THE BOOK--Playing The Percentages In Baseball

I wrote a piece on the Hardball Times a while back asking for feedback on this issue, and settled upon a scale such that your first kept player costs $1 more, your second kept player costs $3 more, your third costs $5 more, and on and on through +7, +9, etc.  No limit to the number of players you can keep, though I don’t think anyone in our inaugural year kept anyone for more than +11.

We’re actually considering moving from +1/3/5/7/9… to +1/2/3/4/5, because under the former there really isn’t much continuity (so far) from year to year.

The plus side of having someone keepable for +1 is that if you get a really good bargain, you could conceivably have them on your team for their entire career (but, the rest of your team will suffer as they are quickly wearing out their initial great value).

In any case, it’s premature to comment on how well this is working (we’ve only been around for a few seasons), but thus far it’s proved to be a fun rule.

I run a keeper Yahoo! league, and we just completed our third full season. We allow unlimited keepers from your final roster, but we use “round inflation” like this:

First six rounds: keep up to any three as your first 3 picks the following year
Drafted round seven and up: costs their draft round + 3

It’s definitely not perfect, but allows a good pickup to provide good value to your team for at least 2-3 years. It also ensures plenty of good talent gets recycled into the draft to keep it interesting year to year.

I created what I believe was the first keeper league to use these keeper slotting system rules.  We created the Monkey Spankers Keeper league, which just finished its 6th season, based off of these rules.

The purpose was to mirror closely the monetary $260 keeper league but without a bidding draft and instead with a standard snake draft format.

The way it works is in the initial year you have a 27 round snake draft.  During the season you add and drop players.  At the end of the season you get to keep 8 players.  You make your draft matrix for the next year (reverse order of finish) and fill it in with the keepers.  A player drafted in the 9th round would get slotted into your 9th round.  A player picked up off of waivers or as a FA would be slotted into the 17th round.  For every extra year (beyond one year) you kept a player, he would get slotted up one extra round.  So if you wanted to keep that 9th round player for a maximum (or max is three years) of three years, then he becomes what we call at 7-1/3 and slotted into the 7th round.  We then have a collision formula for adjusting the slotting if two or more players get slotted into the same round.

Under this system, you can keep a player drafted in the first round for one extra year, a player drafted in the 2nd round for two extra years and a player drafted in the 3rd or lower round for three extra years.  It would of course be ideal to have the players drafted in the lower rounds moved up more than one slot per extra year, but we went with a one size fits all formula for simplicity.  It is already hard enough to administer a keeper league and it had to be easy enough for each person to be able to estimate their own player slotting.

But what it does is it gives a lot of value to getting a late good pick and since our league members are really into researching prospects, it fits our bill.  There is also a penalty for dropping players who are still under a long term contract.  You end up forfeitting a middle to late round pick until the 28th (extra last) round.  Teams are also allowed to trade draft picks and players during the offseason, with the understanding that they do so from within the confines of their 8 keepers and that they aren’t drafting a team of more than 27 players.

These rules have worked out well for us the past 6 years.  We conduct our draft during a two week period in March through an message board forum where each person gets two hours to make their pick.  If they don’t make their pick in time, they get skipped but can jump back in at any time.  We have a dead period from late at night to early morning so the draft clock isn’t ticking against people while they should be sleeping.

A few people have emailed me wanting to copy or use their own version of our system.  I have posted the rules of our system many years ago at the Fantasy Baseball Cafe under this same username. 

Link in my name takes you to our online message board and league rules.
vr, Xei

"Therefore, it would be easy to simply create a chart that shows something like this:
1st -> 1st
2nd -> 1st
...
8th -> 6th
9th -> 6th
...
18th -> 12th “

That would work fine.

I play in one league, it’s an auction with the $5 rule and this would be comparable.

On the subject of fantasy baseball, anyone looking for a league that attempts to model running a real franchise and can occupy large chunks of time everyday from February to October might want to check: http://mendozabaseball.com/ .

I played several years ago and couldn’t keep up with the daily time requirements, but I do highly recommend it.  I would think this site and that one have similar target audiences.

We have three main goals for our keeper league:
1) reward managers who make skillful (ie, lucky) picks to snag players immediately before they see a large increase in value
2) turn the player pool over regularly
3) allow all teams the flexibility to remake their roster after a poor season.

