By the way, the Slugs hit bottom at 13 games under .500. So not every team is done if they get far behind. Of course, I have won 11 out of the past 12 weeks and still don't have the spot clinched. Not too likely of a scenario...

General odds of a wild card spot are (2/12)*(3/4) = 1/8, while odds of winning the division are of course 1/4 (and if you add those you get 6/16). However, in the scenario you were outlining, you were assuming the division was already out of reach. In that case, the 3/4 has already happened, so you're just left with 2/12.

As for you bottoming out at 13 under, I wanted to emphasize that I didn't remember what the actual numbers were. I guessed 10 under but it may have been 20. My point was just that there was a pretty clear point-of-almost-certainly-no-return that could happen within the first third of the season. That said, you may well have bucked that trend this year. I'll have to take a look.

Not exactly an 'other things being equal' analysis.

I make no claims one way or another about what all those teams that fell behind early did in terms of selling or not, but it's worth noting that the big sell trades weren't as much of a thing several years ago. What I am saying is that if you need a comeback of unprecedented magnitude, and you have the option to do something more likely to be productive than rely on that, it can make sense to take that other option.

If you're good at deal-making you can be successful either early or late.

Not if there's no one left to acquire. There are only so many 'known' quantities on good contracts that contending teams can afford to part with. By the time I was done with my trades there were very few available players I had any interest in. I have no doubt that a few of the players involved in the subsequent dozen or so trades will prove to have big surplus value, but I have no idea which ones.

Sharur wrote:As for you bottoming out at 13 under, I wanted to emphasize that I didn't remember what the actual numbers were. I guessed 10 under but it may have been 20. My point was just that there was a pretty clear point-of-almost-certainly-no-return that could happen within the first third of the season. That said, you may well have bucked that trend this year. I'll have to take a look.

Based on your spreadsheet, Freeman bottomed at -12 in week 8.

Not if there's no one left to acquire. There are only so many 'known' quantities on good contracts that contending teams can afford to part with. By the time I was done with my trades there were very few available players I had any interest in. I have no doubt that a few of the players involved in the subsequent dozen or so trades will prove to have big surplus value, but I have no idea which ones.

Were you certain about the surplus value you might gain from your trades? Are you as certain now? You imply that the amount of tradable talent is a fixed sum. I disagree - as a season progresses and players develop (and break down), new value will appear while the value you previously traded for might just disappear.

Regardless of our differing opinion on when to sell the farm, I think we can agree that it would make for a better league if more owners played for the present further into the season. Having said that, I think we have a pretty damn good league now and would be quite happy with the status quo.

I certainly think that the status quo is preferable to most of what's been proposed. But I am surprised at the lack of interest in moving to a year-to-year contract system, both for the logistical simplicity and the fact that there'd be much fewer 'shot in the dark' types of decisions. If you think value fluctuates that much in a month or two, what about in 3 years? I'll bet if we worked from that basic premise we could come up with a genuine improvement, but right now we're all running off in so many different directions that I doubt we're going to settle on anything that can pass.

It's not the year-to-year part of the system I find off-putting. It's the allocation part I'm not sold on. If you (we?) think abandoning long term contracts is a good thing, then let's propose that.

Yeah, this thread is like herding cats right now.

Sharur wrote:I certainly think that the status quo is preferable to most of what's been proposed. But I am surprised at the lack of interest in moving to a year-to-year contract system, both for the logistical simplicity and the fact that there'd be much fewer 'shot in the dark' types of decisions. If you think value fluctuates that much in a month or two, what about in 3 years? I'll bet if we worked from that basic premise we could come up with a genuine improvement, but right now we're all running off in so many different directions that I doubt we're going to settle on anything that can pass.

The goal isn't to get every player's salary exactly where it "should" be. We'd have no keepers and go with a full auction every year if that's what we wanted. (I know that's not exactly what you were saying, but my point is that imperfect information and fluctuating value are desirable to some degree.) Base year salaries and multiyear contracts add a huge dimension of awesomeness to the league by making player evaluation and trading much more complex than just the raw fantasy numbers. I'd also be against significantly devaluing keepers. I think a decent amount of year-to-year consistency in rosters is an important thing to preserve in this league.

