Effect of "bad" players
Years ago I ran a series of sims with 6 players at a 6-deck table, Atlantic City rules at the time (DA2, DAS, Split out to 3 hands, late surrender, hole card dealt but not checked until the end of the round).
The purpose was to see if there was any way to benefit from the like-card bunching which existed in the now-obsolete "stutter shuffle." Thus, in the first sim, all 6 players were mostly playing BS, but making the certain kinds of "mistakes" or "bad plays."
The result of this first sim was that the first player benefitted significantly, the second player was about even, and each player did significantly worse than the one to his right. On a graph, in terms of money won/lost, it was virtually a straight line down from the 1st baseman to the 3rd baseman.
So I ran a second sim, with only the first 2 players making the "mistakes," and the other 4 players playing perfect basic strategy.
The result was that the first 2 players did the same has in the first sim, but the other 4 players all did about the same and all did significantly worse than they should have playing pure BS.
So I ran a third sim, with only the 3rd player making the "mistakes," and the first 5 playing perfect BS.
The result was that the first 5 players did exactly as well as they should have playing BS, and the 3rd baseman (to use an unscientific term) just got *killed*!
My conclusion, which I have used successfully *many* times in real-life situations, is to keep "bad" players to your left.
Bad players to your right DO hurt you! Bad players to your left, even in the assumed-to-be-crucial "anchor" spot, do NOT hurt you.
I know this post is going to draw a lot of fire and riducule from the math purists, but I say, "ignore them."
Instead, the next time you go to a casino, don't play at first -- just watch what goes on at a table with 1 or more bad players and keep track of the effects of their mistakes.