Yeah, a commitee rather than than the whole league. I say start with BBM per game ranks, adjust for 9 cat HtH, and then shuffle a little to account for bounce back. Then just average out the three lists and that's the league list. Easy.
Heres a list I generate with the following method:
1. Use BBM 9-cat per-game predictions for 2011/12 2. Run the predictions 9 time. Each run has one of the categories 'punt' box ticked. Combine the results from each run into a single spreadsheet. 3. When all 9 runs are in a single spreadsheet, sort the whole thing by value. 4. List the players for each position they are eligible, using their highest value on the consolidated list.
I've cut the lists off at values of 0.20, for reasons I'll explain latter. Here's what we get...
C 1. Howard (1.10) 2. Pau (0.65) 3. Amare (0.62) 4. Love (0.59) 5. Horford (0.56) 6. Jefferson (0.52) 7. LMA (0.46) 8. Ibaka (0.44) 9. Brand (0.43) 10. Gortart (0.38) 11. Nene (0.38) 12. Bynum (0.38) 13. Bogut (0.37) 14. Lee (0.36) 15. Monroe (0.34) 16. B.Lopez (0.33) 17. Noah (0.32) 18. McGee (0.31) 19. Frye (0.30) 20. Duncan (0.29) 21. Marc (0.29) 22. Z.Bo (0.27) 23. Bosh (0.24)
There are a lot of players that end up in 2 positions here, but I think there are 61 individuals. You can see that coming up with an essentially arbitrary number (like top 10) isn't really going to work for a couple of reasons: 1. There are clusters of players with identical values. This is seen for both C and PF, with 2 different sets of players. 2. Not all positions are created equal. Even the 24th best PF is better than the 10th SF!
Just from eyeballing the consolidated list, the value limit of 0.2 seemed to make sense. I don't think there are any players above that wouldn't be considered solid starters...well maybe Frye, but he's a bit of a fantasy aberration. And once you hit 0.2 you are getting into Tony Allen and DeAndre Jordan territory and it just doesn't feel right.
For the purposes of tie-break points, I think something like this might work:
1. Add modifier points for every player on your team that doesn't share the target players position, and subtract modifier points for every player on your team that does share the position. 2. List players with a value greater or equal to 0.45 are worth 2 modifier points. 3. List players with a value less than 0.45 are worth 1 modifier point.
From an eyeball of the list, this seems to work. 0.45+ guys are the fantasy all stars, and the guys below are solid fantasy starters. Of course it's not perfect, but it seems pretty good to me. And subtracting modifier points seems to make sense, because if I'm a solid (but not star) SG in the league, then I'm probably not gonna sign for Wades team...just not a good career move. Maybe subtracting is overpowered. Might have to play out a few scenarios to find out.
I guess it depends on what we want to focus on with the tie break. Personally, I was thinking about prioritizing the open 'starting' slots in order to keep competitive balance up front in the process. Essentially, I'd like the biggest factor to be whether or not a team has another good player at that position already, rather that total team strength like you've done. I was thinking about a tiered tie break system something like this (you start applying them at number one until something breaks the tie)...
1. Prime roster openings: the team with an open slot at PG, SG, SF, PF, or C to accomdate the free agent will win if the other team doesn't have a similar slot. In essence this is an open 'starting' job.
2. Secondary roster openings: the team with a open G,F, or Utl slot to accomdate the free agent will win if the other team doesn't have a similar slot. This is the 6th man type bench slot.
3. The team with more players from the above list not at the free agents position will win the tie break.
4. The team with the most regular season wins will win the tie break.
I was thinking we'd just produce a list for each team and have them include it in their bids for ease of reference. Perhaps every time a free agent is signed they fill up a slot on the list starting at the top. Either way we'll have to produce a tie break mod list for every team.
I'd be ok going either direction here based on what the rest of the league wants to see.
Another thing we could do is move around tiers depending on the price range of the FA.
Top guys are assured playing time, so their priority would be best change to win.
Bottom guys pretty much just want to maximise their minutes. Not sure how to reflect this though. Could add avg minutes of all players at that position and team with the lowest number wins. This might be a little fruity though.
They want to win, but they still want to play. Most guys want at least a reaosnable shot at starting if they're any good, or at least the assurance of a prime bench role. Good players don't want to win so much that they're willing to ride the pine. Teams also don't generally go after expensive FAs unless they have a positional need for them. Since fantasy doesn't necessarily work like that we end up having to somewhat enforce it from the opposite end.
Like I said, I think that keeping competitive balance at the front of our wish list here is the best way to proceed, and mostly that means I'm going to be against making roster strength the first tie breaker. I think it should be in there somewhere, but not as the go-to tie break.