I guess I'm just confused about why, if you don't like multiyear contracts, they need to be banned. You could just go year to year with your own team.

I do agree that the status quo isn't bad at all.

Our current contract structure heavily rewards multi-year contracts, because of the way our inflation rate works. But those contracts also involve a lot of additional risk, because projecting out multiple years is so difficult. I think we'd see better competitive balance and less incentive to sell early if we removed the pressure to take on that risk in order to gain an edge. It's not as if people couldn't maintain that continuity with yearly contracts, they'd just have more control over when to let a player go. Right now, the year-to-year inflation rate is too high relative to the 3-year contract inflation rate for a team to be able to retain its players for a significant amount of time that way.

There's plenty of uncertainty just from one year to the next, I think. Pushing it out to multiple years isn't necessary in order to get the effects you describe (there's a reason fantasy experts on FanGraphs are using a yearly contract model, after all- that's not the aspect of their league that makes it too hardcore/esoteric for our purposes).

Ok, I've read through the allocation posts and I believe I now understand how it works.

I hate this idea.

At first, I did think it had promise as a market-like way to "correct" player salaries to some degree. However, it's pretty clear that it's solely a competitive balance measure, and kind of a nasty one. Unlike any other action you can take in fantasy baseball, allocation doesn't do anything to directly help your team. Instead, the strategy here seems to be to identify your perceived toughest competitor and punish him/her as much as possible. It seems like this introduces a subtly aggressive element to the game that isn't currently there. Right now you do better by improving your team and beating others on the field. Allocation is an indirect way to make it easier for yourself by targeting and hurting other teams. That leaves a bad taste in my mouth.

I also don't see how this helps the tanking situation at all, because it doesn't devalue keepers for teams who aren't doing well. Unless I'm wildly misreading how this would play out, teams who did well the previous year will be targeted more, and teams who didn't would be targeted far less.

Regarding competitive balance, it's a philosophical question. Do we want to help teams that lose the previous year and penalize teams that win? I have no problem with facing a penalty or handicap of some kind as a result of doing well in a year or over a period of years, but let's have that be a sane and at least somewhat predictable penalty, not a punitive one where teams are going after each other. I've managed to pile up a great record over the past few years by hoarding good contracts, and if the league wants to take reasonable measures to make this harder to do, let's discuss them.

For example, I would be much more interested in something like a restricted free agency system.

Right now there are two ways to have a player on your roster on Opening Day. 1) Buy him at auction. 2) Keep him for a set increase in salary. Option 1 is pure free market. Option 2 is the reserve clause, just with small but guaranteed salary increases.

Maybe we can put some mechanism in that injects a little more of a "market force" (this should perk Mike's ears up) into the contract system under certain cases. It would have to be simple; no messing around with service time or number of consecutive contracts or anything like that. I don't think any of us want to go down that road.

I'm thinking about a period before keeper selection where teams can submit offer sheets to a limited number of players on other teams, with some restrictions on how high they can escalate the salary. The owning players would then have the ability to match (exactly or at some discount) and keep the player, or release the player to the offering team. This could be done blindly or transparently, in one batch or in some rotating way. I have a few ideas about details, but nothing has come together yet that seems like it would work well.

And apologies in advance for not helping the "herding cats" problem. This is brainstorming time.

I don't think there are any problems with the rules of our league. Exhibit A: Last year Freeman complained about how our salary structure created inequities in our league and bemoaned his chances against Matt. But now, low and behold, thanks to good managing, you could make a case that Freeman has as good a chance as anyone to win the title. And that's in one year building completely from scratch.

Now, if there's a way to improve the experience of the latter half of the season for owners whose teams are out of playoff contention, then it's worth exploring. I'd rather just simplify it and have the remaining owners compete for a $10 auction money first prize and a $5 auction money second prize. That's a small boost. It's not punitive, which works better for the reasons Matt points out.

