This thread is only partly on point but has some interesting conclusions:
http://forumserver.twoplustwo.com/sh...14&fpart=1
To summarize, Jackpot Jay once quoted Lee Jones as saying that only 7%-8% of online poker players end their year in the black. In most PT databases, however, around 40% of ring game players are winners (this number apparently holds up over time.) Therefore, the number of overall losers must be greatly inflated by otherwise breakeven or winning ring game players playing a losing tournament game. In fact, 'greatly' may be an understatement, because from 40% to 8% is a huge jump. [Edit: I should mention that the fact that 40% of the players are winners in PT means that the actual number is smaller, due to statistical sample size issues. Nonetheless, it is clear that the number of tournament winners is much smaller than the number of ring game winners.]
I suspect this is true mostly due to MTT's rather than SNG's (where the 109 and 215 regulars are numerous enough, and the play at the 11's is soft enough, that the 8% number probably cannot be right). A good example of why this is so is the Empire Happy Hour, which is a slightly +$EV tournament (the occasional overlays more than make up for the times they actually collect the $1) that's got more than its share of 2+2'ers. I haven't played nearly enough of them to get an accurate ROI, of course, but I have a large enough sample to know I'm a big winner in that tourney, along with 4-6 other 2+2'ers and a couple of good regulars. The other 90%+ of the field on a day to day basis, however, are clear net losers, and this is in a tournament that averages a slight overlay.
I will go on to speculate that in a normal tournament field, with 8-10% rake, less than 5% of the field are long term winners.
Discuss.