This leads to the following format:
1) Keeper Cost = round drafted / 2 (round up)
2) If two players have the same value (ie, 6th round), one goes in round 6, the other in round 5
3) Players that were first round picks cannot be kept
4) Midseason pick-ups go for a 16th round pick

So if you make a midseason pick-up in 2010 (perhaps Jose Bautista), you can keep Mr. Bautista as your 16th round pick in 2011, 8th round in 2012, 4th round in 2013, 2nd round in 2014 and 1st round in 2015.  He cannot be kept in 2016.

If you drafted Mat Latos in the 16th round, you could keep him as your 8th round pick in 2011, 4th round pick in 2012, 2nd in 2013, and 1st in 2014.

If you draft Albert Pujols as your first pick, you cannot keep him.

This way, a manager who goes for rookies gets to keep the fruits of their investment, while a manager lucky enough to draft Albert Pujols can only keep him for one season.  If don’t like anyone on your team, you basically keep three free agents.

The round/2 looks pretty good.  Based on the data points I posted, I’d have it as round*2/3, but /2 is easier.

Problem my league members would have with something like round/2 would be that if they drafted some prospect like Buster Posey in the 20th round and wanted to keep him for only two rounds, he’d then be slotted into the 5th round - and if you wanted to keep him for 3 years it appears, if I am calculating this correctly, he’d be slotted into the 2nd round.  Not really much reward for making smart late round picks.  The 2/3 scaling would take him to the 6th round.

I think something around 4/5 would be better with leaving the player slotted in his original round if only kept for one year.

This would take my mythical B.Posey drafted in the 20th round to.

20 -> 20 (keep one year)
20 -> 16 (keep two years)
20 -> 13 (keep three years)

I think the leaving the year the same if the player is only kept for one year is key.  In the $260 format, I don’t believe a contract is charged the $5 extra if the player is only kept for one year.  A $1 player would be $1-1 if kept for one year, $6-1/2 if kept for two years, and $11-1/3 if kept for three years.

Tango’s 2/3 scaling would take B.Posey from $20 to $9-1/3 if you don’t penalize for the first year kept.  A little more reasonable.  Maybe this is what you meant all along, but it is not clear.

Thanks for all the good ideas.

Tango, a minor correction.  We “bump” guys up by 1 round only if they are in the top 100 players (rounds 1-5 in our 20-team league).  Guys from later rounds get bumped all the way to 5th round in their second “kept” year (3rd year of ownership).

I like the idea of having a more continuous function, though.  I’d have to think it over, because I think Mitch’s round/2 would be too severe a leap (again, 20-team league, so each round is 20 players).  But some other number might work well.

Round conflicts...the round conflict issue is mainly a problem with the undrafted players--but we solve the issue by promoting a player’s draft position by a round as Mitch does.  I think it’s a good solution, as managers should be allowed to keep multiple waiver pickups that work out well.  You still can only keep 1 first-rounder. 

I do prefer to limit the number of keepers to some degree to prevent absurd dynasties from forming.  The keeper cost rules do help, but I wanted to err on the side if caution here.  We may up it above 4 after this year, but this seemed like a way to both keep some team continuity while still allowing managers to make up ground and not be stuck behind indomitable dynasties.
-j

Consider making it a dynasty league where you keep everyone (you’ll want to provide for a minor league system too.) We are heading into year 9 of our 20 team league and we have only turned over 1 team in the last 4 years.

It seems to me that a simple way to handle a keeper league that manages to create turnover at all talent levels while still providing real continuity from year to year is to tie a player’s keeper status to his real world contract. So if a guy is signed to a multi-year deal or is still under club control, you get to keep him. If at any point in the offseason he becomes a free agent in real life (even if he re-signs with his previous team), he enters the draft pool in your fantasy league.

This allows a last place team to snag an ARod or Johann Santana caliber player in the draft, yet at the same time everyone gets to keep most of their players. It also replicates real world dynamics, with weaker teams having a strong incentive to deal a star player on the verge of free agency like Cliff Lee mid-season in exchange for younger guys that can be kept long term. You’d probably want to cap how long one person can keep a guy (so someone doesn’t get to keep Ichiro forever, for example). I think this would be a good way to keep things fun and balanced without much complication.