Ultimately, I don't think we can prevent some owners from making poor decisions, and I'd rather not make rules that are basically just designed to save owners from themselves. Todd's going to think it's dumb to trade a top rental for an unproven prospect whether it comes 5 weeks or 5 days before the trade deadline. Maybe if we encourage owners to stay competitive by offering an incentive, we'll see fewer teams trying to hoard prospects and more teams targeting top existing contracts and filling out the rest of their teams with players who give them a shot to come in 7th or 8th.

schulni wrote:Ultimately, I don't think we can prevent some owners from making poor decisions, and I'd rather not make rules that are basically just designed to save owners from themselves. Todd's going to think it's dumb to trade a top rental for an unproven prospect whether it comes 5 weeks or 5 days before the trade deadline.

I'm very curious to know what the relationship between these directly adjacent sentences is meant to be.

And for the record, it's less 'dumb' and more 'damaging'. I can go into detail on that if people want but I'm not sure it's relevant.

Matt, I assume your idea is derived from the same concept in the NFL. I actually think that your idea might be another good (perhaps better than allocation) way to move to a year-to-year contract system. We could get that multiple inflation rates thing going in a way that isn't punitive.

In fairness to allocation, it's not very punitive, as you can only allocate so many dollars to a given team or player, and we could presumably also have a cap (as Nick described) for how much can be allocated to players and teams in total. I'm also not sure whether you would tend to target the teams that had won recently, or just the teams with the best contracts (in the aggregate, that is; I'd tend toward the latter personally). However, your idea makes additional salary inflation not "free" to the teams that are driving it, which is probably preferable from a strategic standpoint.

Quick proposal time, combining my earlier idea with Matt's:

1. All contracts are for a single year. All prices increase by $2/year (for simplicity's sake I'm ditching rookie contracts for now).

2. During the offer sheet period, each team simultaneously makes offers on up to N players. Those offers may represent a salary increase of up to X dollars. Teams should also rank-order their offers from most important to least important.

Possibly X is only limited by a team's ability to assume all of the contracts in question, but likely in order to preserve some amount of surplus value it would be something <= 10 (perhaps 8 so that max total growth is 10). Maybe N is only limited by ability to assume the contracts, under the assumption that you win all your bids AND lose no players? Though that would still leave open the possibility that your own players are bid up and you can't afford it all, so maybe it's also limited by an assumption that ALL of your players are bid up.

This could mean that teams with a lot of keepers can't put in many claims; in fact, a team could be in a situation where it can't afford all of its own players if enough are claimed, which may mean teams need to rank-order their own players as well, possibly even in one master list along with their claims on other teams. And as such, maybe N is unlimited, and it's just a question of whether you can afford your bid on a player any more by the time the queue gets that far. Though we still may want to limit it just for the sake of our own sanity.

3. If the original team is willing to match the highest bidder minus some discount Y (I'd suggest Y = 2), they retain the rights at the new price. If not, the player goes to the highest bidder. Ties go to teams that finished lower in the standings, but it works sort of like waivers, where once you win a tie you get bumped to the bottom of the list (unlike waivers you don't get bumped if there's no tie).

Someone let me know if this, in combination with #2, would result in non-optimal/manipulative bidding behavior (I don't think so but I haven't thought it through carefully).

I think that's it? Seems to me it would be pretty complex during the offer sheet phase and very nice and simple outside it.

Is every player eligible to receive an offer from other teams? If so, how is that different from the auction? I need more explanation on the point of the offer sheet period.

I'm not as opposed to allocations as Matt is. There's so little money involved under my proposal that I don't think it would create hard feelings or seem punitive. I like the elements of it that accelerate contracts with a ton of value and add some competitive balance for teams fortunate (and/or skilled) enough to have several of these contracts. And I like that there's a lot of other strategy possible, like targeting more expensive contracts or expiring contracts to make them less keepable.

