An Idea for a Two-Tiered Fantasy Baseball League: GMs and Agents

I had this idea a few years ago and wrote it up somewhere else, but in hope of actually making it happen now, I’m dropping it here. (To give an idea of how old the idea is, the example player I used in the original version was Albert Pujols.)

Short version:

  • Take a regular fantasy baseball league – 5×5, fantasy points, whatever. Doesn’t matter. Has to be an auction draft and a keeper league though. Determine the winner however you want. The competitors are the owners/GMs of the teams. You know, standard fare.
  • Add a second league next to it. The competitors are agents. Winner is the agent who makes the most money in any given season.

Longer version:

Let’s start with step 2 because it’ll make it easier to understand if we momentarily skip step 1.  Oh, and all the numbers – salary cap and stuff – they’re just for explanation’s sake. I didn’t try and balance them to make sense.

Step 2: It’s the first year of the league. The GMs have their auction. Each GM has a $120 million budget for this year. Players are auction off as you would in a regular league auction.

Straightforward, right?

Before we get to Step 1, let me explain how agents make money, at least in this fictional world for the game.


1) Take paid a percentage of their players’ contracts and

2) Maybe get some money if a player does well and/or if his team does well.  Think of this as a proxy for endorsement deals – if an agent represents the guy who lead the league in homers and his team won the World Series, the player would probably get some big deal from Gatorade or Nike or someone, and the agent would probably get a cut. I don’t know if this one is workable, though. It’d be cool, but may be too hard to pull off.

So anyway, Step 1.

Step 1: Agents have their auction. I don’t know how it’d work exactly, but here’s what I’m thinking:

  • Each agent gets $5m or $10m, not sure. Let’s use $5m for example. It’s not a salary cap because remember, they’re being measured by who makes the most money, and whatever they spend, they lose, so keep that in mind.
  • Agents bid on players in a bid-or-pass style auction, with two numbers in play: dollars and percentage points. Dollars go in $10k increments, starting at $10k and going up. Percentage points go in 1% increments, starting at 25% and going down.
  • Basically, agents spend $$$ to wine and dine players, so that’s the dollar stuff. They also negotiate with players by taking a smaller and smaller percentage of the contract.
  • An agent can outbid another agent by upping the ante in either tranche during their turn.  So if Mike Trout is up, and the current bid is $100,000 and 8%, you can go to $110,000 and 8% or $100,000 and 7%.

So taking Step 1 and Step 2 together, let’s say you’re Mike Trout’s agent – and to make it simply (and stupid), he’s your only client – and you spent $1m entertaining him and agreed to take only 3% of his contract.  The GM who won him at auction bid $20 million. You’d end up with 3% of that, or $600k.  So you have $5m – $1m + $600k = $4.6m in your war chest… if it ended there. (And you’d clearly suck as an agent.)

But wait, there’s more.

Step 3: GMs and agents negotiate long term contracts.

Consider the auction bid an offer sheet, and one which either side can unilaterally impose upon the other at any given point in the negotiations, ending it right then and there.  It’s not a floor, necessarily, although practically I guess it would be.

What’s up for negotiation? I’d keep it mostly simple:

  • Years
  • Salary per year
  • No trade clause
  • Options, buyouts, opt-outs, etc.

But you can get creative if the league allows. Imaginations can come up with really neat things.

After that, it’s mostly a regular league, except that agents are involved in transactions.


For regular player salaries, players – and therefore agents – get paid in quarter increments, one each on Opening Day, June 1, August 1, and October 1.  This has an effect on both GMs and agents. For GMs, it means that if you have $20m left in your budget on August 2, you can take on effectively $80m in player salaries, because 75% of those salaries have already been paid out.

Agents get paid when the player gets paid. However, that gets locked in once the deal is signed. So as long as the deal isn’t renegotiated, the agent gets paid even if he no longer represents the player.

The implication here is that agents have to manage their cash flow a bit – and of course, they’ll prefer signing bonuses to buyouts.  (Whether the league allows the agents and GMs to put in other $$$-laden clauses is up to them; again, your imagination is the limit.)


Agents can try and steal players from other agents. I have no real good way to do this, but here’s what I’m thinking: a morale check, followed by an offer and option to match (maybe?).

Each player has a morale number. It’s calculated – and I’m doing this off the top of my head – as follows:

  • Start with 10.
  • Subtract one if the player’s team (in the league, not in real life) is currently in the bottom half of the standings.
  • Subtract one more if the player’s team is in the cellar.
  • Subtract one the player is not one of the three highest played players on his team, by both current year salary *and* average annual value (excluding performance bonuses and options and buyouts?).
  • Subtract one more if the player is not one of the eight highest played players on his team, using the same formula as above.
  • Subtract one if the player is not one of the five highest played players at his position, same math.
  • Subtract one more if he’s not one of the 10 highest played players at his position, ditto.
  • Add three if the current agent represents three or more other players on the player’s current team.
  • Subtract three if the above bullet isn’t true *and* the stealing agent has three or more other players on the player’s current team.

(Note that the last two bullets encourage agents to try and manipulate rosters.)

The morale check: If the player’s morale is 7+, he can’t be stolen.

After that, the terms of the steal attempt are up to the stealing agent. He spends some dollar amount on a steal and if the current agent matches (with morale notes below), the current agent keeps the player and the stealing agent can’t try again for six months. If the current agent doesn’t match, the player switches agents and gets a +4 morale boost which wears off, 1 point per payday (as above) until it hits 0 in a year.

The terms of the agreement between the new agent and player – as well as what it takes for the current/old agent to stop the steal – are determined by the morale check described above:

  • 4+, he can be stolen, but the stealing agent has to decrease the percentage the player negotiated with the current agent by one point *and* the current agent only has to spend half as much to match and therefore block the steal attempt.
  • 0+, the stealing agent has to offer the same percentage and the current agent has to match the full money offered.
  • Under zero, the stealing agent can offer 1% under *and* the current agent has to double the money the offered by the current agent.

Finally, the minimum money offered by the stealing agent has to be at least equal to the dollars last paid by an agent to acquire that player.

If the steal is successful, the money goes from the stealing agent to the original one. (I’m trading realism here for some balancing.) If it’s unsuccessful, the original agent’s counter-bid just evaporates.

Again, using the Mike Trout example, let’s say his morale check is a 5. To recap, the agent acquired him for $1m and 3%. Someone tries to steal him, so they have to offer $1m and 2% by rule. He makes that offer but the original agent blocks by spending $500,000. That $500,000 evaporates.

Another agent makes a try right after but offers $2 million and, again by rule, 2%. The original agent decides to let him go. That $2m goes to the old agent.

Two years later,  Trout’s morale is negative somehow. Another agent offers him $2 million (by rule) and 3%. For his current agent to block, he’d have to pony up $4m.

Too complicated maybe. Too D&D, for sure. Maybe it needs a few d20s, dunno.


I think this could be fun. Stuff like no-trade clauses would be awesome, where you basically have to buy out the agent, and the agent may really really not want a player traded because of the morale hit other players would take, etc.

I don’t know how the second year would work. Part of me thinks free agents should negotiate contracts via their agents; part of me thinks it should be done via auction, just like the initial year. I also don’t know how unrepresented players would enter the league. None whatsoever. Further, there has to be some mechanism for agents to renegotiate deals for their clients, but there should be a way to prevent doing so in order to take over the value from the existing agent. (Probably some present-day-value calculation?)

But that’s the basics. What do you think?

Originally published on January 3, 2016