THE BOOK--Playing The Percentages In Baseball

9 years, 286 million, A-Rod forevermore spelled with a Halo over the A.

At the *most*, I see something in the nine-year, $260 million range, and I would not be shocked to see him get a lot less than that. Maybe Boras figures by shopping A-Rod to the Mets or Sox, he can get the Yanks on a bidding war, but I don’t see it.

After almost 20 votes, the median has already settled to
(Only look if you haven’t voted. 
Highlight from here: 8.2 / 267 MM (32.5 per): to here.)

You will see once the final votes come in, that the numbers won’t move much from that.  This is what I try to convey with the balloting.  That if you get 100 or 300 people, you’ll still get means somewhere close to what you’d get from 20 people.

EDIT: made a fix in the spoiler section.

Median was too difficult.  Here’s the weighted average after 33 votes
: 7.7 yrs, 250MM, 32.4MM/yr:

(Highlight between the two colons.)

Voted closest to my entry in JC’s contest (see link in name), which was 8/235.

Or I could mess up the link in the name… let’s try this again…

http://www.sabernomics.com/sabernomics/index.php/2007/10/what-will-a-rod-get-2/

Here’s the weighted average after 60 votes
:7.67 yrs, 241MM, 31.4MM/yr:

(Highlight between the two colons.)

I voted 5/175.  I would guess that the variance in dollars/yr is small.  The real question is time commitment and risk.  8 years is a long time for a 32 year old.

#8, I voted 8 years initially, but afterward had second thoughts.  I’m doubting teams go past 6 years for him this time, but the $ per year could get interesting.  I’d go with 6/200 after second thoughts.

It seems like this is more of a qualitative question than a quantitative one.  What are the factors that go into our guesses?  I think the most important variable is the number of teams in the bidding.  With the Yankees out, and presumably the Red Sox out (since they’re very opposed to long contracts, declining fielders, and old people), the degree to which Alex’ contract reaches his true market value is going to depend on how many bidders there are, I’d think.  I see the Angels as potential bidders, but who else is really going to make a stab at signing him?  Unless Boras gets the Angels to outbid themselves (as Bill Simmons might say), I’m not sure this signing is going to reach the mammoth expectations that it has.  Other factors include my belief that Boras and ARod probably want at least $30/year as an average salary as sort of a psychological level of achievement (similarly, if it’s at all possible, he’ll shoot for >$252 million for the life of the contract so it can be the new “biggest contract ever").

A true market value can mean anything.  If all players were free agents, there’s no way his true market value will be 4.5MM per win.  If he’s the only free agent, and there’s two teams needing an infielder, both are contenders for a pennant, and both have a surplus in their budget, his true market value will be 6MM per win.

The “market” is whatever seems to exist at the moment.  Market is not the right word, certainly not if you are trying to tie it in to stocks, futures, or other products or services.

Here’s the weighted average after 68 votes
:7.6 yrs, 240MM, 31.5MM/yr:

(Highlight between the two colons.)

***

In terms of Fantasy dollars: a 30MM per year, where the average team payroll is 85MM means paying him 92$ for a 260$ team payroll. 

The question is not if ARod is worth it.  He’s not.  No free agent is worth anything they get.  The question is how much of the budger surplus do you give free agents?  According to Studes’ article:
http://www.hardballtimes.com/main/printarticle/2007-net-win-shares-value/
Free agents got 1.7 billion$.  And, the nice rule of thumb is that free agents get double what they should get based on what they produce (i.e., you are paying Lexus prices for a Camry with a nice finish).  So, there’s about 900 million$ that teams are literally throwing down the toilet, as they fight each other for the better finish.  That is, there’s 30 million$ of inefficiency per team (on average).  Some teams don’t dabble in free agency, so they save that money.  Other teams don’t.

If you pay ARod 200 million$, that’s 100 million$ flushed down the toilet.  Teams have budgeted themselves to flush 30 million$ down the toilet every year.  Some team will flush more than 100 million$.  They are literally preventing themselves from dabbling in the free agent market for two additional years.  (On average anyway.)

Here’s the weighted average after 81 votes
:7.5 yrs, 236MM, 31.5MM/yr:

(Highlight between the two colons.)

Here’s the weighted average after 92 votes
:7.435 yrs, 234MM, 31.5MM/yr:

(Highlight between the two colons.)

If we compare to post 4 after 33 votes, we see that as time goes on, people have lower expectations. Not that much lower mind you, just enough that you notice.

Ok, after 99 votes, you guys think:

7.4 yrs @ 31.5 = 233MM

This means either a 7/221 or 8/252 deal.  This implies a 7 WAR player.

From 2002-2007, his WPA/LI has been +31 wins in 6 seasons, or +5 WAA.  In that time period, he was an above average fielder.  You can give him +0.5 for his fielding and position.  Knock out 0.5 for aging, and add in 2 for replacement level, and you get 7.

Two different ways, same result.  Let’s wait for the third way (owners).

I get 5.4 WAR.  Above average, -3 runs for fielding, +2 for baserunning, +41 for hitting, +1 for positional adjustment (Tango uses +5 I guess), for a total of +41.  I use -18 for replacement level (these are all per 150 games, not 162), so that is 59 runs above replacment which is around 5.4 wins above replacement, using 11 runs per win, which I think is closer to accurate in the AL.

