Game Theory and the FA Market

Really, I think Sandy has the best low risk approach to building a team.  Find out which prospects shake out in the wash and only after that process fill the confirmed holes with veterans.  If you do it in reverse you might find you've blocked or traded the wrong prospect.

Lower risk it is indeed, in the sense of avoiding extremely visible blunders...

The question is whether the problem is simple enough, to allow us to use one principle to the exclusion of the other ones.


=== Ceteris Peribus ===

Of course I agree - from a Game Theory standpoint, you'd much rather fill in the 4-of-a-kind and Full House yahtzee slots before you fill in the Chance and 1's.   Other things being equal.

Jumping on an opportunity FA in 2009 puts pressure on you to forecast your 2010-2011 minor leaguers correctly.   You slide the continuum lever over towards "more talent judgment required."

That would be a Golden Principle if MLB were a "mental sport" like high-stakes poker.  Buying FA's later would be a (decisive) advantage.

... in a vacuum.


=== Competing Principles Dept. ===

However, our problemo is that other Golden Principles come in from the side and compete with our first one.  :- )

The FA market is a game-within-a-game:  if John Lackey says, sure, I'll take 4 x $13m and you pass on that ... then two years on a Grade A starter is 7 x $20m?   Or not even available?

So now we realize, that if Gillick has 3 free agent markets (2010, 2011, 2012) from which to pick his players, and we have only one (2012), that now he has a different kind of big advantage. 


How much is it worth, to obey an absolute that we have to buy FA's last?  What if you save $30M with a value buy on the FA market? 

How many prospects balance a $30M savings?   That's why real GM's balance the FA market more heavily than any other factor, including the minor leagues:  the huge dollars at stake dominate the board game.


One approach (FA's at any time) gives you the luxury of more options on the "real" FA market. 

The other approach (target FA's later, when you're sure what position you need) gives you the luxury of having more options when deploying your minor leaguers.

Targetting FA's later would slide the continuum lever over towards "more luck required in FA opportunity."


=== Blending the Two ===

A Game Theory guy is aware of both principles, and (hopefully) in a masterly fashion, leans toward one or the other as circumstances warrant... this type of third-order strategy is what separates the elite of poker, backgammon, chess, etc...

A poker player is aware of the fact that if he's holding two pair and his opponent is trying to draw to a straight, that he should bet hard and push his opponent out of the pot before he can get lucky.  This is a golden principle in poker:  pressure the underdog to quit.

But he's also aware that Opponent X can be easily trapped when holding three-of-a-kind to a better hand... waiting for this specific showdown can easily provide the knockout blow with great reliability ... hence, he doesn't want to risk his bankroll when he knows the Magic Trap is coming along shortly...

As Fischer said, chess is a matter of delicate judgment, knowing when to punch, and when to duck... the mind factors many principles all competing against each other, and if fortunate, the player picks his path of least resistance... the principle that applies to the given setup of pieces on the board, Roto ballplayers available in a keeper league, etc.


=== In Real Life Dept. ===

It's awfully seldom that we see real GM's, given $30M to spend, not spend it.   The fact is, real GM's (fairly) consistently spend their money now.

Even more variables apply:  next year there might not BE money, if you don't win now (winning in 2010 creates more $$$ in 2012).   There certainly might not be FA's you want.  The GM's job is at stake now.   etc., etc.

Waiting to see which minor leaguers are good?  Saunders or Tuisasospo, Ackley or Trinfuel, Halman or Moore?   When will we know that for sure?  2013?


Chuck Armstrong used to use the "we kept our powder dry this winter" line a lot.  I think he used it about six years in a row one time :- )

I get the impression that the Atlanta Braves are another team that is very willing to just sit out the FA market until it's quite good and ready -- one year on, or three years on.

Maybe Sandy can speak to that...



