THE BOOK--Playing The Percentages In Baseball

One thing to note is that while he’s always been an RF in Toronto (and always would have been as long as the albatross was next to him,) he will apparently play CF in Chicago.  Depending on how well he can play CF every day, that could change things I imagine.

The fielding+positional value presumes that he will be able to leverage himself similar to anyone else.  If he’s been horribly miscast as a RF, and has superspeed that can be leveraged in CF, then yes, he’ll get a gain.  Or, if he’s been greatly cast as a RF, and doesn’t have enough skills to leverage in CF, then he’ll be at a loss.

I’m trying to wrap my head around this. So his remaining value and remaining contract are about the same (62 million), which means the Blue Jays should be able to replace his value as long as they spend the same amount of money? Or are the Jays assuming that they need to have players that bring more value than their contract is worth for them to be competetive on their budget?

They should be able to get that kind of deal on the free agent market, no different than you buying a house at fair market value.

The only benefit the Jays really get here is to remove a contract obligation, if budgeting was a concern.

The Jays might be deciding that they can’t compete next year, so getting rid of one year’s worth of Rios is money in the bank (assuming little attendance decline).

I’m surprised more teams don’t try that on a larger scale: spend $0 one year, and $200 MM the next year, so as to at least make the playoffs occasionally.

I think teams do do that, but not to such an extreme scale.  The fire-sale Marlins for example (exception?) did it.

I wonder how easy it is for a team to go from minimum salaries to 200 million?

There aren’t always enough really good players in the free agent market.  And the ones that give you the great value - like CC and Tex last year, demand 6-8 year deals.  There’s a big risk in thatone of those players loses his value quickly and you’re stuck with him, like Vernon Wells or Barry Zito.

I don’t want to say impossible, since the Marlins more or less did it, but it seems a very difficult task.

I think the only way to do it is in trades on guys with a year to go. 

For example, you would trade (or sign as FA) these players:
C: Varitek
1B: Branyan
2B: Hudson
SS: Cabrera
3B: Crede
OF: Manny, Cameron, Abreu
SP: Washburn, Harden, Bedard, Lackey, Piniero
RP: Putz, Izzy

I don’t know what their WAR was in 2008 (or currently in 2009).  I’m obviously cherry-picking (but not totally).  And the cost of minor leaguers I don’t know.

But, you might be able to do it, if you are committed to it.

I don’t know how fans would react, but then again, fans are used to turnovers in college sports aren’t they?

You’re putting forward a plausibility argument that the expected value, in $ terms, of Rios’ contributions over the remainder of his contract is roughly equal to his contract.  However, don’t you need to factor risk into the equation? 

In general, shouldn’t a player’s market value be less than his expected value, because there’s uncertainty in projecting future performance and because teams are risk averse?

Gennaro’s WSJ article last week represented one attempt to quantify this theme, arguing that Strasburg should be valued akin to a CCC-rated corporate bond and hence merits only a $11 million bonus.  (Apologies if that article has been discussed here already, I’m not yet a religious reader of this site.)

I don’t agree with Vince regarding the discounting.

Teams are not buying one bond, but 300 bonds (players per team in pro ball), none of which are related in fundamentals or technicals.  So, the diversification of such bonds, as a group, turns a high-risk per player bond into a limited-risk group bond.

If a team were all buying the same kind of bond, say all their best players are 21-23 years old with zero MLB experience, then maybe I can agree with Vince.  Or they are all 32-35 years old in MLB.

But, that’s just not the way teams are built.

Put differently, you’re saying that there’s a portfolio effect involved in assembling a MLB team, much as there is in assembling an insurance company’s book of business, and that therefore the team is diversifying away much of the player-specific risk, which thereby minimizes the amount of risk adjustment you need to make to a player’s expected value. 

That’s a very interesting thought, worthy of further rumination.

Right, you said it well in only one sentence!

Just remember that you are GUARANTEED to have 2430 wins to be generated every year.  It’s even better than insurance, because you could have a bunch of people die disproportionately from year-to-year.

I really, really want to make a crack about AIG or mortgage bonds, but Tango is making too much sense here.

As long as we’re confident that fans will watch MLB at about the same rate, and individual team attendance rises/falls with their win totals, there is virtually no risk in what MLB pays to the population of players.  Even if a huge group of superstars get hurt all at once, you’re covered, as their replacements only make the league minimum and you’ll still wind up with your 2430 wins.

I said something like this on THT Live and got a handful of responses from people who believe the free agent market has completely collapsed. I said the same thing about using the 4.4 number in the future, and these commenters were convinced that the number should be even lower than that.

I don’t remember where you figured out the dollars per win in the past (if you could link it, I’d appreciate it), but do you (or anyone) know what the number came out to for this past off-season? There’s more than one way of doing it, but I’ll take what I can get.

I used studes’ work as one basis.

Basically, you just work it out.  There’s 77MM of marginal dollars per team’s payroll.  There’s 33 WAR for each team.  That gives you about 2.3MM per win.

Free agents get paid almost double the “normal” rate.  2.3 x 2 = 4.6

I had used 4.4 for 2008.  And that’s probably what should be used for 2009 and 2010.

Just play around with it, and come up with something reasonable.

Assuming the team in #8 started with enough talent to trade for the guys who were not free agents, like Lackey, Washburn, Cameron, etc., and filled out the bench and bullpen with replacement level players at the league minimum:

They’d spend 116 million on the 25 man roster, plus an additional amount for fill ins when guys are on the DL, suspended for roids, or whatever.

They’ve accumulated 28.5 WAR by Fangraphs, so with replacement level at .300, this team would be 62-50 through 112 games.  They’d be behind the Dodgers, Yankees, and Angels if in those divisions, but contending for the division or wild card otherwise.

If they had convinced themselves to spend a bit more on 1B, and would up with Giambi instead of Branyan, they’d be 4 games worse, and not have much to look forward to except a ton of comp picks.

NIce post, as I’ve been thinking along similar lines… generally speaking, that is. My own comments were that even if you assume that the FA market has fallen back the 2006-2007 levels (Tango’s own chart), if Rios is a 3.5 WAR player with a “normal” decline curve (0.5 per season), then the contract is about market value.

Still, I’m not sure he’s 3.5 WAR. I’m curious as to where the 0.5 win baserunning comes from. I’m not doubting it, but I was away from my laptop, and when I checked his EQBRR numbers, he was never more that 1 or 2 runs above average per season… I’m sure there are other baserunning numbers out there that I’m not seeing. Rally has him at between 0-2 runs every season but 2007 (+4), and I assume that’s the kind of thing that’s going to be declining at Rios’ age.

He’s +8 total for baserunning over 5 years, but -3 for hitting into DP’s, and -1 for reaching on errors.  Speed overall is worth only about 1 run per year in his case.

I do have him averaging 3.4 WAR per year though, and that includes TotalZone which does not like his defense as much as more detailed metrics.  So I can see bumping him up to 4 WAR or so, but then declining each year from there.

He has 112 SB and 33 CS in 5 full seasons.  That’s about +0.2 wins.  I figured a bit more for running the bases.  I probably should have said “+0.25?” instead of “+0.50?” .

Regardless, I was more interested in laying out a reasonable estimate for him to discuss the point that his contract is pretty much equivalent to his expected value, so that half the teams could have been justified in picking him up, and the Jays were justified in dropping him.

’K. Thanks. I agree about his value and the principles in general (although J.P. is getting roasted up here for it). I was just wondering if I’d missed something with regard to his baserunning.