THE BOOK--Playing The Percentages In Baseball

And while I won’t add the Cubs to the opposite list yet, for signing Ramirez, another deal like this, and I will:

For whatever reason, mainstream analysts simply fail to grasp the importance of defense, especially at the extremes. Beltre’s one of the best gloveman at third base around and Ramirez is probably the worst. That fact wipes away the whole of the difference between their 2006 offensive performances.

Cub Fans have Ramirez as a somewhat below average 3B, while Mariner fans (and fielding metrics) have Beltre as a top-fielding 3B.  Here is Ramirez against the average 3B, according to the Fans’ Scouting Report:

In.Fi.Sp.Ha.Re.St.Ac.
56 53 48 53 53 60 55 MLB average 3B
55 41 29 47 41 70 51 Ramirez

Other than his arm strength, Ramirez is a step behind the competition, fielding-wise.  Beltre is probably +1.5 to +2.0 wins ahead of Ramirez with the glove.  Hitting-wise, Ramirez was +2.2 wins ahead last year, and +1.3 wins over the last 3 years.

As Dave rightfully points out, that’s a wash.  Someone who is +2 wins above average should get about 5 million$ above average.  A full-time average position player should make 5 million, so that means 10 million per year.  Offering 73/5 is crazy, but typical, for a free agent.

***

It also seems to me that MLB must be thinking revenue are going to grow at a faster clip that ever before.  Maybe MLBAM money is coming in?  Soon, MLBAM will be worth more than all 30 MLB teams combined (welcome to the NFL party, MLB, just 40 years late).

The marginal $/win may be growing quite fast, as is the average salary, so that in 5 years, a 73/5 deal may seem reasonable.

***

Going back to the Beltre/Ramirez comp, this is similar to my comp of Polanco/Jeter.  The gap in hitting and fielding of these guys is similar.  (The only other relevant consideration is the durability factor, and, I’m not convinced either way we can estimate that reliably.)

Some people simply refuse to believe the amount of impact great fielding can have.  That the Astros have Everett in the field shows me that the Astros do believe it.  Redsox fans seem to not believe that, with their calls to oust Gonzalez.

I would take Beltre over Ramirez.  I may be in a minority of 2 on this (with David Cameron).

I agree the defensive advantage of Beltre cancels the offensive advantage of Ramirez, but what if Beltre hit as well or better than Ramirez?

I don’t quite know what to make of 2004.  Throw it out, and Beltre = Ramirez.  But I wonder if Beltre can recapture some of that magic, and hit .280/.350/.525 or so.  If he does that he’s a top 20 player.

Its kind of like Erstad in 2000.  Eventually I had to give up any hope that anything like it would happen again.  But its too soon for me to give up on Beltre.

Ramirez is getting paid 4.5 million per marginal win.  Valentin is certainly a bargain at around 2/win.

The 4.5 might wind up being reasonable given all the cash out there.  That would mean Pujols is worth over 30 million and the next group of elite players (Beltran, A-Rod, Sizemore, Mauer) are worth a little over 20.

At that level JD Drew is worth 15 million, while Soriano and Carlos Lee between 10-12.

For what its worth, I wasn’t even really trying to criticize the Ramirez signing that much.  I don’t think its an awful deal.  It’s not one I’d want to sign, but it’s not a brutal overpay like the Carlos Lee contract will be. 

Really, my point was (as always, since I’m a broken record on this issue) that the Beltre signing wasn’t a bad one.  I just used Ramirez’s contract as a platform to get back on my soapbox. 

The new revenue coming in is certainly going to inflate the costs of talent, and we’re going to have to adjust for that.  Beltre at $39 million for the next 3 years is probably a better bargain than what any of the “premier” free agent hitters are going to sign for this winter. 

And guys like Sheehan, Law, and Neyer will still call it one of the worst contracts in baseball.

I have ARam at -5, -5, -8, -9 in defense the last 4 years.  I think -5 is a reasonable projection.

Beltre is 12, 3, 24, and 12. 11 might be a reasonable number for him.

So let’s call it a 1.5 win difference.

ARam in off lwts was 17, 25, 31, 6 the last 4 years.  +16 is reasonable for 07.  He is a terrible baserunner.  -4 for that.

Beltre is 6, -4, 49, -10.  +6 is reasonsable for next year. His baserunning is around -1.

So ARam has only a 7 run edge in hitting, which gives Beltre an 8 run advantage overall.

BTW, I have ARam at only 7 runs over the average third baseman, which is, say 2.5 wins over replacement, 2.7 if you want to be generous (and call “replacement” 20 runs below average).

That is 5.5 mil per marginal win, which is in fact, a very bad contract.

By my calculations, $4.5M was the marginal cost of a win from free agents last year, let alone next year or five years from now.

Seems to me that the other thing going on with Valentin is the whole second base thing.  Either GM’s undervalue second basemen, or the replacement level is higher for second basemen that other positions.  Question: if you could get Loretta or Valentin at the same price, who would you prefer?

My choice would clearly be Valentin, more for the glove than the bat.