I don't think there is much down side to signing stars to there first big free agent contract, particularly if they didn't sign away their first couple years of free agency.  Alex Rodriguez's was worth the contract he signed with Texas, Barry Bonds was worth the contract he signed with SF in '93, Beltre was worth his contract as well.  So this is the no-brainer category, but it is also rare. Unfortunately, no players of that quality and age profile are available this year and only two were available last year and they cost 7X$23M and 8X$22.5M. 
I grant you Lackey would be worth the numbers you listed, 4X$13M, and I would gladly sign on the dotted line, but such a bargain with a premium talent is also very unusual (the Vlad Guerrero contract comes to mind).
I think the key to the situation for the M's is that they need to figure out how to best use their resources, to maximize their market advantage against the A's and Indians of the world, without bidding head-to-head with the Yankees. This means not necessarily holding out for the Teixieras of the world since that involves direct bidding against the Yankees, though you can always know there roster holes ahead of time and find cases where they aren't going to bid.
If I'm Jack and I'm sitting on a lot of money this year as well as a lot of roster uncertainty, I make short term commitments with big money up front to improve the short term without costing me longer term flexibility.  I don't want to give Lackey 5X$16M, but would be go for 3X$21M?  If he's as good as he thinks he is, he'll more than make-up the difference with the next contract, but the M's avoid the albatross contract.  Isn't that the way to do stars and scrubs without having your star turn into Chan-Ho Park?

2 to spend your free agent dollars continuously on guys who don't require long term commitments UNLESS your very sure that the long term commitment is worth the risk.  Lackey might or might not be worth the risk depending on years and cost.  But we can be fairly certainly that Russell Branyan is worth the risk since he wants two years.  No long term commitment there to gum up your budget planning.


Yes, I think "given the current situation" in Seattle, playing Tui/Saunders/Moore would be a perfectly acceptable reality in April.  But, as Doc so wonderfully spells out ... there are COSTS associated with any approach.  My assessment of the Seattle reality is that at this moment, they have a core of 1.5 young players, (Gutz and Lopez).  Call them both .75 -- as Lopez is viewed as defensively suspect, while Gutz has a grand total of one solid year of production in his resume.
I often see threes in baseball.  And I think thinking in terms of thirds is a very reasonable mindset toward roster construction tenets.  I believe, based on watching dozens of teams through history, that it is ultimately foolhardy and detrimental to pay "big buck" FA money BEFORE you have a "solid" young 1/3 of your lineup in place.  Note -- this doesn't count 1-year fliers, low-cost stopgaps, etc.  You've gotta field a team every year, no matter what your plan.
But I cannot remember ANY team in the history of the FA period that "began" an expensive FA push that succeeded who didn't ALREADY have at least three young regulars already firmly productive and under club control.  I view Phillies travails as illustrative ... as they did a multi-year FA push ... failed repeatedly ... then dumped some of that high priced talent, waited for the kids to develop ... and then went back into the FA market to FINISH off the job.  That's the pattern I see as repeatedly most successful. 
The Braves pitchers got all the glory.  But in 1990, the year BEFORE they began their run, they had Justice (.908 age 24); Blauser (.747 age 24 SS); Gant (.896 age 25) emerge as a YOUNG core to build on.  They also had only 2 guys over 30 that season.  That first playoff team brought in one pricy FA, (Pendleton).  But, they also pushed Lemke, Hunter and Deion Sanders into the mix, too.  (Blauser was the only one who established himself in 1989).
The one problem with Doc's intro is that the team that ops to start building via FA actually does NOT get to play in three FA classes.  When you sign those first "high priced" FAs, the ability to repeat the process is compromised.  The result is ... you end up with the Yankees payroll, or, you end up frustrated because if you weren't paying $12 million a year to Silva and $10 million to Batista, you probably WOULD have had the money to land Teixeira.
Wlad failing doesn't prevent you from bringing up Saunders.  Silva failing DOES prevent you from going and signing Cliff Lee.  Paying $12 million for a mistake like Silva is WHY you end up "Frenching" Wakamatsu. 
But, all that said ... there is always the possibility that you can pull off a deal that helps build your young core w/o bloating your payroll.  The Putz/Gutz trade is a perfect example.  But, this takes a great deal of serendipity, where you find someone bloated with talent a position of need for you, while you're bloated with talent at a position of need for them.  Snell is Jack's second attempt to bring in a lightly regarded talent that "might" blossom into a core piece for tomorrow.  But, trading Clement thinned out the one farm spot that was clearly deep.  I'm not sure that Z has the chips he needs to pull off another such mini-miracle.
I just know I firmly do NOT want to bypass giving the youth a fair shot for the likes of Scutaro or Jackson.  (Crawford ... a tougher call -- but that's probably just because I'm softer on Saunders than most of the blog-o-sphere). 