I think Aaron/10’s idea is neat, but I think 6 years is probably longer than most leagues will be comfortable allowing players to be kept--and that’d be all of the non-FA players, which is a huge chunk of the population.  It also requires a lot of bookkeeping by either the managers or the commissioner.  It’s do-able, I just don’t think it’d work for us (not that this discussion is all about my league, of course).
-j

I looked more at Mitch’s round/2 system, and I actually like it a lot.  Tango’s system (round*2/3) works well too, but I would like to continue to let folks keep players for a full year at original draft position before we start moving them up.  Therefore, once they start moving, I want them to move faster--hence round/2.  If we were going to immediately start moving players up, then I think round*2/3 is a better pacing.

I’m probably going to propose this to my managers.  Thanks for all the great ideas, folks.
-j

Here is the chart, using my league parameters (550 players, 22 teams, 262$, increasing the keeper cost by 5$):

If we focus on say a 16th round pick, we see he is worth a 11th round pick in year 1, 8th in year 2, 6th in year 3, and 4th in year 4.

Now, what if we follow a round/2 simple model, with a repeat of round=round in the first year.  So, the 16th round pick would look like this:
year1: 16th (instead of 11)
year2: 8th (same as my 8)
year3: 4th (instead of 6)
year4: 2nd (instead of 4)

Overall, it would seem to have a similar kind of value.  So yes, having a round/2, where the first year is round=round is a very fine and simple way to do a keeper league.

Kudos to whoever thought that up…

The round=round was created 6 years ago in the Monkey Spankers League to mimic the process in a $260 bidding draft keeper league, where…

Example 1:
Player drafted for $3

Keep one year: $3-1/1
Keep two years: $8-1/2
Keep three years: $13-1/3

Contract is not increased for keeping the player for one year, but goes up $5 for every year after that.

Example 2:
Player drafted 20th round (snake format).

Using scale of 0.8

Keep one year (round=round): 20-1/1
Keep two years: 16-1/2
Keep three years: 13-1/3

What does “-1/3” mean?

x/y means…
x = current year of contract.
y = number of years under contract.

If I have a 20th round player that I keep for 3 years (using scale of 0.8) then he will become.

2011: 13-1/3
2012: 13-2/3
2013: 13-3/3

oic, you are not allowed to go year-to-year with him, you have to sign him to his next contract, and after that he’s a free agent. 

That’s different from what I was thinking, that you can go year-to-year, but it costs you an extra 5$ each year to do that.

#17. Typically you cannot go year to year with any player.  You draft a player (or pick up via waivers or FA during the season) then during the offseason you must decide to either.

A) Drop player
B) Keep for 1, 2 or 3 years.

This adds risk to keeping a player for multiple years, because “most” leagues have a penalty (some more severe than others) for dropping a player under a long term contract.  That is part of the reason why I don’t feel the 1/2 or even the 2/3 ratio is ideal, as it only further lessens the value of a keeper.

If contracts were year to year, then you would know which players to keep for 2 or even 3 years, where if you have to make the decision up front (like in real baseball) some serious risk analysis must take place.  I’ve never heard of year to year contracts in fantasy baseball keeper leagues.  Players kept for one year should be thrown back into the hopper to be re-drafted by the entire league, not held on to by one owner on a year to year basis imo.

To limit total keepers, consider a separate budget system for “contract years”, where each team has an allowance that they allocate across their players in any way they see fit.  For example, my league has a contract year budget of 48 years for 24 man rosters.  All players require at least one year, giving me 24 extra years to assign.  I can have an assortment of 2-year deals and 3-year deals, and maybe one or two 4/5-year deals.  Or I can have a bunch of 1’s and a handful of 5’s. The key is that I must sign the contracts immediately after the draft/auction but before the season begins, not after it is over.

Along with this is a rule that you cannot break those contracts, similar to what Xei had mentioned. So, if you drafted Stephen Strasburg going into 2010 and assigned him to a 2-year deal (at $15 or 10th round or whatever, again IN ADVANCE of 2010), you are effectively stuck with him in 2011 at that price even though he isn’t going to play for you much if at all.  Likewise, if you drafted Jose Bautista in Round 23 but only signed him to a 1-yr deal, you cannot simply “keep him” for any price—you would have had to have had the foresight to sign him to a longer-term contract BEFORE he got good (this is where my league’s method differs from Xei’s).

I strongly recommend this method.  It is not hard to track, just another column on your roster sheet.  Also makes trading much more interesting and keeps a fresh flow of interesting new talent in the draft/auction each year rather than always having surprise players be “kept” almost by default.