In theory I don't like it. In practice, I have done it this year (Abreu, Burrell, Lieberthal and Helton, Walker, Payton). And I've never really felt like it didn't work. But as someone mentioned, a rainout or two could really hurt, or perhaps bad matchups could impact you more than normal.
I'm still working on swinging some trades to diversify a bit, but I have to agree with the consensus, it doesn't really seem to matter.
I would theorize that having players on the same team hurts you more in H2H leagues than it does in Roto leagues. Since games are made up in Roto leagues and slumps play themselves out much better, a poor run by a team (affecting it's players) won't kill your stats for a week, leading to a loss.
i think it's problematic esp. if you pick up chunks of AL west teams or NL east teams.
i had soriano, jeter, johnson & mondesi at one point and really hurt when they ran into a buzzsaw sqad.
that said, it worked many of the times. i've concluded though, you don't want lots of rangers or braves b/c they're going to finish the season against the phillies or the a's and 3 days of getting burned by mulder-zito-hudson may lose you the year.
diversification is better as a general rule.
lots of jays, stros or cardinals is pretty safe but variety is better.