I agree with Mr. Winberg. Money winnings emphasizes to the donators, who are the life of your tournaments, that they aren't doing well. Without those players willing to keep coming, your game dies, whether cash or tournament. Fortunately, some (most?) people play for reasons other than winning money. If they didn't, poker would dry up fairly quickly. I try to learn why those people come and provide it for them.
For games I've tracked over the years, I find that 28-35% of players are net winners, even if it's only a small amount; 30-50% in any given year don't cash. In some cases, they've attended very few games. I've found those who attend all the games and never get a cash out tend to quit, and there is no bringing them back. They might do that for a year or two, but eventually they quit. If the measure of how they are doing is just money, those who never cash get 0. If points is what determines who makes the final game, or stacks in the final game, I'd be curious to know how you keep those donators coming year after year. Having said that, we are in our 7 year of tracking all kinds of stuff. Our top player in any year has never repeated as the top player another year. It will eventually happen. Sometimes we've had a top player one year who was a net loser the next.
Few want to admit it, but the money payouts in tournaments is as arbitrary as any other method of measure. All tournament systems are somewhat arbitrary, even if well thought through. For example, EFDenny's formula, 40/25/15/12/8, or stated in percentages apart, 1.6x/1.667x/1.25x/1.5x/1. Why not 39/26/17/11/7? Those numbers are fairly consistent in percentages apart, 1.5x/1.529x/1.546x/1.571x/1. I wouldn't say the first one is wrong, especially for a payout, and it's much easier to figure. However, in terms of quantifying quality of play, I think the second is a better measure. Cash games are easy -- the best player over time makes the most money over time. The best players are going to have more winning days, and their winning days are going to be bigger than their losing days. For example, they may win 70% of the time. On those 7 out of 10 wins, they may win 3x what they lose on losing days. Tournaments have a lot more variance. They are harder to measure.
We use a flatter payout structure, and at lower attendance, pay about 1 slot more than a lot of other games. However, our lowest payouts are smaller, and the lowest usually just gets their money back. Players who rarely cash though get very excited about a cash, and that helps bring them back. Our top 2 slots get at least a little more with each additional attendee, even when we add a payout slot.
I like rewarding them something for coming, but my main reason is I like to compare each player to all other players. If someone gets 0 for coming because they finished low, you can't really compare them until they win something. While we compare them all, our elite performances (this year our top 5) score a lot more. Up to now, I've been scoring the top 7. After December, I'll compare both and see if it matters to determine who the top players are. Maybe I'll decide only the top 5 doesn't work, I don't know. I suspect either way, our top players will still be our top players.
Others only reward the top performances. That's not wrong -- just different. You need to know what YOU are trying to measure. I wanted to measure who played the best over several games (generally at least half + 1 in a given year). I wanted every game to count (kind of like college football). Some throw out worst performances. I don't because I think it skews the results, and we have a relatively short season -- 14 games. For weekly games, I think if you are measuring for a year (50 games), money won is probably a great way to measure who the best players are for most of those games. If that's what you want, match your point system to the money, but don't tell people that. If you post a list of all players, nobody really wants to see their name at the bottom of the list, or even near the bottom of the list for that matter. When you reward participation some, those at the bottom can be told the reason they are at the bottom is they didn't come enough, which sounds a lot better than "you were a horrible player."