I still think this is a topic worth exploring. Or at least, I think keeper overhaul is a topic worth exploring generally, but it's hard to keep track of what's going on through 9 pages of different types of proposals, so I'm starting a new thread to focus discussion, since this idea seemed to have some traction. If people have other ideas for how to change the system (they did months ago, at least), feel free to start other threads.
WHAT RFA (Restricted Free Agency) MEANS
Owners can bid on each others' players, to a limited extent, as part of the keeper process.
WHAT'S WRONG WITH THE STATUS QUO?
I believe that the risk/reward model we have with 3-year contracts encourages relatively uninformed gambles. I would prefer that teams not to have to commit up front to long-term contracts, and thus avoid being punished across multiple seasons for poor player evaluation, failure to anticipate injury, etc. However, many players would not remain keepable for more than a year or two at +$5/year, and lowering the inflation rate would make elite players too valuable for too long.
WHY USE RFA
I believe RFA can solve the problem of having a fixed, single year inflation rate. RFA can drive up the prices of elite players faster than the baseline rate, allowing us to lower the base inflation rate so that teams can keep favorite non-elite players longer. At the same time, the limitation on volume and pricing of bids should allow teams to retain significant value from elite players.
HOW I WOULD IMPLEMENT IT
First, change the base inflation rate to +$3, and eliminate the rookie contract distinction.
Second, introduce a RFA signing period, to take place after the keeper signing period. It would look like this:
1. Keepers submitted
2. RFA offer sheet phase
3. RFA matching phase
RFA offer sheet phase:
*Every team can submit claims on any players rostered by each member of their division, ordered by priority
*Offers must be $4-9 more than the price at which the player was kept
*Owners must have enough cap space to accommodate their largest claims for each division member combined, plus an additional $18 (the maximum possible) for potential matching. If they do not, all of their claims are voided.
RFA matching phase:
*Owners are given the opportunity to match claims on their players at the claim price minus $3.
*A claimant wins a player when the owner declines to match.
*If multiple teams claim the same player, the higher bid wins. In the case of a tie, the lowest-finishing team wins.
*A claimant can only win one player per division owner. Having won a claim, the claimant's remaining claims on the owner's players are void.
*A claimant can only force an owner to make one match, but can be involved in multiple matches in the case of multiple claims on the same player.
*Therefore, an owner can only be forced to match 3 players.
*Owners respond to the claims of the lowest finishing team first.
Here's an example that should help illustrate how things would work in a case of maximum confusion. Most of the time it would probably not be this complicated, but hopefully this helps explain why some of the edge-case handling exists.
Teams A, B, C, and D are all in the same division.
Team A has Mike Trout, Clayton Kershaw, Giancarlo Stanton, and Paul Goldschmidt (Team A is a jerk).
Team B submitted +$9 claims on Trout, Kershaw, and Goldschmidt, in that order.
Team C submitted +$9 claims on Kershaw, Stanton, and Goldschmidt, in that order.
Team D submitted +$9 claims on Stanton, Trout, and Goldschmidt, in that order.
Order of finish was A-B-C-D.
The lowest finishing team's bid goes first, so A has to decide on D's claim on Stanton. A chooses to match Stanton. Now D can't force A to do any more matches.
Next, A has to decide on C's claim on Kershaw. A chooses to match Kershaw. Now C can't force A to do any more matches.
Next, A has to decide on B's claim on Trout. A decides NOT to match on Trout. However, since D has not won a player yet, and D finished lower than B, D is currently winning the bid for Trout, so D receives Trout. B is still considered to have forced A to match, so his Goldschmidt claim is ignored.
Note that, had B put Goldschmidt at the top of his list, and A decided not to match, B would have received Goldschmidt, not D, since D had a higher priority claim on Trout.
My goal with this way of handling tiebreakers is to incentivize everyone to rank the most valuable players the highest. I'm not 100% sure this system accomplishes that perfectly, but I think it should come pretty close. It's hard to come up with a scenario in which it makes sense for a team to duck their responsibility to bid up a player like Trout (on the assumption they won't get him) in order to win a different player. Or at least, you'd only do so if you'd be willing to do so in a vacuum, rather than because you expect someone else to take care of it.
The other goal is to make this only a two-step process, rather than the 3+ steps it threatens to be. I think it works because the resolution order is fixed, so owners can just receive an email instructing them to respond to a set of bids. In the example above, Team A would be instructed to match or decline on Stanton, Kershaw, and Trout, and wouldn't need to know anything about who submitted the claims or how those players were selected. And I'd be happy to handle the logistics of this if people were interested in the fundamental concept but scared of the bookkeeping.
