FF2010: Keeper League discussion thread

Pages

Note: This is not the signup thread. This is also not related to the two "regular" leagues that I've been running for the last (I don't know, what?) three or four years.

Now...

I'm thinking about firing up a 3rd fantasy football league this year. Since I'm doing it mostly out of my own personal curiosity, it will include the two features about which I am curious because I've never done them before - keepers and an auction draft. Note that the important part of that least sentence is the part where I admit the "I've never done them before." Consider yourselves forewarned.

So...

Because of the way I plan to implement this, it's going to start a little earlier than the other two leagues. The signups for this league (to ostensibly be named GWJFFLK) will open some time on Sunday, 7/11 and will remain open for exactly one week. Admittance will not be first-come, first serve but will instead be slanted towards those that have participated in the other leagues the longest as well as other factors determined by me. Sorry if anyone disagrees with this method but given the fact that it's a keeper league, I'm looking for people that will be around a while and the most objective criteria I can think of is to look for those who've already proven their ability to remain committed.

Anyway...

First, the similarities: Scoring and almost everything else about the league after opening day will be the same as the other leagues. If you've played with us before, you already know what that means. As far as the differences, well that's really the main point of this thread. Anyone and everyone is welcome to toss in their two cents. I'll share the rest of my personal thoughts in the next post.

For reference, here's last year's thread on the topic. Those who participated and showed interest then will also be considered favorably for selection into this second attempt.

This draft type is becoming increasingly popular as is FA acquisition both of which are seen as the fairest way of doing things. We implemented both of these features in one of my new baseball keeper leagues this year and so far, that league is the most competitive and subsequently the most fun. It's really made me want to try it in a football format.

Sucks in regards to signups. I missed the boat last year. Hopefully I can get in on this one.

Any and all comments, questions, or suggestions are welcome. Additionally, these plans are not set in stone; I reserve the right to make changes as I see fit.

The draft: Everyone starts off with X virtual dollars. Someone nominates a player and the bidding begins. Unless I scrape together a web app, bidding will be done by email to me (if it's blind) or in some semi-private Google Doc (if it's open). Everyone will have X hours from when bidding opens to either bid or officially pass before they forfeit their opportunity to bid. Obviously, if everyone is active and engaged and passes in a timely fashion, rounds can move faster. I expect blind bids will also move faster since everyone only gets one shot but open bids might be more exciting. So what do y'all think? Blind or open bidding?

Keepers: We've obviously got a bit longer to figure this out but given that I expect to have this be an auction draft every year, keepers will have a price in subsequent years that is a factor of their original acquisition price. Given that NFL contract usually go up every year, you can expect that same thing to happen here. Also, since we use bidding for free-agent acquisition, this will also address the problem of "what do we do with mid-season pickups?" that can be a factor in snake-draft leagues. Anyway, I actually have less concern about this since I'm a lot firmer on what I plan to do here. The only real questions are along the lines of how much should salaries escalate each year and should the maximum contract (the most seasons you can keep a guy) be three, four, or five years? (Basically, I'm leaning towards four but could see going a year either way.) Obviously, the latter issue can be addressed among the members of the league over time.

I think blind bidding sounds rather interesting. I really only know about open bidding but with mid-season pick-ups being done via blind bidding I'd say most of the regulars have a decent amount of practice.

I know with keeper leagues without auction it's usually 1 round higher then when you drafted the player the first year and 1 more round each subsequent year. I heard of one where they did it via 2 rounds each year but I perceive that as generating too much turnover.

I'm sure there are some standards out there but what about a 50% increase in price paid each year kept? Also you can only keep 50% of your roster from the previous year? Does the WCOFF have any rules we might be interested in adapting?

50% feels a little high. I think folks should be able to keep even their premium players over a couple years without it hamstringing their entire roster... but we'll certainly see what feels right once the draft has been completed and we can see where the prices have ended up.

I admit I haven't looked since last year but while they do some sort of auction, I don't think WCOFF does keeper leagues.

Grumpicus wrote:

and should the maximum contract (the most seasons you can keep a guy) be three, four, or five years? (Basically, I'm leaning towards four but could see going a year either way.) Obviously, the latter issue can be addressed among the members of the league over time.

I don't think there should be a maximum at all. I think players re-entering the draft pool should come naturally from the annual price inflation of the player.

