I like that crazy... I feel like I've done something similar before but I can't remember the details.

By the 5th team you'd probably just be drafting your shortstop and 4 bench guys, though. It might work with 3 teams or 4 teams instead of 5.

And I worry about the team with the first pick having too big an edge, but maybe not.

You could also require a certain minimum number of PA or IP from each team, which combined with a salary cap would force people to use at least some talent from each club.

Not sure if schwarze is reading this, but this would be a cool idea for Round 2 of the 2011 WISC....

An idea I've been kicking around for a while would be a multi-season "understand the SIM engine" theme.  Each season we'd pick an aspect of the SIM that's not well understood (like exactly how much fatigue impacts performance) and design the league rules to test that hypothesis.  By definition, we'd have to limit the  amount of roster variation allowed.  Kind of like what you just did crazystengel with the hitting vs pitching theme. 

The primary object of the theme would not be to win, per se, but to learn more about how the engine works and be able to quantify things more specifically.  But of course, winning would still be a secondary objective...

Ideas might be:
-- Fatigue: everyone uses WIS park; half the teams draft limited PA, the other half draft limited IP, and we see what happens
-- Defense (fielding): everyone drafts the same set of 19th century position players; 1/3 of the teams use 19th century pitchers; 1/3 use early-mid 20th century pitchers; 1/3 use pitchers from the last 30 years.  What's the variation in errors committed?
-- Defense (range): Same idea
-- Defense (double plays): everyone drafts the same set of pitchers, with minimum WHIP (probably 1.3 or higher).  Teams can use different combinations of IF range and we see how DPs vary
-- Hit and run: everyone drafts similar (or maybe even identical) rosters; Four teams set H&R to 0; four teams set it to 1; four teams set it to 2; etc.  But every team sets every player to "steal = 0."  How many SBs are attempted?  How many DPs are hit into?  How many times does Sparky call for the H&R?

Again, for this to work, we'd need a set of owners who were really interested in undertsanding the game, rather than being primarily in it to win.

I don't have time to do this right now, but if my work-related travel ever calms down, I may try it out.

For anyone interested, I've started a thread in the Theme Classifieds on my team draft idea.  Look for the thread titled "CONNECT FOUR."

I'm going to add a few more to this list... I'd like to do these eventually but don't have time right now. So I'm putting them here so I'll remember. If anyone else wants to run with them, I'd be OK with that.

Last name twist (probably 100mil) ... pick a real life team, and twist 3-5 players to other players with the same last name. So the '11 Phillies would turn Michael Martinez into Pedro Martinez, Joe Blanton into Cy Blanton, etc. The other 20-22 players would not be twisted at all. The names would have to be spelled exactly the same. There's no other players named Ruth :) Probably would use something like 1960-present or 1980-present for the starting teams, just so no one picks the '06 Cubs or the '27 Yankees. 

I've had an idea for a progressive for a long time. I even tried to start it up earlier this year, only to have real life cut back my available WIS time to the point I couldn't run a league. The interest seemed to be there, but I'd like to open it up for someone else to run with the idea if they would like.

The basic idea is that it would be a limited span progressive (I was going to use 1980-1989). Each owner takes the opening day roster of a real-life MLB team and acts as though they just took over the GM job and has to see if they can out-manage the real life team execs. Draft positioning would be determined by the percentage better (or worse) each team performed than their real-life counterpart for that particular season. For instance, if your team won 80 games in the sim compared to 70 in real life, you would be +10 and would draft higher than an 80 win team whose real-life counterpart won 90 games. This would completely remove tanking and would incentivize winning otherwise meaningless games late in the season.

The biggest issues I was running into were:

- Is there a reliable way to determine opening day rosters when players played for more than one team in 1980?
- How would we discourage people from dropping after a few seasons if they are doing terrible and getting terrible draft picks?
- Exactly how would we present prizes? Team that wins the WS in 1989? Team that improves most upon real-life?

Anyway, I thought it was too interesting an idea to let die, but I just don't have the time to commit to make it work.

I want to do an Albatross Progressive.   

You have a normal progressive but then there would be a one round Albatross Draft every year.

My next idea is a simple one...

Trade Machine (probably 100mil would be enough). Take any RL team, and trade all of their players for someone cheaper. You could trade an $8mil player for a $7,999,999 player or a $5mil player, but each player has to be at least $1 less. It's tough to verify, since owners would have to submit a list of all 25 players and who they replaced. Maybe there would be restrictions (you can only trade for a player from 1950-present, or only a player in your given league/year, or whatever). 

The Sunset Jimmy Burke League

 I've been kicking an idea like this around for years.

- Everyone has to start with the same player, in this case 1901 Jimmy Burke.
- The next player has to be on the same roster as the previous player - so the 2nd player has to be a teammate of 1901 Jimmy Burke (baseball reference is the authority here.)  BUT you have to use a version of that player from a year later than that player.  Jimmy Burke was picked because he played on three teams in 1901 - Milwaukee Brewers, Chicago White Sox and Pittsburgh Pirates.
- Obviously that 2nd player has to have a WIS db season for the one you pick.  Continue to repeat the process until team is filled.
- No clones. 
- Use a stadium from one of the teams one of your players played for.
- DH in both leagues

I tested out the theory on another 1901 player - p Bill Bernhard.  This is how far I got.
To walk it thru -
Lajoie was a teammate of Bernard on the 1901 A's
Joss was a teammate of Lajoie on the 1904 Naps
Turner was a teammate of Joss on the 1907 Naps
etc.

I stalled out trying to find a short jump to the Yankees.

Big Len used to run something like that sheller...

Wonder if I could talk him into running another one.  It's a heck of a lot funner being an owner than it is a commish.