Interesting.  You use 1.6 per 150 for repl level, while I use 2.1.  That’s 0.5 wins right there.

For fielding+position+baserunning, you get a 0, and I give him +0.5.  (I figure he’s an average fielding SS, which is 0 plus 0.5 bonus, or that he’s an above average fielding 3B, which is 0.5 with no extra bonus.)

That’s 1 win right there.

***

However, if you are using 1.6 as opposed to 2.1 for your repl level (per 150), then you $ per win has to compensate.  That is, for nonpitchers, I have 2.25 wins per 162, meaning that a team of nonpitchers (8.65 per team), that’s 19.4 wins above replacement.

In your case, 1.64 per 150 implies 1.77 per 162, or 15.3 wins above replacement.

Since the dollars above replacement must be exactly equal, then you must value each win 19.4/15.3= 1.27 times more.  If I’m valuing it at 4.4, you are valuing it at 5.6.

And 5.6 $ per win x your 5.4 wins is 30 MM per year.

Note MGL’s replacement level. He calls it 18 runs per 150 G, which as we’ve seen is about 15 to 16 wins per 162 G for 8.65 nonpitchers.  Basically, MGL has the replacement level nonpitchers at around .405.

He’s also called the replacement level pitcher as a starter at 0.75 runs below the average starter.  With the average starter at the .485-.490 level, that puts MGL’s replacement pitcher as a starter at around .415.

He hasn’t described his replacement pitcher as a reliever, but .485 would probably be about right for him.

Two-thirds of .415 and one-third of .485 gives us .438 for all pitchers.

***

So, MGL has the total nonpitchers as being +.095 wins above replacement per game and the total pitchers as being +.062 above replacement per game.  (40% of the value being pitchers according to this).

Anyway, the total average players are +.157 wins above replacement per game, while I use +.210 wins.  So, each MGL win is worth about 34% more than each Tango win.  That makes an MGL free agent win 5.9 wins.

ARod is worth 32MM according to MGL, which is very similar to what I have.

The chatter is for ARod to return to the Yanks for 10/275.

I just re-updated my model to include the 380,000$ minimum salary (on my computer, not website); a 7 WAR player, which I previously said was 8/252 is now 8/255.  Not a big difference obviously.  Anyway, a 10yr contract would imply 310MM$.  Presuming that the Yanks follow my model to a tee, if we subtract the 21MM house money that the Yanks lost, Yanks would go for a 10/289 deal, and be in the exact same position that they wanted to be in.

If they sign him for 10/275 (meaning they would haev done 10/296 with the Rangers money), that would imply a WAR of 6.8, which is quite fair.

ZIPS forecast for ARod for the next 10 years:
http://www.baseballthinkfactory.org/files/oracle/discussion/2008_2017_zips_projection_alex_rodriguez/

What I did was translate those numbers real quick like this:

_1_ _2_ _3_ _4_ _5_ _6_ _7_ _8_ _9_ _10_
2008 4.9 +0.5 683 4.8 +0.5 2.2 7.5 4.4 33
2009 4.2 +0.2 683 4.1 +0.2 2.2 6.5 4.8 31
2010 3.4 (0.1) 649 3.2 (0.1) 2.1 5.2 5.3 27
2011 3.0 (0.4) 639 2.7 (0.4) 2.1 4.4 5.9 26
2012 2.5 (0.7) 628 2.2 (0.6) 2.0 3.6 6.4 23
2013 1.8 (1.0) 606 1.5 (0.9) 1.9 2.6 7.1 18
2014 1.4 (1.3) 549 1.1 (1.0) 1.8 1.9 7.8 15
2015 1.0 (1.6) 493 0.7 (1.1) 1.6 1.2 8.6 10
2016 0.4 (1.9) 447 0.3 (1.2) 1.4 0.5 9.4 5
2017 (0.) (2.) 396 (0.) (1.) 1.3 (0.) 10.4 (1)
TOT xxx xxx 5773 20 (5.9) 19 33 70 187

This is what each column means:
1. YR: forecasted year
2. OPSwins/162: (1.7*OBP+SLG-1)*.025*700
3. FldWins/162: start at +0.5, and drop by 0.3 per year
4. PA: forecasted PA
5. OPSwins: take the value in OPSwins/162, divide by 700 and multiply by forecasted PA
6. Fld: same thing
7. repl: 2.2/700*forecasted PA
8. WAR: add up the above three fields
9. $/win: start at 4.4, and increase by 10% every year
10. sal: value of performance

Under this scenario, ARod is worth 187 million$ over the next 10 years.

My standard aging is to simply drop the value by 0.5 wins, which is a rule of thumb.  I start at 7.0 WAR for ARod (Dan has 7.5), and in 2009 I’d have him at 6.5 (same as Dan).  Then I go to 6, whereas Dan already has him at 5.2.  Then I’m at 5.5, and Dan has him at 4.4.

Essentially, Dan forecasts a drop of 0.9 wins per season, whereas my rule of thumb implies 0.5 wins per season.

When I look at my more rigorous model, which looks at the specific age and specific talent level in question, ARod drops his WAR at a rate of 0.7 wins per season.  Under that scenario, he’s worth 10/225.