Having year limits means we have to track that information. It's inelegant. Instead, I think the only thing that should matter is the player's dollar value.

At the end of a season, everyone has a roster with players who all have a dollar value, be it from this year's draft, an inflated number from being a keeper from a past draft, or from free agency. The point is, it's a dollar value and everything else flat out doesn't matter. The dollar value is the only "state" value a player has.

It's very clean and easy to grok. You look at your roster and the dollar values and that's all the information you need to know. The idea then becomes coming up with the right inflation formula to achieve the desired effect: allowing players to be kept for a bit but not indefinitely, while still allowing savvy owners to acquire good bargains where they get a guy below market value for a year or two because of their foresight.

Previously, we discussed the idea of a salary floor, so that a $1 guy doesn't take forever for his annual percentage increase to amount to something.

So, something like:
$1 - $10: Normalize to $10 (salary floor)
$11 - $39: +30% increase over last year's value
$40+: +20% increase over last year's value

The numbers are basically pulled from my ass, just a theoretical example.

Given a sufficiently aggressive cost increase, guys aren't going to be kept beyond the ~4 years you're looking for. I don't think it's necessary to track years or any other extraneous stuff to achieve the desired effect (or at least pretty close to it)

I don't think there's one perfect set of rules - I think we could happily play under a myriad of systems. I also think every system is going to have inefficiencies in edge cases where someone is more/less of a bargain than perhaps they should be, and I'm OK with that - a little fuzziness is a good thing IMO.

Given that, I strongly lean towards simple and easy to understand. Even if the alternative isn't necessarily complex, I find myself very strongly wanting to push towards the elegant, low-upkeep approach.

Eh, most of the extra work would be on me... and I'm not crazy about the thought of someone having CJ for 8 years (a number I admittedly pulled out of my ass) just because they got lucky. By comparison, just because the Titans paid him crap last year doesn't mean they'll get to do it forever.

Of course, that's just my knee-jerk reaction and I can appreciate the simplicity of your proposal. I tried to do some research so I could run numbers based on last year's leagues and it seems Fleaflicker isn't showing any historical data at the moment. At least not transaction logs and draft results. I'm sure once I see some scenarios, it'll all work out in the end.

Grumpicus wrote:

Eh, most of the extra work would be on me... and I'm not crazy about the thought of someone having CJ for 8 years (a number I admittedly pulled out of my ass) just because they got lucky. By comparison, just because the Titans paid him crap last year doesn't mean they'll get to do it forever.

Indeed. I'm confident, though, that this scenario could be addressed with market mechanisms.

Just throwing an idea out there: what if every team owner was allowed to make 1 or 2 "restricted free agency" offers to players on other rosters?

Say, if you have CJ for $5 somehow, I might have the ability to make a $50 "offer" on CJ. You have the right to match the offer and keep CJ, or refuse the offer and lose him. CJ is now a $50 player, either for me or for you.

It would presumably be limited, mainly a mechanism to address extreme edge cases.

Perhaps there would even be an additional cap penalty (say, 10% of the deal's value is clipped from your cap number for the year - so $5 on the $50 deal) to encourage people to only use it when the value proposition is high - in other words, I'm willing to take a cap penalty so that you don't get to enter the season with CJ for only $5. It's small, just something to ensure that I don't go poaching without good reason.

I haven't fully thought it through, but it's an idea. I really like the idea of lightweight rules that turn everything into a money issue, where you make a value decision, rather than rules that enforce a strict absolute.

Apparently, I'm a real "free market" kind of guy when it comes to fantasy football. I feel like a Milton Friedman book.

Yeah, that doesn't sound complicated at all.

Grumpicus wrote:

Yeah, that doesn't sound complicated at all.

Smartass.

It's a simple game mechanic. Just like bidding waivers, it takes an hour to explain and about two seconds to understand.

It might work if there's also the possibility of protecting certain players from being plundered. Also, there's the fact that "restricted" generally means something like...

Wikipedia wrote:

If the old club elects not to match the offer, it may receive draft-choice compensation depending on the level of the qualifying offer made to the player.

Yeah, it sounds like we might as well just go with "contract lengths" by that point. In three/four years, then the RFA rules can kick in.

GWJFFLK - for the HARDCORE!

I think compensation would probably be taking the idea too far.

