82games post on ideal fantasy league setup

Interesting post by Eric Wong, the resident fantasy expert at 82games, on the best structure for a fantasy league. His criteria:

Rotisserie. If you're not familiar with fantasy sports, the two other options are points (discussed more below) and head-to-head. I really don't like head-to-head, although points is pretty good because of the flexibility. Rotisserie means teams are ranked in order and assigned points based on where you are in the league: first place in assists gets 10 points in a 10-team league, while last place gets 1 point. This is done on a weekly basis and the highest overall score wins (More on rotisserie scoring here)

No turnover penalty. This is more controversial and we don't adopt it in my league. His argument is that since turnovers are highly correlated with overall skill (e.g. Kobe Bryant and Steve Nash are usually near the top of the turnover per game category), it doesn't make sense to mute the positive effects of these players. So you only count positive categories, not any negative ones.

Auction drafting. This is a new concept to me and I can definitely see the appeal. Rather than a simple ordered draft, each GM receives a fixed amount of money and has to bid to complete a roster. Blue-chip players like Kobe may cost you one-quarter of your entire budget, while the last few players on a 12-team roster may cost the absolute minimum since no one else wants to big on them at all. This ostensibly adds another layer of skill to drafting. Almost everyone knows that picking LeBron overall is smart; but how much are you willing to pay for him? Twenty percent of your entire team budget? More? In a pure draft, there's no trade-off. Just draft Lebron first overall (that's what I did) and move on down the line hoping to get Kidd or Kobe next.

Auction free agent pickups. Similar to auction drafting, each team gets a budget to spend on free agent acquisitions. This adds more skill as well; when Gilbert Arenas goes out for several months, it's the fastest to the computer to pick up Antonio Daniels and his added minutes. When you play across a lot of time zones, as I do, this makes even less sense, since there's no way we'll be online at the same time so it's just a coin flip regarding when the news comes on the wire. In the auction format, teams must bid for free agents on the wire - so how much is Larry Hughes really worth now?

Broad position categories: I really agree with this one, although we haven't implemented it in my league. Rather than separate roster spots for PGs and SGs, just have G, F and C. I would take this even further if I could and just have "big" and "small" as the two categories, but I don't think that's possible, at least not on the Yahoo league where I play. Since many of the dominant players at any position are hard to categorize - Is Iverson a point guard? Is Shawn Marion a four? - broaden the categories to avoid roster juggling problems.

I like all of these ideas. However, the league I play with a friend is quite different. We do points only, standard draft and waiver pickups. We do have turnover penalties, though I'd rather we didn't. There are a few fun but pointless wrinkles, like added points for flagrant fouls and ejections (giving Stephen Jackson and Rasheed Wallace a little extra value), a small bonus for minutes played, and an interesting format that penalizes players shooting less that 40% while rewarding those shooting above it. Here's the full format:

Overall Fantasy Points = 0.1*Minutes + Field goals - 0.4*Field goals attempted + 0.2*Free throws attempted + 0.5*Free throws made + three pointers + points scored + Offensive Rebounds + 2*Total Rebounds + 3*Assists + 2*Steals + 2*Blocks + 3*Ejections + 2*Flagrant Fouls - 0.5*Turnovers

Whew.

What you can see from this format is that high assist/high three pointer players are extremely valuable. A single three pointer is worth 4.6 fantasy points: 1 field goal made + 1 three-pointer made + 3 points - 0.4*1 shot attempt. One three is worth the equivalent of 46 minutes, or a 6/8 night at the free throw line.

All-arounders have pretty great value in this league: Caron Butler, Josh Smith, Antawn Jamison, Ron Artest and Andre Igoudala are all great top-30 player because they contribute in all of the high-value categories: assists, steals, shot-blocking as well as decent scoring.

Jason Kidd is an absolute monster in this league, but Baron Davis is the dominator: he's currently the #3 overall player in the league, far ahead of Kobe, Yao and Garnett.

Here are the top 10 players in this format:

1. Lebron
2. Dwight Howard
3. Baron Davis
4. Chris Paul
5. Carlos Boozer
6. Steve Nash
7. Kobe
8. Kidd
9. Iverson
10. Chris Kaman (!!!)

In the past two years, I've played in two very weird formats. The first was 4 teams, 12 players per squad. In other words, we only played with the top 50 players in the NBA. Now that may not sound very challenging, but it was, for two reasons. First, last season had such an insane run of serious injuries to premium players - a partial list would include Yao, McGrady, Artest, Pierce, Amare, Ray Allen, Gasol, Arenas, Shaq, and on and on - that we ended up constantly juggling our lineups and swapping in second-tier players. Secondly, there was very little margin of difference between the teams, since while one squad have Nash, Kobe and Kidd, the other would have Dirk, Yao and KG.

Still, we decided this was just too shallow a way to play for the whole season, so we doubled it: 4 teams, 25 players per squad. Now we're digging a bit deeper into the lower tier of the NBA: some pretty middling players like Marko Jaric, Rajon Rondo and John Salmons have roster spots in this league. With only two bench spots, there's not really any art to constructing your rosters for the week, since there's room for everyone to play every game. But there is a lot of scrabbling at the bottom, swapping out marginal players for someone hot. We've had 47 waiver wire transactions so far; I've already swapped Ben Wallace on and off my team several times as I tried out Brendan Haywood, Sam Dalembert, and Troy Murphy. Squeezing something out of the wire is essential because the draft has locked up every quality player imaginable, and we're left fighting over Ricky Davis, Randy Foye and Darko.

Fantasy Basketball Rankings, Projections, and Analysis by Eric Wong AKA Roto Evil: ""

(Via 82 games.)

Technorati Tags:
Fantasy basketball, NBA