The Beltre contract was an average free agent contract at the time, meaning it was a typically bad contract (i.e., no bargain like Craig Counsell, but no bust like Cristian Guzman… Beltre was the going rate for a class of players with whom demand exceeds supply.)

Valentin has been underrated for years, basically because no one believes he is such a good defender.  Most people believe he is a BAD defender or just don’t know anything about him.  Somehow he has managed to sneak under the radar for some time now.  I tried to get the Cards to sign him at SS or 2B a few years ago.

Corner OF’ers and first basemen are the predominate players who are overpaid, because most teams fail to understand or simply cannot help but ignore or underestimate positional differences.  Therefore most of the bargains are going to be at the other positions.

And yes, Lee is going to be grossly overpaid, as was Konerko, yet few people other than me believed that.  Of course I look bad with him having a very good year at the plate (+25 lwts) when I had him projected at +13.  I’ll take under 20 next year for any amount of money if anyone wants to wager.

Gonzalez, 3/14:

Fans see Gonzo as one of the best fielding SS:
http://www.tangotiger.net/scouting/pos2006_SS.html
The Fielding Bible has him as slightly above average.  Let’s call him almost +1 win, relative to the average SS, or +1.5 wins relative to the average fielder.

His hitting is one of the worst in the league, meaning -2 wins relative to the average hitter.

So, he’s a -0.5 wins relative to average.

The average players get 5 million a year, per 162 GP.

So, you start at about 4 million, and subtract 1 million, making 3/9 as fair.

A 3/14 is a typical FA deal, perhaps better than typical.

GMs love HR.  How else to explain Soriano?  Here are your choices: sign Beltran at age 28, to play CF, for 7/119, or sign Soriano at age 31, to play OF for 8/136.

Soriano is a good hitter, a +1.5 to +2.0 wins above the average hitter, Beltran is a bit better hitter, a +2.0 to +2.5 wins above the average hitter.  Beltran is also more clutch, for whatever that is worth to you.  Let’s call it a +0.5 win advantage to Beltran.

Fielding-wise, there’s no comparison.  Beltran is one of the best fielders in the league, and Soriano would be a below-average CF and above-average LF.  This is probably a +1.5 win advantage to Beltran.

Beltran is likely the superior baserunner.  For basestealing, Beltran is 35 SB and 11 CS, while Soriano is 71 SB and 19 CS.  Assuming Beltran’s been injured to the point that his outstanding pre-2005 basestealing won’t come back, let’s give Soriano that one.  So, running-wise, let’s call them even.

Overall, that’s +2 wins to Beltran, meaning that he’s worth +4 million$ in an overall market, and +8 million$ per year in free agent prices. 

If Beltran’s 7/119 can be classified as pretty much the most you should have paid two years ago (not a bargain, but not outlandish), and if we add in 20% in baseball-inflation, that puts Beltran’s contract at 7/144 today.  An 8th year would have put Beltran at almost 20 million/yr in today’s dollars.

Soriano should have gotten 8 million/yr less than that, or 12 per year.  At 8 years, that’s 96 million$, which means the Cubs overpaid by 40 million$.

Tangotiger.

Just to clarify you are using $2M per marginal win and $4M per FA marginal win. Do you have a link you can post to where you calculated those numbers.

Thanks
Alex

Another way to look at this is to compare Soriano to Juan Pierre.  Knock out Pierre’s rookie year, and he has 4207 PA (excluding IBB, SH).  Soriano’s career is 4179 PA. 

Let’s see how they compare, performance-wise.  Soriano has 196 more career HR and 50 more doubles+triples.  Pierre has 42 more walks+hitbatters, 337 more singles and 103 less outs.  That’s a 108-run hitting advantage to Soriano. 

Pierre had 43 more triples, giving him an extra 13-runs here. For SB, Soriano is 210-60, and Pierre is 318-110.  This one comes out as a wash, a bit more toward Soriano.  For baserunning, Soriano is a very good baserunner, but Pierre is one of the best in the game:
http://danagonistes.blogspot.com/2006/09/more-baserunning.html
This gives us a career advantage of 10-15 runs to Pierre. 

These three running categories is about +20 for Pierre.  That’s makes Soriano about +90 offense, on 4200 PA, or +15 runs per year, or +1.5 wins per year.

Fielding-wise, Pierre is an above-average CF and Soriano would be a below-average CF (above-average LF).  Though Soriano has the arm.  Let’s give Pierre only a +0.5 win advantage here.

So, Soriano is +1 win ahead of Pierre, per year.  That’s +4 million in free agent dollars.

I don’t know what Pierre is going to command, but I’ll guess 8 mill per year, putting Soriano at 12 mill per year.

***

Put another way: you are better off with Beltran and Juan Pierre than with two Sorianos.

Alex/11: yes, that’s correct.  As for a link, try doing a search on this site, or looking through threads here:
http://www.insidethebook.com/ee/index.php/C14/

You’ll also find useful data at Hardball Times (look for studes’ articles).

***

The “2 mill” is easy to figure. 