I realize I didn't mention it in my last post, but nice work on the list. I think the .2 cut-off would work just fine, and I didn't see any glaring ommissions from the list. (although I'll have to go over it in more detail to be sure).
With every rule change or process overhaul this we need to consider how much additional work it brings. Mimicking the NBA becomes counterproductive once the time required to manage and regulate the rules, by either a commish or a committee, is ramped up. In a league where noone has yet put their hand up to act as commish, even if just for a season, it only becomes more important that the rules be commish friendly.
And once again the clock is ticking.
The draft lottery
New rules are easy to follow, odds mimic the NBA's. Helps teams needing a rebuild.
This one is make-or-break for me.
Teams in the NBA can hold anywhere up to 6 guys who rarely or never hit the pine, because they are simply not ready for the NBA, are injured, or whatever. Conversely in our league every roster spot is used so to maximise quantity of games played, making all projects and even injured all-stars prohibitively damaging in competition. We could mimic the NBA's salary cap, free agency process and trade rules, but if we don't address this fundamental shortcoming then it is a joke to suggest this league mimics the NBA.
Simply put the NBA has a real bench, and we don't.
OPTION A We go with an inactive list where GM's would be able to stash 1) injured players, 2) projects, and 3) backup scrubs without it hampering them every day, every matchup. The challenge then becomes ensuring GM's abide by the rules and players remain inactive for the week's matchup barring injury. That means checking yahoo! daily. Expecting a commish to do it is unrealistic, even if we had one. The only realistic option to me would be GM's keep an eye on one another, and in case of a breach some automatic penalty that makes the infringement counterproductive, preferably during the same weekly matchup. In the event a breach wasn't reported within ie. 3 days it would simply go unpunished. GM's would need to merely check their opponents lineups, checking that Monday's inactive players remained benched barring injury, which is easy enough to do.
OPTION B Extend the number bench positions by ie. 4-6 spots. In this scenario GM's could still rotate players in/out of their lineup but as players 15-20 would be scrubs the gains from ramping up games played would be minimal. Meanwhile teams that wished to hang onto injured players and projects would actually have an option. The minimum number of players would remain at 13 so filling out your bench would simply be an option, not a requirement, for teams that felt the change was unnecessary.
OPTION C Reduce the number of active spots. We could drop a UTIL spot, maybe a C spot too, but aside screwing with player values with regards to how many centers we play, this alone is not going anywhere near far enough towards fixing the problem in my opinion.
OPTION D Leave things essentially as they are. (FAIL)
Option A is more realistic, option B is simpler. Personally I prefer option A, but would be willing to try option B and see how it plays out, see if having a stack more bench spots encouraged GM's to retain their picks and younger players. Options C and D are non-starters for me, I would cease to consider this a real keeper league and essentially stay on merely as a caretaker GM. I'm into fantasy keeper leagues because they offer realism, and maximising games played instead of having a bench to carry projects and injured guys is simple not going to cut the mustard for me. Aside the need for a commish, this is hands down the biggest issue this league faces.
No small change please. Bids should go up in increments of $100k for simplicity reasons, regardless of whether we are using blind bidding or not. If we are using open bidding, $10k increments makes an auction drag on. With regard to blind bidding, it's rarely necessary. $10k achieves little in terms of seperation between offers for a player, and is just a PITA, especially for the commish (and Scully assuming he manages the data).
Free Agent Process
I'd like to see a totally blind bidding process. All offers would be made by the Friday or Saturday midnight, with the results announced ASAP afterwards in time for the following matchup. GM's wouldn't even need to announce if they had placed a bid on a player. Once a week a committee member would check a dummy email acct and sort through any offers. I could flesh it out, and we'd modify the process for the offseason, but that would be the crux of it. It's relatively simple, and it mimic's the NBA.
Regarding waivers, teams would have the option of picking up a waived player on their current contract for 24 hours following their release. In the event two teams apply their current standings would determine who gets the player. For the first 6 weeks of the season though last seasons final standings would determine the tie-breaker. Any player whose contract was taken up would no longer require a buyout, otherwise they'd receive 75% of their deal. Future season buyout debts could be reduced with cap space under the luxury tax level at the end of each season (NBA doesn't do this but I think it's a good rule for fantasy purposes). After 24 hours if noone had offered to take on the deal the player would enter unrestricted free agency. The GM that waived the player would be unable to place bids on the player for ie. two full weeks. Once again, this is fairly simple and mostly mimics the NBA process.