But I go back and forth, because once you've gone 3x21 ... the next two are only 8.5m each, and that's 4 and 5 years out :- )   so it's tough to decide...
You're right, of course, that my "reductio ad absurdum" 4x13 contract figures are fine in theory but difficult to find IRL.


that it is ultimately foolhardy and detrimental to pay "big buck" FA money BEFORE you have a "solid" young 1/3 of your lineup in place.
Now this is a telling blow:   even if it were okay to develop four players, buy three, and then develop two more ...  THIS team has developed 1.5 players, so buying three leaves you needing to pound 4.5 square pegs into round holes.
Good show.  :- )   Am not saying that closes the discussion, but it's a body blow.


The one problem with Doc's intro is that the team that ops to start building via FA actually does NOT get to play in three FA classes.  When you sign those first "high priced" FAs, the ability to repeat the process is compromised.
How true.
Still, if I have the option to bet on 5th street, 6th street, and the river, and you can only bet on the river .... well, my deciding to seize 5th street doesn't mean I had zero advantage...
That said, teams DO buy FA's now, have it backfire, and then are lamed on 6th and 7th street... it happens...


And Zduriencik must know that.
Earlier, KGaff pointed out that there's always a glut at 1B/DH.  I suppose Zduriencik is exploiting his position-of-strength to further underpay Branyan, but that leaves me feeling a little sour.  Russ Branyan was a great org guy in 2009.
More understandable if Zduriencik is trying to get his long-term 1B/DH in here this winter.


Really, all I've done is clarify how difficult it is to be a successful GM in baseball (or any other pro sport).  I've never been more frustrated by a sports executive than I was with Wally Walker (the Sonics were my first sports love... boy did I love the Gus Williams 1-on-3 fast break).  No matter how much I wished to challenge his intelligence, he was surely qualified on paper. Graduated from UVa, got an MBA from Stanford, and worked at Goldman-Sachs.  Oh, he also has two NBA championship rings from his playing days.  But he absoluted lacked the Midas touch with talent evaluation, not to mention many other questionable moves.  In the end he sucked, but not because he was a moron, but because the job is really hard.
While we might only have developed two pre-free angency players, per Sandy, by the time we know what we have in Saunders, Tui, and Moore, Lopez and Gutierrez will be getting expensive.  Most players only have 2-3 years between being established and getting expensive.  We invested a lot of time Lopez, he's now an asset (not a big one, but an asset nonetheless), but he's a FA in two years.
It really blows my mind to go through this exercise and then have the standard opinion be that great players are over paid.  Have Albert Pujols gives you the production of Ichiro and Branyan from one position for less money.  It's why the mid-nineties M's should be eternally ashamed for blowing the chance at greatness with the Griffey, Rodriguez, Johnson, Martinez, and Buhner M's.  Utterly terrible mismanagement, worse than the Lincoln/Bavasi mismanagement from my perspective.  This is also why I do not have a crush on Piniella... I really believe if he had been more patient with young pitchers we could have developed a bullpen.