A replacement level team will be .300, and needs about 10-12 million$.  The average payroll is say 75-80 million$.  So, the extra 65-70 million$ will buy you (.500-.300)*162=32.4 wins, or about 2 million$ per win.

That’s the one-paragraph explanation.

Assuming 10% annual inflation, Soriano will be paid $12M per year (current dollars) over 8 years.  You seem to think he’s worth about 4 wins above replacement:  1.5-2 wins above avg on offense, good baserunner, and about a league-average fielder.  So the Cubs are paying $3M per win, which is $1M less than your estimate of the current FA market rate for wins.  What’s your complaint?  All this really shows is that the Beltran contract was perhaps a bargain. 

(Now, if we assume declining performance with age, as we should, maybe it becomes $4M per win, but still probably close to fair market value.)

Tango - thanks.

If I drop Soriano by 0.5 wins per year, and start him at +4 wins above replacement, his 8-yr total is +18 wins. 

A perpetual league-average player would be +16 wins over 8 years.  If I start the league-average free agent price baseline at 8 mill per year this year, and add 10% each year, that gives me 91 million$ that +16 wins will buy me.

Soriano is therefore being paid 45 million$ for those +2 wins over the life of the contract.

On the other hand, if his decline is 0.25 wins per year, then he’s at +25 wins above replacement, compared to the perpetual-league-average player of +16.  In that case, he’s getting 45 million$ for those 9 extra wins, in which case, that’s ok, in a free agent world.

***

If we had a Beltran-type talent, where he’s starting at 2 wins ahead of Soriano, a 0.5 win decline per year means that a superstar would be +34 wins over those 8 years.  That would be worth over 200 million$.  Either Beltran was a steal, or Soriano is way over.

Actually, that would be 183 million$.

***

The same calculation that makes a +2 wins above average player in his decline phase a 92 million$ player makes a +4 win above average player in his decline phase a 183 million$ player.

***

Here are contracts for various quality players, assuming decline phase, 4 million$ in today’s dollar per free agent win, and 10% baseball-inflation:

+1 above replacement: 2yr/6 mill
+2 above repl (i.e., avg): 2/15, 4/22
+3 above repl: 4/41, 6/50
+4 above repl: 6/81, 8/92
+5 above repl: 6/111, 8/137
+6 above repl: 6/142
+7 above repl: 7/173

Soriano is a +4 being paid like a +5.  Therefore, if someone wants to argue that Sori is actually a +5 and not a +4, that’s where you can win the argument.

Using the same process, Gonzo’s 3/14 implies that he’s -0.4 wins relative to average (I had him a -0.5).  Therefore, Gonzo’s contract is in-line with this framework.

***

Ramirez’ 5/73 means he’s valued at +2.1 wins above average.

Beltran’s contract two years ago can be justified as him being valued at +3.5 wins above average, and in his decline phase.

***

Those guys from two years ago, who got the 5/65 deals can be justied if they were valued at +2.3 wins above average, and in decline.

I’d put Soriano closer to a +3 than a +4.

But just wait till Carlos Lee signs.  He’s a +2 who might be paid like a +4.5

Matthews signed for 5/50.

He is a slightly above average hitter, maybe a bit more than +0.5 wins.  His career BABIP was .304, including last year’s .349.

His fielding is either above average (Dewan), or fantastic (Fans):
http://www.tangotiger.net/scouting/pos2006_CF.html
Let’s call him +1.0 wins as a fielder.

I pay a premium for a CF, an extra +0.5 wins.  All told, a +2 wins above average player (+4 above repl), in his decline phase. 

Based on post #17, a +3 above repl, which is 1 win worse than Sarge, should get a 4/41, or 6/50 (meaning about a 5/45).  That he got a 5/50 seems reasonable, and depending how real his 2006, a bargain (in free agent prices).

Way more optimistic than I have him.

I see at best an average hitter and average fielder.  An awful signing, and from what Red Sox reports are saying, JD Drew would not cost that much more.

ZR and PMR have him as a below average fielder.  I wonder what MGL has for him.

How can you say “at best” average hitter?  You have to go back to his 2003 season to get a below-average season out of him.  So, he’s “at worst” an average hitter.

As for his fielding, Dewan had him as a plus each year, 2004-2006, and Fans had him as avg CF last year, and great this year.  Again, “at worst” he’s an average CF.

The pessimistic view is that he’s an average player.  He’s likely at least a +1 win above average player.

As for JD, he’s a better hitter and worse fielder.  He’s at least a +2 win above average, and better than Matthews.  His injuries work against him.  A +4 above repl, for 2 years, is 2/30, according to the model in post #17.

I think the hard part to understand, which is what Guy is saying, is that 10% inflation will makes a long-term contract look reasonable.  Hard to see that right now, but in 7 years, a 10% inflation means an average 5 million$ a year contract is 10 million$ in future dollars.

I’m looking at projection stats.  I get almost the same thing as Dan Szymborski on BTF.  The aging certainly hurts.  I have him at -2 runs for 2007.  I should say about average instead of at best average.

I hope Dewan’s system and the fans are right about his defense.