WHAT RFA (Restricted Free Agency) MEANS
Owners can bid on each others' players, to a limited extent, as part of the keeper process.
WHAT'S WRONG WITH THE STATUS QUO?
I believe that the risk/reward model we have with 3-year contracts encourages relatively uninformed gambles. I would prefer that teams not to have to commit up front to long-term contracts, and thus avoid being punished across multiple seasons for poor player evaluation, failure to anticipate injury, etc. However, many players would not remain keepable for more than a year or two at +$5/year, and lowering the inflation rate would make elite players too valuable for too long.
WHY USE RFA
I believe RFA can solve the problem of having a fixed, single year inflation rate. RFA can drive up the prices of elite players faster than the baseline rate, allowing us to lower the base inflation rate so that teams can keep favorite non-elite players longer. At the same time, the limitation on volume and pricing of bids should allow teams to retain significant value from elite players.
HOW I WOULD IMPLEMENT IT
First, change the base inflation rate to +$3, and eliminate the rookie contract distinction.
Second, introduce a RFA signing period, to take place after the keeper signing period. It would look like this:
1. Keepers submitted
2. RFA offer sheet phase
3. RFA matching phase
RFA offer sheet phase:
*Every team can submit claims on any players rostered by each member of their division, ordered by priority
*Offers must be $4-9 more than the price at which the player was kept
*Owners must have enough cap space to accommodate their largest claims for each division member combined, plus an additional $18 (the maximum possible) for potential matching. If they do not, all of their claims are voided.
RFA matching phase:
*Owners are given the opportunity to match claims on their players at the claim price minus $3.
*A claimant wins a player when the owner declines to match.
*If multiple teams claim the same player, the higher bid wins. In the case of a tie, the lowest-finishing team wins.
*A claimant can only win one player per division owner. Having won a claim, the claimant's remaining claims on the owner's players are void.
*A claimant can only force an owner to make one match, but can be involved in multiple matches in the case of multiple claims on the same player.
*Therefore, an owner can only be forced to match 3 players.
*Owners respond to the claims of the lowest finishing team first.
Here's an example that should help illustrate how things would work in a case of maximum confusion. Most of the time it would probably not be this complicated, but hopefully this helps explain why some of the edge-case handling exists.
Teams A, B, C, and D are all in the same division.
Team A has Mike Trout, Clayton Kershaw, Giancarlo Stanton, and Paul Goldschmidt (Team A is a jerk).
Team B submitted +$9 claims on Trout, Kershaw, and Goldschmidt, in that order.
Team C submitted +$9 claims on Kershaw, Stanton, and Goldschmidt, in that order.
Team D submitted +$9 claims on Stanton, Trout, and Goldschmidt, in that order.
Order of finish was A-B-C-D.
The lowest finishing team's bid goes first, so A has to decide on D's claim on Stanton. A chooses to match Stanton. Now D can't force A to do any more matches.
Next, A has to decide on C's claim on Kershaw. A chooses to match Kershaw. Now C can't force A to do any more matches.
Next, A has to decide on B's claim on Trout. A decides NOT to match on Trout. However, since D has not won a player yet, and D finished lower than B, D is currently winning the bid for Trout, so D receives Trout. B is still considered to have forced A to match, so his Goldschmidt claim is ignored.
Note that, had B put Goldschmidt at the top of his list, and A decided not to match, B would have received Goldschmidt, not D, since D had a higher priority claim on Trout.
My goal with this way of handling tiebreakers is to incentivize everyone to rank the most valuable players the highest. I'm not 100% sure this system accomplishes that perfectly, but I think it should come pretty close. It's hard to come up with a scenario in which it makes sense for a team to duck their responsibility to bid up a player like Trout (on the assumption they won't get him) in order to win a different player. Or at least, you'd only do so if you'd be willing to do so in a vacuum, rather than because you expect someone else to take care of it.
The other goal is to make this only a two-step process, rather than the 3+ steps it threatens to be. I think it works because the resolution order is fixed, so owners can just receive an email instructing them to respond to a set of bids. In the example above, Team A would be instructed to match or decline on Stanton, Kershaw, and Trout, and wouldn't need to know anything about who submitted the claims or how those players were selected. And I'd be happy to handle the logistics of this if people were interested in the fundamental concept but scared of the bookkeeping.