Anonymous's picture

The trick is to only pay Stars with more than 2 years. Everyone else should be either a developing young player or Civic rental at 1-2 years. The second year of the Civic needs to be viewed as a sunk cost. If the Civic fails to meet expectations, he should be able to be replaced on the roster (traded or released). The short, fat pyramid paradigm in play with replacements who can play multiple positions. We should plan on 1-2 positional player failures and be able to replace them with our current payroll. This works if we have a good pipeline of young, developing players. Ideally, we pay the stars with more money, not more years. If we are successful, players will want to stay with the club and extend their contracts (Jack's "We have a nice story to tell here in Seattle.").
God bless, Michael


As Doc so expertly notes, there are no absolutes here.  If you buy early, you "can" buy late, too.  It's just going to be more painful if you miss.
And the note about waiting for the 2nd crop of youth to sprout means the 1st crop is aging is another absolutely on-the-mark assessment.  BUT ... one shouldn't expect that EVERY home grown success is going to leave at the end of their 6-year apprenticeship.  In point of fact, if you're a top flight organization, then you should expect on paying "some" of your success stories handsomely and for a long time.  Having Chipper and Andruw and Maddux and Glavine and Smoltz didn't seem to seriously hurt the Braves for a decade plus.
The reality is that home grown talent often, (but not always), is willing to sign for under market prices to stay with a club.  Pujols should be the highest paid player in baseball today.  He isn't *BECAUSE* the Cardinals are a class organization who have treated him exceptionally well from day one, and handed him a truck load of money when the time arrived.  The club that DEVELOPS talent is the *ONLY* club likely to get that superstar to play for 80% of value.  But saving 20% of value (or more) on the best player in the game IS, in fact, a boat load of money.
A team that has already won championships might have some hope of getting a 'ring' discount in some situations ... but that tends to be isolated to late career guys who are already clearly on the way down.  But, Seattle doesn't have that plus.  Seattle is going to pay market OR ABOVE for any FA.  That's the reality of their current situation.  Trading for arb-elligibles can limit that exposure -- and targeting the hungry and unwanted also helps.  But, it gets easier to lure the expensive impact players if you ALREADY have a foundation that they can see.  The Ms aren't there -- at least not yet. 
But, if in 2010, Moore and Tui --- or Carp and Saunders -- (doesn't matter much who) -- jumps up and has some genuine success, while Gutz and Lopez continue developing.  Then, looking at 2011, the club is in a position to make a stronger push forward that is a lot less duct tape and paper clips, and looks more like an actual Major League organization building toward a SUSTAINABLE future. 


and that's exactly Capt Jack's stated plan, BTW.  He traded for Wilson, e.g., with the idea that he could convince Wilson to put down roots here.  That's what happened.
Same with Branyan, who was pleading to return.
Absolutely, hometown discounts for org players is part of the program :- ) ... look forward to your stuff as always San-Man...

Taro's picture

The beauty of that Pujols deal is that the Cards are only paying like 40% of his value.
Pujols is really, really good. :-)
Thats one pricy FA I'd love to see the Ms pursue. Hes so good that its going to be difficult to hand him a contract that overpays him, even in FA. The Yanks already have Texiera at 1B right?


His AVG is exactly what Ichiro's is, bolstered a bit by the HR's out of play, of course.
Imagine contact like Ichiro's when you're swinging for an HR every pitch.  Pujols' success is hard to grasp.
I could argue that you'd be better off with 1 Albert Pujols and 2-3 league-average players, than with 3-4 Carl Crawfords.  Pujols is essentially worth 3-4 minor stars by himself.
I'd like to see the parallel universe in which Ichiro weighed 200 lbs.

M's Watcher's picture

Branyan may make his 2/$10m in 100 games, but what about the remaining 62?  Park his backside on the DL for the last third of the season.  Actually, he was awful after June, not even 100 games into 2009.  What was his value after that?  We can't afford an encore, not at a BIG bat position.