By Baseball Prospectus batting runs above average, which is park adjusted, here’s Matthews over the last 4 years:

-11
+2
-3
+16

So given his age, average hitter for 2007 seems reasonable to me.  For this money he better be a great (+10) defender.

Wow. According to your framwork, Tango, Carlos Lee’s 100/6 deal pegs him at about 4.5 wins above replacement. His offense perhaps gets him close to +4 wins above replacement but the fans don’t like him in the field. Even if you think that is a little high it looks like the ‘Stros have overpaid for Lee (perhaps you get +2.5 overall). Though at LF in the Astrodome I suspect that most flyballs will sail over his head for home runs. Have I got that right?

The amazing thing is that they could have signed Beltran for similar money a couple of years ago. What an error that is looking now.

Lee 6/100:

His hitting is +2 wins above average, and Fans detest his fielding for the last two years.  Dewan has him as a bit below-average.  Let’s call him -0.5 wins to be nice.  So, +1.5 wins above average (+3.5 wins above repl).

According to post #17, that’s a 6/65 deal, meaning he’s overpaid by 35 million$.  (Remember, post #17 factors in free agent prices, and inflation, and aging.).  His overpayment is similar to Soriano.

Oops, forgot about the position.  He’s down to +1.0 wins above average, meaning a 6/50 deal.  Sarge is better than this guy.  What a horrible signing.

Here’s handy-dandy chart:

http://www.tangotiger.net/salary.html

Baez, 3/19.  Using the handy-dandy chart, a 3/19 corresponds to a +2.0 wins above replacement.

Funny thing about relievers is that the average reliever is a below average pitcher, but he gets above average stats.  That’s because it’s easier to relieve than to start. 

So, when we look at Baez, he’s got slightly above average stats, but, he’s got a built-in advantage, as do all short-relievers.  The advantage is enormous, almost 1 run per game.  In Baez’s cases, that works out to about 8 runs.  While he’s been getting a leverage of 1.50, he probably deserves a bit less.  Let’s call it a 1-win advantage to Baez.

That turns Baez’s +0.5 wins above average to 0.5 wins below average.  While a full-time pitcher is worth +2 wins above average, that’s hardly the case here.  In short, he’s +0.5 wins above replacement, meaning he’s been over-valued by 1.5 wins.

At least he only got 3 years out of it.

Alou: 1/8.5

This one is very interesting.  He’s a player who has performed at or above his career level, and he’s over 40 years old.  A Hank Aaron if you will.

On a full-time basis, he’s +3 wins above average as a hitter, below-average as a fielder, and he plays an easy position.  Let’s call him +2 wins, in full-time play, and since he’s on a steep aging decline, let’s make it +1.5 wins.  Still pretty good.

We don’t expect him to play full-time.  Marcel thinks he’ll play about 60% of a season.  So, his +1.5 wins above a full-time player, which is also +3.5 wins above a replacement player in full-time play, means that he’s +2.0 wins above a replacement player, if he plays 100 games.

And, what does the handy-dandy chart say?  A 1/8.5 deal corresponds to a guy who is somewhere between 2.0 and 2.5 wins (if you work it out, 2.1 wins).  Bingo!  Good job on the Mets.

***

Shawn Green is a lefty, Alou is a righty, and Endy Chavez is a lefty.  If Green was a righty, and Alou a lefty, this would be a great platoon combination.  I’d want to give the most PA to Alou, then Endy, and then Green.

Green, yech.  Last year, Endy was .348/.431 in 390 PA.  Shawn Green was .344/.432 in 588 PA.  We still have to consider Green as likely the better hitter because of past performance.  Say give Green a 1.0 to 1.5 win advantage with the bat.  But, fielding-wise?  No contest.  We are talking about a 3-win difference between the two.

If they can make all three OF start 100-120 games each (with Endy taking over from Beltran for 10 games or so), this would be an excellent OF.

Juan Pierre: 5/44.

Handy-Dandy says: +2.9 wins above replacement (almost 1 win above average).

As a hitter, he’s -1 win.  Average CF, excellent baserunner.  Overall, a league-average player (or +2 wins above repl).  That makes him overvalued by almost 1 win.  His fielding better be better than I’m allowing.

Delluci: 3/11.5, paying for 1.4 WAR.

An overall slightly-above-average player, who if I consider that he piled his stats almost exclusively against RHP is probably right around average.  Call him +2 WAR in full-time play, and if you consider that he’ll be platooned, will be right around +1.2 WAR.  Let’s say: bingo.  Appropriately paid.

Eaton: 3/24, paying for 2.3 WAR.

Average pitching stats, making him an above-average starter.  Bingo.  Appropriately paid (in free agent dollars), if he’s part of the regular rotation.

Wolf: 1/8, paying for 2.0 WAR.

48 starts over the last 3 years.  Hard to see him as a a full-time league average pitcher, or a great part-time pitcher.  I wouldn’t put him in Alou’s class, but when you sign a player for just one year, there’s not much you can lose.

***

Woody Williams: 2/12.5, paying for 1.75 WAR.

Padres play in a pitcher’s park.  He is 40 years old, but pitched at least average in 3 of the last 4 years.  Another bingo.

