I've been monitoring BTCGuild Hall of Fame for over a month now, and it seems to me something is not right.
There have been many discussions about predictability when it comes to hash required to solve block, and
even though they all ended up with "proofs" that it's impossible to predict a hash, this table clearly shows
different picture. Users colored in red are not even listed on top 25 fastest list, but it
somehow happened
they solved blocks every difficulty for 1+ month. Also check users colored in blue, (dark) green and yellow.
Now, I don't really care if you think I'm wrong and start coming up with "proofs" one more time. The purpose
of this post is to let developers and think-tanks realise the issue and eventualy fix the problem, which clearly
points that attacker does not need 51% but noticably less hashrate share to overtake the Bitcoin network.
You simply misunderstand the statistics. The "Blocks Found" is a life-time figure. Those blocks could be found long time ago and these people may have stopped mining in BTCGuild. The only thing you should look at is the number of block found in current difficulty, which is shown in the bracket. Also, you make a mistake in filling in color for the 54578.