Anonymous's picture

Pujols offensive value using SLGBBSB
2k9 Pujols
.658 + 115/(115+568)+ (16-4)/(115+186)
=0.86624192661844607773989094429014 base/AB
2k9 Ichiro
=.465+32/(32+639) + (26-9)/(225+32)
=0.57105577365799347045758986819139 base/AB
2k9 Carl Crawford
=.452+ 51/(51+606) +(60-16)/(185+51)
= 0.71606624874235740267781131491371 base/AB
1990 Rickey Henderson
=.577+97/(97+489) + 65-10/(97+119)
=0.99715863986853747945898116546581 base/AB
Basically Rickey Henderson found a way to get on base regardless what the pitcher did...
So yeah Pujols isn't worth two crawfords. Comeon do the math! ;-)
Hmm I really should make a excel macro for this... 


He's working to strengthen his core muscles this I don't expect him to tail off that badly again...but obviously it's possible.  Which is why he can't ask for zillions of dollars despite 40 HR potential.


At least above a certain threshold, value becomes non-linear.  So you can't compare Pujols' .880 score to Carl Crawford's .760 score and go "yep...Pujols = 8/7 of Crawford!"

M's Watcher's picture

This is another case where playing through pain hurt the team.  It took him and the team two months to figure this out, far too long before taking appropriate action.  But when you think that you have a huge drop off in talent without him, you lie to yourself that the guy can work through it.  So the problem starts with a lack of talented depth.  What isn't to like when he's healthy?  OK, so he's not the greatest glove.  I'll forgive that.

M's Watcher's picture

You know I understand.  I would have rather paid him $200k/game for the first half and made him sit for the second half on the DL, watching Carp (not Hannahan) play the second half.  Please, just take his second half stats off the books.  What's a .600-.700 OPS 1B worth for two months?  Sorry, that is Sexsonesque 2007-2008.

Anonymous's picture

"At least above a certain threshold, value becomes non-linear. So you can't compare Pujols' .880 score to Carl Crawford's .760 score and go "yep...Pujols = 8/7 of Crawford!" Well simple kinematics( will tell you v0+v1dt. Which of course has t being fed a square of itself so yeah their is your non linear value. And you can also say that If you put a entire lineup of Pujols then you can't really pitch around Pujols which will lead to more hits overall. How significant would this be? Well if Pujols never walked, you could estimate a lineup of Pujols ceiling by just taking OBP-AVG*(0.25*2+0.5+0.25*4) and then add to current SLG which would translate the overall slg to 0.89. Factoring in the sb of 0.03986710963455149501661129568106 you would get ~0.902 bases / ab. Oddly enough that's just a 6% difference from his current SLGBBSB. But of course no team can afford a lineup full of pujols. Now keep in mind baseball stats always have at least two players(one for each team). This of course means within this information you are capturing the moves done for each respective side. For example pitcher intentially walks Pujols. Which of course means that if you don't have a lineup stacked with sluggers then the running game on bases needs to be weighed equally(since the pitcher has the choice whether to pitch to pujols). And like hitting, running around the bases always has a non linear dynamic in the sense that if you are 10 percent faster you have a bigger window of oppurtunity. So yeah I think their is the fact that more power will allow the player to have more hr oppurtunity which is the same as the faster runner have the more SB oppurtunity but the baseball data is capturing two moves at once. And even if you eliminated the BB's all together from Pujols(which is a difficult feat) you would only gain 6% which isn't a big amount in the context of the SLGBBSB total. However let's look at one more player that could benifit most from what your trying to describe. 2001 Barry Bonds. Let's assume you have a whole lineup of Barry bonds and you can't pitch around him. .863 + .515-.328 *(0.5*4+0.25*2+0.25+1) ~ 1.19025 bases / AB. In other words, you want to walk bonds because otherwise he will get more then a single base.

Add comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd><p><br>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.


  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.