If Wolf would have signed 1/7, that’d also have been a 1.75 WAR expectation.  I guess with all these question marks (slightly above-average starters, durability/age), we should see these kinds of deals.

Adam Kennedy: x/yy.

Terms are unknown, but let’s take a guess.  A slightly below average hitter, and above average fielder (slightly above average for a 2B).  Add in some aging, and he’s probably an average player, meaning +2 wins in full-time play.  He’s been getting about 70% of the PA for a while now, so let’s call him +1.5 WAR. 

So, I will guess he signed for 1/6, 2/10, 3/13.  Definitely not more than 3 years.

ESPN has a free agent tracker, so let’s take a look at others I missed:
http://sports.espn.go.com/mlb/features/freeagents

***

Casey: 1/4, paying for 1 WAR.

Average fielding for a 1B, below average hitting for a 1B.  Call him +1.5 WAR in full-time play, or +1 win in expected playing time.  Bingo.

***

Catalanatto: 3/13, paying for 1.5 WAR.

Standard LF, if ever there was one.  Full-time is +2 WAR, meaning 1.5 in his typical playing pattern.  Bingo.

***

Cora: 2/4, paying for 0.7 WAR.

Average fielding SS, poor hitter, below-average for a SS.  1.0 WAR in full-time play, 0.5 in expected.  I’m surprised Boston would want to tie him up for 2 years.  What’s the point really?  To me, players like this, and Bloomquist should be signed for month-to-month deals.  Bingo.

***

DeRosa: 3/13, paying for 1.5 WAR.

Pretty much like Catalanatto in value and expected playing time.  Bingo.

***

Edmonds: 2/19, paying for 2.5 WAR.

Fans still see him as a great fielder, but, Fans are definitely biased to once-great, now aging fielders.  Likely still above-average.  Excellent hitter, but injuries and bad (for him) last year, makes this a tough call.  Without the injuries, and in full-time play, he’s over +4.0 WAR.  Could have reasonably signed him to 2/10 or 2/27.  I wouldn’t want to call this signing, but, Cardinals seem fair about it.

***

Easley: 1/0.85, paying for 0.2 WAR.

Poor fielder, horrible as a SS.  37 years old.  A bit below average-hitter.  Call him 1.0 WAR in full-time play.  If he plays 20% of the time, he’s 0.2 WAR.  I guess he’s a Julio Franco-type situation.  Bingo.

***

Nomar: 2/18.5, paying for 2.5 WAR.

Above-average fielder, pretty good at 1B.  As a hitter, +1.5 to +2.0 wins above average in full-time play.  Overall, his off+def is +2.0 WAA in full-time play, or 4.0 WAR.  Knocking him down to 3.0 WAR in expected playing time.  Seems to me that he’s a slight bargain, unless I’m wrong on his playing time.  A 2/23 would have been fine.

Love the way he’s been working on short-term contracts. 

***

I’ll catch up with the rest of the gang next time…

Kennedy signed for a 3/10, meaning 1.3 WAR, which is a bingo.

Royce Clayton signed for 1.5/1 with the Blue Jays.

I have him as below replacement level, so he should be happy to be playing at all.

In fulltime play I have Clayton at -26 bat, -7 defense, +1 baserunning, +9 for position, and -6 overall.

John McDonald rates at -24 batting, +1 fielding, and 1 run over replacement, but I doubt the projection system here.  He’s never been a full time player, and if forced into that role he would probably wear down or be exposed, and make Clayton look like a good hitter by comparison.

Russ Adams is -14 hitting, -8 fielding, and +4 vs replacement.  He’s also a lefty batter, so an Adams/McDonald platoon could work and looks to be a win better than playing Royce Clayton.

Jays fans have McDonald as a *great* fielder.  I wouldn’t trust a PBP metric against the fans in this case.

McDonald’s defensive projection has more regression to the mean built in than the other guys, since he’s a part time player.  And any shortstop who rates as above average at his position should be considered an elite defender compared to all players.

In any case, I don’t think there’s any disagreement that McDonald is the best defensive SS the Blue Jays have.

It must be a virtual certainty that he’s a great fielder, since he’s never had 300 PA, has come into the league at almost age 25, is now 32, has 1146 career PA, his career OPS+ is 56, with a career high of 75.  His career high in salary is 500,000$. He’s also had SIX different managers on three teams.

That a player can fashion that kind of career is a testament to his glove.

Soriano is not being paid evenly.  This is what he’s going to get:

$8M signing bonus
07:$9M
08:$13M
09:$16M
10-14:$18M annually

The total is 136, but not evenly spread out.

This is the same thing as:
07-14: $16.5M annually

I’m assuming the signing bonus is paid upfront.

For those who don’t understand “time value of money”, read this primer:
http://www.investopedia.com/articles/03/082703.asp

Not sure this is the right place for this, but the rumored J.D. Drew -> BOS signing got me to thinking about his Fans’ Scouting Report results:

2004-ATL: 74
2005-LAD: 75
2006-LAD: 56

2005(30 Ballots-.81 Agreement)/2006(13 Ballots-.61 Agreement) SKILLS:

Instincts: 76/55
FirstStep: 79/48
Speed: 77/58
Hands: 69/52
Release: 72/65
Strength: 73/45
Accuracy: 79/58

Yikes.