As I've said, I don't want actually want to change anything. I'm just trying to think of ideas to address concerns other owners have brought up.

I really don't like the idea of mandatory year-to-year contracts. In leagues I've been in with that system, the keeper element feels wasted. I think the three year model works perfectly. There's a great balance between the risk of long-term contracts and the potential reward. Because George has shown that it's possible to be successful without signing many contracts period, there's room for owners to take many different approaches. With year-to-year contracts, everyone pretty much has to take the same approach.

schulni wrote:Is every player eligible to receive an offer from other teams? If so, how is that different from the auction?

It probably isn't different from the auction if you allow unlimited price increases and an unlimited number of offers. Probably we'd want to limit both, though. Maybe something like max 1 offer per team and 5 total. I think it wouldn't work if we allowed players to be protected, though, so they'd all be eligible.

schulni wrote:There's a great balance between the risk of long-term contracts and the potential reward.

Is there? You're right about George, but how many people just go year-to-year, as Matt suggested I do? For the most part it's either 3 years or nothing.

A base inflation rate of $2 or $3 should be low enough to allow people plenty of opportunity for continuity with players they like. I don't see how the keeper element would be 'wasted' in that context. Can you explain why it worked out that way in your experience?

How about this?

( 1) owners can offer 1-3 year contracts as before (rookies can be kept indefinitely)
(2) however, for all expiring contracts other teams can make offers prior to draft. These offers will be blind. (And will include whether they are 1, 2 or 3 years) owner will then have opportunity to match by offering the player a contract that is equal to or greater than the total dollar value of the top blind offer (for example--if the top blind offer is $50 for one year, then the owner could match by offering $50 for one year, a two year contract at $23 and $28 or a three at $13, $18 and $19)

I think this could add some interesting strategy, allow teams to recoup some investment in keepers without it being ridiculous , does not penalize some teams, and it is pretty simple. This is similar to Matt 's idea, I think.

I like the idea behind this, but I think the mechanics are off, and would allow the retaining owner to potentially set the yearly salary below what it previously was. I originally thought of something very close to what Todd proposed, but I think it relies too heavily on the bidding system to correctly police inflation on all but borderline keepers (the guys who are questionable even at +$2).

The idea of using expiring contracts as your "restricted FA pool" is a good one.

I'd suggest something like this:

1) Eliminate rookie contracts, but keep our normal +$5/+$8/+$9 inflation rate the same.

2) Restricted FA offers: Owners submit offers of any number and any amount to players in the restricted FA pool (expiring contracts). The only constraints are that the bids must fit in the owner's budget for the following year (bids + existing contracts can't be higher than $260) and all bids must be at least Base Year + $5. These offers are submitted by each owner in a single email and not revealed until this period is over.

3) Highest bids for each player are awarded. The current owner then has the option to sign the player to a 1-year deal that is either equal to the bid, or +100% of current Base Year, whichever is lower. (Rookie contracts are gone, so the lowest this can be is $11, for a player coming off a 1-year $6 contract.) If the owner declines to sign the player, the player is automatically released to the bidding team and signed to a 1-year at the amount of the bid. The match/release decisions are submitted by each owner in a single email and not revealed until this period is over.

4) Keeper phase: All other contracts are tendered the same way they are now.

Once again using Mike Trout as an example, here's how his salary would progress under this system if he were bid up to incur max inflation at every opportunity:

2012: $6
2013: $9
2014: $10 (initial 3-year contract ends)
2015: $20 (high bid, kept for +100%)
2016: $40 (high bid, kept for +100%)
2017: market price

Compare that to his current arc over the same years:

2012: $1 (Yep, he got 123 AB - 135 PA - in his initial call up and was still a rookie. I got so lucky here.)
2013: $6
2014: $9
2015: $10
2016: $15
2017: $18

Some details to be worked out here, such as the order things happen in the signing phase. Under this set of rules it's theoretically possible that an owner could have enough bids accepted, and match enough offers, to go over their $260 limit, so that would need to be addressed.