I'm also still of the opinion that a properly weighted inflation scale + salary floor is probably all that is needed, without any other sort of restrictions or time limits. I don't necessarily think that there needs to be additional rules to guard against fairly remote edge cases, and I don't think it's the end of the world when some edge cases do pop up on occasion. It's the responsibility of all of the other owners to prevent one guy from having a $5 Chris Johnson. (Also, Chris Johnson? Totally coming back down to earth this year. But that's a different topic.)

But this is kick-ideas-around time, so I am throwing out ideas. Mister Critical.

But since you brought up compensation and, by extension, qualifying offers, there's an interesting idea.

What if your keepers are determined by qualifying offers? By that, I mean:

* keeping a player at current dollar value would equal a no compensation qualifying offer
* keeping a player at $X over current dollar value means $X in compensation

So, say, I've got MJD at $15 (as the already inflated value for this new season). I can:

* keep him at $15, allowing someone else to offer him a higher value, which I can either match or lose him with no compensation
* keep him at, say, $30, so if someone makes an offer on him that I don't match, they also forfeit $15 that goes into my pool for that year.

I've raised the price of MJD from $15 to $30 for myself, but in doing so, I've made him cost at least $46 to anyone who would attempt to poach him. Making qualifying offers like this becomes a value proposition: how much "protection" do I buy to keep the player, while still making him a worthwhile value to me?

The desire to protect players would be an upward force on player values. Severely undervalued players would tend to get protected at higher values to prevent easy poaching, which in turn helps inflate them back to normalcy.

GWJFFLK - for the HARDCORE!

We've already got leagues for the normal folk. If ever we were to do a league in total rotisserie math nerd style, this would be the one.

I'm in for a keeper league.

I've raised the price of MJD from $15 to $30 for myself, but in doing so, I've made him cost at least $46 to anyone who would attempt to poach him. Making qualifying offers like this becomes a value proposition: how much "protection" do I buy to keep the player, while still making him a worthwhile value to me?

The desire to protect players would be an upward force on player values. Severely undervalued players would tend to get protected at higher values to prevent easy poaching, which in turn helps inflate them back to normalcy.

This is a great idea. The problem with all this will really be in administration, unless there is a site out there that does this already.

I have to admit, all the number playing that we could do in this keeper league has me excited. And I think if worse comes to worse we could just build some Excel spreadsheet with macros to automate things a bit.

People wrote:

stuff

You guys have fun with that. Right now, I just want to get past the draft. Then we'll have four more months to figure out the rest.

Elliottx wrote:

I have to admit, all the number playing that we could do in this keeper league has me excited.

Any crazy idea you think of, throw it out there.

I haven't fully thought it through, but it's an idea. I really like the idea of lightweight rules that turn everything into a money issue, where you make a value decision, rather than rules that enforce a strict absolute.

I'm in hearty agreement here.

I liked the idea of a keeper "floor" last year, but I do think it has its flaws. Chris Johnson would likely be a $17-20 player this year if his owner had drafted him in his rookie year, which is just insane. The restricted free agent idea solves this problem to a degree...in a perfect world, the top 24 players would be normalized to their market value. However, this incentivizes trying to cheaply draft players who are just outside of the elite (outside the top 24 picks in a normal league), which may create other issues.

In the end, a lot of this may be moot beyond the first few years. In every keeper league I've ever played in, the trade market served as the great equalizer as the years go on, and teams out of contention dump their high salary players for high upside players.

Landshrk83 wrote:

I liked the idea of a keeper "floor" last year, but I do think it has its flaws. Chris Johnson would likely be a $17-20 player this year if his owner had drafted him in his rookie year, which is just insane.

Maybe, maybe not. I think the draft values of rookies would adjust in the face of events like that. Let some guy get Chris Johnson for cheap, and you won't be letting 1st round running backs go cheaply in next year's auction, I bet.

In the end, a lot of this may be moot beyond the first few years. In every keeper league I've ever played in, the trade market served as the great equalizer as the years go on, and teams out of contention dump their high salary players for high upside players.

That's a good point as well. As the league unfolds, the guy more likely to end up investing his payroll in young players and ending up with a "steal" is the guy at the bottom who desperately needs a break.

I like ESPN because of the free live scoring (I don't think Yahoo does that, at least they didn't used to).