Numbers like that make me wonder if JD was playing at half speed in the field to avoid injury and get a new contract.

Or if fans couldn’t stand him.

Numbers like that make me wonder if N=13 gives you a sample size problem.

As I’ve shown, n=13 is very representative of the readers to my site.  I’d have gotten the same response if n=39.

***

Clayton: 1/1.5, paying for 0.4 WAR.

Clayton is one of the worst hitters in the league, a -2 wins below the average hitter.  Clayton, according to Fans, is one of the worst fielding SS in the league.  Dewan has him as below average as a SS.  Overall, that makes Clayton at replacement level.  That makes him worth 0 dollars above repl.

Just noticed that all my salaries on my handy-dandy chart should be bumped up by 400,000$.  Anyway, he’s being overpaid by 1.1 million$.  There is no reason for Royce Clayton to be playing MLB. 

This is an embarrassing signing.

Manny: 2/38, paying for 4.8 WAR.

Not a free agent signing, but this is what he’s got left, and the teams trading for him will need to consider if he’s being underpaid, and if so, then they need to trade some value back to Boston.

Anyway, Manny is of course a great hitter.  +5 wins above the average hitter in full-time play.  He is also one of the chokiest hitters around, likely in the bottom 20 in the clutchiness scale, to the tune of an average of -1 win per season (which is enormous).  His behaviour gives us reason to believe it.  He is a terrible fielder according to Redsox fans and PBP metrics.  His position and poor fielding must knock out 1.5 wins.  I’d guess he’s a terrible baserunner too.  Let’s call him +3 wins above avereage in full-time play, or +5 WAR.  He’s been fairly durable, but missed time last year, so that +5 WAR becomes +4.5.

He’s being paid for 4.8 WAR, so this is a huge bingo.

So, first off, if the Redsox were to put him on waivers, he would be claimed.  He’s actually being paid properly.  Secondly, the Redsox should not be in a rush to trade him.

This is one of those things (placing him on waivers) that made perfect sense a few years ago, and makes no sense now.

Drew: 5/60 to 70, paying for +3.5 to +4.0 WAR.

Unsigned.

He is an excellent hitter, +3 wins above average in full-time play.  Dewan has him as the best fielding RF in the league.  The Fans have concurred, until 2006, where he’s noted by Fans as an above average RF.  Let’s make him overall +3.0 to +3.5 wins above average, or +5.0 to +5.5 RAR in full-time play.  He doesn’t play full-time, though, which knocks him down to say +3.5 WAR to +4.0 WAR.

Whoever he signs with, will give him around a 5/60 to 5/70 deal.  Knock out about 10 million$ for each year less than that.

Let’s see…

"As I’ve shown, n=13 is very representative of the readers to my site.  I’d have gotten the same response if n=39.”

Fair point.  But the drop from 30 to 13 ballots, and from .81 to .61 on consistency, made me wonder if something unusual was going with the voting in LA.

Agreement level overall was fairly high for Dodger fans in 2006, and not so much in 2005.

At the bottom in the agreement level are Kent, Drew, Mueller, Furcal, Lofton, old and/or injured guys.

Guillen: 1/5 to 8, paying for 1.3 to 2.0 WAR.

Guillen is a slightly-above-average hitter, who had a terrible year last year, in under half-a-season.  Fielding-wise, he was below-average for a RF last year, and above-average in 2005 and 2004. 

Who is this guy?  He may end up having the biggest gap in performance in a long-time.  His OPS+ went from around 70-85 in 1997-2002, to 125 in 2003-2005, back to the 70-85 last year. 

A one-year deal is the most sensible when dealing with such an uncertainty about what you are getting.  I think his contract should have been totally-incentive based: a flat 1 million$ deal, plus incentives that could bring him to 15 million$, if he plays like Vlad, or his 2003-2004 self.  A 5 to 8 deal seems to express alot of certainty.

Drew was covered in post #51. 

Lugo: 4/36, paying for 2.75 WAR.

Lugo is a league-average hitter, making him a good hitter for a middle infielder.  According to Fans, his fielding is average to a shade above average, for a SS.  Dewan has him as average to a shade below average.  Let’s call him an average fielding SS.  Overall, that makes him somewhere between +0.5 to 1.0 wins above average in full-time play, or +2.5 to +3.0 in WAR.  Let’s call him +2.5 WAR, if we include a slight playing time adjustment.

A 4/32 deal would have made more sense, but, probably splitting hairs here.  Same issue with Drew.  Redsox get to overspend a little.

***

Other than Soriano, Lee, and Ramirez, it seems that all the signings are “logical” within the universe of “we’ve got a budget to spend, we need to spend it on somebody, and there’s no sense in giving it to slave and arb players”.  The illogical signings all have a HR-bonus attached to them.

Unlike most other analysts who are shocked and dismayed at Matthews, Pierre, etc, color me unimpressed.

