Let's explore this a little so I can get on with the points I want to make.
Internet Pundits claim that X army is bad....
Heard about this one before I hope! You'll get constant, cliche, and boring arguments on either side of this paradigm just about daily on the blogosphere. Just today, I've already seen 3 articles written that touch on this subject. One was an author explaining how XXX army is uncompetitive and two more articles where authors threw in a couple snipes about how ______ army performed well at a tournament, as proof that the "internet pundits" are wrong.
A couple points to clear up here before I continue.
A) I don't think all army books are balanced and do believe that some armies are not as competitive inherently as others. So, you know what camp I'm in. (I do a codex countdown on the subject).
B) Seriously, the camp that keeps saying "internet pundits are wrong" are themselves "internet pundits". Seriously guys, you are just in the camp that thinks they are right in the opposite direction. The "internet" doesn't think Orks suck. YOU are the internet too. I am the internet too, and ALSO an internet pundit because I put my unfounded opinion on this BLOG and Podcast all the time.
(For whatever it's worth, I like Orks a lot! Heart attack! I think some books aren't balanced and Orks aren't one of them... so I guess I'm a third camp of internet pundits?)
First, the entire argument is subjective.
Second, using tournament results as proof of how all books perform is fundamentally flawed. We all know that. Match-ups, luck, players, and on and on. In fact, if we really wanted to see anything even remotely resembling evidence, we would be much better of looking at lots and lots of tournaments over time... you know.... a statistical study.
BECAUSE GAMES OF 40K ARE NON-DETERMINISTIC. :)
Third, it always funny because even though tournament results get used to prove the case that XXXX army does work, the argument almost always looks like this....
"Big GUNZ Tournament was won by Grey Knights BUT Orks came in 3rd! So, as you can see, Orks still work...."
End of the day....
Consistency is what wins GTs
One of the reasons why you see the proposed "top tier" armies (generally considered GK, IG, and SW at the moment... although if you ask me Necrons should be there too) generally coming in First and the non-top-tier coming in 3rd (as the example above) is it requires consistent performance to win in a GT.
Can Tyranids beat Grey Knights? SURE! Especially if you get to play against a GK player who isn't very good, has a poorly thought out list, or simply just a good match-up. Heck, you can still win even if your opponent is awesome with a completely tailored anti-Tyranid list. (non-deterministic....)
Can you beat 7 Grey Knight players, at random, where each subsequent player gets better and has a better list, where the odds are 1-2 of the lists you will face are terrible match-ups for you, in a row? Hmmmmm........ maybe not so much?
Most avid GT goers will tell you that their performance in a GT is directly related to match-ups. Top tier armies tend to WIN GTs more often because they match-up better versus MOST armies.
Whereas a player who brings an army with more bad match-ups, dramatically increases his odds of getting a bad game. Now, you can play through bad match-ups, especially if your opponent doesn't leverage that, but the longer you win in a GT, the harder this gets.
Difficulty of match-ups generally scales as you win because the opponents get better, their lists get better, and the potential for a non-top-tier list to be abused also grows because of that.
At the end of the day, the only point I'm really trying to make here is that the performance of your army should not be judged in one-off games but in if it can consistently win progressively harder games over 6-8 rounds. That's what makes a GT winning army.... not an army that comes in the Top 10.
BUT WAIT THERE'S MORE!
Lists which meta well also perform well. One thing I've noticed, but haven't bothered to prove is that good players with "off" lists seem to win a good bit more than good players with "standard" looking lists.
I'll cover this in another article, but I wanted to at least bring up the idea in this article that lists which can "screw with" the standard meta consistently perform well for obvious reasons as well.... at least until they run into another list which is attempting to do the same thing as them.