Here's a spreadsheet we use in one of my FF leagues. This is used for a Dynasty league hence it's robustnessess. It's used in conjunction with Yahoo FF.

Feel free to pick through it and take anything you like.

http://www.fseven.net/documents/NJFF...

Grumpicus wrote:

So what do y'all think? Blind or open bidding?

I think blind bidding would speed the draft up by a huge amount, but it would diminish the feel of drafting against live opponents. What about using a Vickrey auction instead of a straight blind bid? The winning bidder would pay the amount of the second highest bid, possibly plus some percentage. This keeps bidders from paying bids that are wildly out of line with the market (this is how eBay works).

Sign-ups opened. More details to come later. Keep the discussion in this thread.

I'm very interested, but I'm curious if my general lack of football prospect knowledge is a barrier to entry. I don't mind if it's a barrier to winning, just a barrier to entry . Auction day research aside, I'm a little intimidated by the other people who would be in this league.

I'd like there to be a Franchise tag for at least 1 player so that, regardless of how the $ value changes he's YOUR player for the year after you draft him. While smart [dollar management] should be a very large factor, I'd prefer if smart [drafting] has a way to trump that in at least one area. The Franchise tag could be part of the Blind Bid, so that you can still associate a market value to the player. All coaches put in a bid, with the current owner just putting "Franchise Tag" on their bid. The value is the second highest bid + X% similar to what was discussed above.

Jolly Bill wrote:

I'd like there to be a Franchise tag for at least 1 player so that, regardless of how the $ value changes he's YOUR player for the year after you draft him. While smart [dollar management] should be a very large factor, I'd prefer if smart [drafting] has a way to trump that in at least one area. The Franchise tag could be part of the Blind Bid, so that you can still associate a market value to the player. All coaches put in a bid, with the current owner just putting "Franchise Tag" on their bid. The value is the second highest bid + X% similar to what was discussed above.

I was mulling over some franchise tag type options, there's definitely some interesting possibilities. What about going with something similar to the real franchise tag rules- you'd get to keep one player at the average price of the 5 highest paid players kept at his position? Of course, this might just lead to everyone keeping their stud RB as a franchise player rather than paying his salary increase, and it would add complexity.

RE: The applications thus far - Bah, you guys suck for creativity.

Grumpicus wrote:

RE: The applications thus far - Bah, you guys suck for creativity. :P

Well, your post motivated me. Stop being so Grumpy and go record a podcast or something.

I like the options legion has presented. I prefer a standard auction to a blind auction. If it was done in a spreadsheet or with some type of software it could be done over a period of time. We could say that people had 48 hours to bid on a player after he was nominated. There could be multiple players being auctioned at the same time to speed things up.

For keeping I like the idea of a minimum amount for a keeper and increasing the amount by x% each year. I think that if someone does some research and gets a player cheap they should be able to keep them for a while. Not forever but for a while. I don't think rookies will go cheaply. In the keeper league I'm in rookies go quickly, especially running backs. We keep 4-6 players in my keeper and Moreno went in the first round. WRs go a little after backs but Calvin Johnson also went round one. QBs don't generally go early but they usually take some time to develop and aren't worth keeping as rookies anyway.

It could also be a graduated salary cap. If a players salary is less than x their increase could be higher than someone who is above x. That increases cheap players faster than more expensive players but doesn't require going back years to determine cost. If you combine the minimum with graduated increases I think you've got a pretty good solution to keeping a cheap player forever.

In the other thread which is now closed, ukickmydog wrote:

Sigh, get an invite to join, then don't get in. :(

Just to be fair, you got a PM to make you aware that applications were open (i.e. the thread was created). If you want to think of it that way, you got an invite to apply. As for not getting in, the committee decided to go a different direction. Sorry.

BlackSheep, kaostheory, Landshrk83, and garion333, check your inbox.

Grumpicus wrote:

BlackSheep, kaostheory, Landshrk83, and garion333, check your inbox.

Jeff Goldblum's Unfertilized Egg has joined the league.
2011 - Jeff Goldblum's Fertilized Egg
2012 - Jeff Goldblum's Hatchling
2013 - Jeff Goldblum's Chicklet
2014 - Jeff Goldblum's Maturing Young
2015 - Jeff Goldblum's c*ck

2015 is going to be a great year for Jeff Goldblum

Pages