Zito: 6/81, paying for 4.0 WAR.

Unsigned.

Barry Zito has thrown 34 or 35 starts every year since 2001, 213 to 231 IP each year.  Durable.  His FIP is 4.32, but his actual ERA is 3.55.  Zito is as extreme a FB pitcher as you’ll find, and as we know, FB have as a byproduct less hits per ball in play.  From 2004-2006, as many runs have scored in all Oakland games at home as on the road.  Oakland pitchers however benefit from good fielding.

All told, his stats is something like a +1.5 wins above average pitcher.

Starters also have a disadvantage against relievers.  A starter with average stats is +.110 wins above replacement level (per 9 innings), while a reliever with average stats (and average leverage) is +.020 wins above replacement level.  Clearly, most of the money should go to the starting corps.

Anyway, a starter with average stats for 225 innings will be +2.75 wins above replacement.

This makes Zito +4.25 wins above replacement, in full-time play.  Let’s knock him down to +4.0 wins above replacement.

What if he signs for 6/96?  That means he’d be paid for +4.5 WAR.

However, Zito will turn 29 in May, meaning his decline phase shouldn’t be as pronounced as the other players.  If we give him a decline of only 0.3 wins per year, a +4.0 WAR gives him a 6/98 deal.

You can reasonably argue Zito as anything between a 6/81 to 6/113, depending if you think he is currently +4.0 or +4.5 WAR, and if you think his decline is 0.3 or 0.5 wins per year.

Jason Schmidt: 5/71 to 83, or 4/59 to 69, paying for 4.0 to 4.5 WAR.

Schmidt is probably at most 0.5 wins better than Zito, but is 5 years older than he is.  So, he’s in a steeper decline phase than Zito.

It will be interesting to see if the free agent market prices them the way the handy dandy chart says it will.

If I would have shown the 3/x to y for Schmidt, I would have shown:
3/46 to 53, paying for 4.0 to 4.5 WAR.

Schmidt signed with the Dodgers for a 3/47 deal (paying for 4.1 WAR). 

Bingo.

***

What other bigshot is on the horizon?

Ted Lilly: 3/26 or 4/31, paying for 2.5 WAR.

Unsigned.

He’s compiled league-average numbers, maybe a shade better, which makes him about a +2.5 WAR. 

If he’s really a +3.0 WAR, then look for a 3/33, 4/41 deal.

Greg Maddux: 2/27 or 3/39, paying for +3.5 WAR.

Unsigned.

He is on a steep decline phase, as far as his age is concerned.

Maddux signed by Padres for 1/10, paying for +2.5 WAR. 

Wow, quite a difference from my off-the-cuff estimate.  A league-average starter with 225 IP is +2.5 WAR.  If you make Maddux a 200 IP pitcher, and make him slightly above average as a starter, you are still at +2.5 WAR. 

I dunno… unless I’m way too optimisitc, this looks like a great deal for the Padres.

Lilly just signed, 4/40.  Probably a little overpaid (it is the Cubs, after all), but looks like another win for the chart.

Agreed, I would call it a little overpaid.  The Lilly/Maddux deals I don’t like.  I had Maddux as 1 win above Lilly, but the market has made Lilly as 0.5 wins above Maddux.  Was I too quick (even moreso than the quick ones I’m doing) in looking at this one?

But, as you said, the Cubs (Soriano, Aramis) are not valuing players the best.

Gil Meche: 5/55, paying for 3.35 WAR.

Is there a more average starter around than this guy?  His WPA is pretty much exactly the league average starter.  We know that the league average starter is +.100 wins per 9 IP above replacement.  If we give this guy 180 IP, that puts him at +2.0 WAR.

This guy should have topped off at 4/22.  A 5/34 deal would have meant you are valuing him at +2.5 WAR.

Misvaluing a guy by 1.35 wins is huge.  Compounding that misvaluation by also giving him 5 years is… well, what’s huger than huge?  HUGE!

Right now, the market has valued, in order:
Schmidt
Meche
Lilly
Maddux

Maddux and Meche need to be swapped.

I think Meche is pretty clearly being overpaid, but at least he’s only 28, so the .5 WAR decline built into your model is probably not quite appropriate.

Thanks for reminding me. 

With no aging at all, he’s being paid at +2.25 WAR.  With mild aging (0.2 wins lost yer year), he’s being paid at +2.70 WAR.

Still overpaid, but not as bad as I thought.

Meche probably has to be overpaid to get him to go to the Royals.  A decent team would probably offer him about what Ted Lilly got.

The Royals are so bad they have a tough time attracting free agents.  Last year they got below average starters in Elarton and Redman.  To get even an average guy, they have to offer an extra year and a little more money.  David Glass has the money.  I’m sure they are sick of losing 100 games all the time.

Bonds signed for 16 million, good for 4 wins over replacement.

I’ve got Bonds at +56 runs per 600 PA.  Subtract 1 win for position + defense, And he’s +63 vs replacement in full time play.  We all know the old man isn’t going to play every day, but the Giants will get their money’s worth if he hits my projection and plays 2/3 of the games.