The Amnesty, I think we have to seriously consider whether or not it's fair on teams like TheRobSays (I think?), who have already taken their lumps and ie. assigned draft picks etc. to move bad contracts. An amnesty would let whoever took the contract to simply dump it but still retain the benefits of the trade. While taking over a team laden with a bad contract or two is tough, noone was forced into taking over a team here. I personally think the only fair way to approach this is to either dump the amnesty idea and make GM's with bad contracts, regardless of whether they or a previous GM signed them, tough it out, or reboot.
I'm in favor of expanding the number of roster spots to 15(or 16) so that stashing draft picks is easy to do and doing weekly lineups where we start 10 guys. This allows for more competitive match ups between the have's and have not's, but it still rewards teams that have stronger top to bottom rosters through better management. I'm also in favor of moving to a straight hard cap, but bumping the number up to account for extra roster space and the rookie scale. I'd set the straight hard cap at 72.5 million(half way between our current soft and hard caps). A hard cap will create more player movement and allow for a more competitive league.
Bidding I'd keep it as it is.
Free Agent Process
I like the idea of blind bidding, especially for the offseason. I agree with putting players in groups, and then giving everyone a deadline(I like 48 hours, but because of the time crunch i'd be fine with 24 hours for this season) to submit contract proposals for anyone in that group. If two or more teams offer the same top contract go to the re-designed tiebreaker system. For in season free agency, set a day each week that all contract proposals must be submitted to the blind email account.
Markos wrote: Regarding waivers, teams would have the option of picking up a waived player on their current contract for 24 hours following their release. In the event two teams apply their current standings would determine who gets the player. For the first 6 weeks of the season though last seasons final standings would determine the tie-breaker. Any player whose contract was taken up would no longer require a buyout, otherwise they'd receive 75% of their deal. Future season buyout debts could be reduced with cap space under the luxury tax level at the end of each season (NBA doesn't do this but I think it's a good rule for fantasy purposes). After 24 hours if noone had offered to take on the deal the player would enter unrestricted free agency. The GM that waived the player would be unable to place bids on the player for ie. two full weeks. Once again, this is fairly simple and mostly mimics the NBA process.
Agree with this setup for waivers.
I'm in favor of a one time amnesty, that any team in the league can use. For fiscally responsible teams you could say the one time amnesty could be used anytime during the next 2 or 3 seasons(whatever deadline you wish to impose). It's either league wide or nothing, though.
- Increase the total number of roster spots and/or reduce the number of active roster spots. The purpose of this is to reduce the % of a team’s roster that can actively be played during the week.
I didn't really hate the bidding process either, to me it is closer to the real thing. Each players bids on them people so known they are actually public knowledge, no way the GM's talking to the agents don't hear them. We just need a rule to help out the lower teams who are trying to sign players. For instance, a tie breaker for min contract auto goes to the worse team. That or first come first serve for a min contract tie breaker.
I'm still on board with ONLY new teams getting exemptions. Everyone else made their own bed (and that includes me, who could really use the exemption). Each and every time someone takes on a new team they have 1 free waive, no $ limit.
Markos - 100K increments doesn't really work for the lower end salary guys. I do think that upping the increments would speed up the bidding process though. Maybe we want to go with 50K increments up to 500K or so and then 100K increments after that? the 10K thing is brutal though, no doubt.
On the roster front I think we shoud add a bench slot and change either one or two other slots into bench slots. We oculd either drop the Utl slot for one more bench slot, or drop both G and F for two more. Personally, I think that adding one slot and dropping Utl should be enough. Not only does it increase the ratio of bench to active slots, but dropping the Utl slot also takes away the most flexible spot used to roll players through the lineup.
One way to resolve this that I suppose we oculd consider is to change the league to a roto league, which does have games limit available. Poeple may not like the idea, but it does solve one problem.
I'm fine with the waiver system Markos suggested too.
Switching 1 utility slot for bench and adding a 15th slot, also bench, is only going to help marginally. Guys will still be pushed to use all their slots to rotate guys in/out of active service to maximize games played, any non lotto rooks or injured players will remain a heavy burden.
Don't forget, reducing the active slots is only going to make extra games played more important. We need either an inactive list of 2 or 3 guys, or to expand the bench to the point the last 2 or 3 guys are sitting most of the time before using the bench to bring along projects or stash injured players isn't punished in the number of games played. It's a big enough hit losing a key guy to injury and having to sub in a scrub in their place without also taking a hit in games played. Not to mention the burden it places on any team wishing to retain a guy for a season or two and see if they develop.
Really don't understand why guys are opposed to extra bench slots, they don't have to use them if they don't want to. For me bringing along a few young guns over the course of a season or two is half the fun, any fool can grab the flavor of the week off waivers.