (because it's emotional instead of rational), or rely on stats too much (because they're not actually analyzing the context of each of the stats).
The best solution really is to be able to understand the game, understand what's going on, and reference stats as a piece of that puzzle. However, if the stats don't support your conclusion-that doesn't automatically make you incorrect (although it certainly can). There could be another cause for the stats not aligning with your observations.
Correlation =/= Causation (all of the time).
The stats become very useful when you're able to distinguish which stats correlate with causation--and which ones don't.
I think that's a valuable lesson/component which many "stats people" (for lack of a better term), are going to have to define a little more clearly in the future. Just like we all know that having a poor +/- doesn't mean that a player is necessarily poor on defense. I think it'll take a few seasons to iron out all of that information, and it will be valuable once that happens.
I guess my point is simple: you need to watch the game and refer to the stats (I know you've said this hundreds of times), however I would add that we have to also be able to understand the game in terms of on-ice action and what the stats actually tell us about players (while keeping in mind that correlation =/+ causation automatically, as we try and rule out all other reasonable possibilities).
With that said Bailey's a nice player to have, but if we lost him I wouldn't lose a lot of sleep over it.