Great job Rally.  I was shying away from Bonds because of The Chase.  But, I suppose all the extras, The Chase, the breakdown of the body, the chance of indictment all balanced out.

I have Bonds at +6 wins per 700 PA.  -1 win for fielding/position is good.  That makes him +7 WAR in full-time play.  I’d put him at playing 70%, or 5 WAR, making him worth 20 million$ for a 1 year deal.

But, this is a pretty tough call, with the injuries, age, and being at such a high level to begin with.

Next Season

Ichiro: 3/65, add 20 million$ for every additional year, paying for 5.0 WAR.

He’s +1.5 wins above average the hitter, who is a also great fielder.  Putting him in CF will further add to his value, similar to Soriano being put in LF.  I’d call Ichiro a +3 wins above average, meaning +5 WAR.  Who knows, I may even be conservative.

(I’m assuming that next year the free agent price per win increases 10% to 4.4 million$ per win.)

What other big names are there in next off-season?

Pettite: tough call, isn’t it?  I’ll guess he’ll be valued as +4.0 WAR, meaning a 3/46, 5/59 deal.  That is, just like Schmidt.  Why is this taking so long?

***

Erstad.  I like this guy.  Probably -0.5 to -1.0 wins as a hitter, very good fielder.  Has been clutch.  Call him +1.5 to 2.0 WAR in full-time play.  What’s his injury situation?  Do we expect him to be +1.0 to +1.5 WAR?  A 2/6 to 2/10 deal would suffice?

What other big names are there in next off-season?

Vernon Wells, Michael Young, Andruw Jones, Carlos Guillen, Dunn; Smoltz, Mo Rivera, Joe Nathan, Carlos Zambrano, Schilling (I think)…

I may have missed news of some extensions, but I’m pretty sure they’re all up next year.

One thing I wanted to ask about, that took me by surprise in the Manny evaluation and is there again with Erstad: clutch.  We don’t need to get into the whole debate, but I’m just a bit surprised to see it taken to the extent of being a factor in how much a guy is worth.  Are you really comfortable saying you’d factor clutch into a contract valuation?

Erstad is clutch, I have no doubt about that.  But over the last 4 years he’s tried to play CF twice and couldn’t make it through half a season.

As a 1st baseman he’s still a great fielder but his bat makes it hard to even want him on the roster.

I’d be happy with him in a backup role, but no more.

Wells and Andruw are going to make some staggering piles of cash.  Enough to make Soriano jealous.

Center fielders seem to be valued above shortstops.  How else to explain Pierre getting more than Lugo?

Dan, I’d be comfortable with a +/- 0.5 wins at the extreme, and typically inside the +/- 0.2 wins, for clutch.  Considering I’m doing off-the-cuff estimates, and I’m always reporting to the closest 0.5 wins, I think that’s ok.  i.e., when it’s close, I push them up or down to the next 0.5 level.

***

Wells and Andruw will be interesting, especially Wells.  He’s 1.5 years younger, and we’ll see how real his power his this year.  Wells is likely the better defender.  As of now, I’ve got Wells as +1.0 wins as a hitter, and +1.5 as a fielder.  Andruw is +2.0 and +1.0. 

All three, Wells, Andruw and Ichiro should get around the same in terms of WAR (4.5 to 5.0), but Wells will get the longer deal because of his age.  So, if we are talking about giving these guys an 8, 6, and 4 deal next year, I’ll say that as of right now:
Wells: 8/125
Andruw: 6/120
Ichiro: 4/85

But, as you can see, the 0.5 win difference I have right now makes it pretty big between Wells and Andruw.  Such is the power of compounding.

If Wells has the season we hope from him, enough to make him a 5.0 WAR player, he’ll get 8/137.

I disagree that he’ll get more than Soriano, even though he’ll deserve it.  Soriano’s contract was really out of whack of the rest of the group.  However, that doesn’t mean someone won’t make Wells out of whack.

And if I was Toronto, I would NOT trade Wells.  This guy has a free agent price of 18 million$, and they’re probably paying him what 8-10 million$.  The “you need to get something for him” doesn’t make sense, since they are getting 8-10 million$ of value out of using him right now (using free agent prices).  In normal dollars, they are getting what they are paying for.  So, use him for this year, and bid him goodbye.

Why were Soriano and Lee so overpaid ? (relative to all the overpaid players).

You mentioned a homerun premium in post #54, and that just might be it.  If so, then Wells and Andruw will get more than Soriano

And who might pay them?

1. White Sox will need a CF, and have some big contracts expiring.

2. Giants, after Bonds breaks the record and retires.

3. Orioles will offer big bucks for both.

4. Nationals decide to become players as they get closer to the new stadium.

5. Rangers

Tango, I wish I had noticed you’d already done some back of envelope stuff, I wouldn’t have bothered to repeat it.

Wells: 7/126, paying for 4.7 WAR (starting in 2008).

In post 73, I called the three Willie Mays (Wells, Ichiro, Andruw) as 4.5 to 5.0 WAR, with Wells being 8/125, which would also be 7/118.

I’d call that one a huge bingo.

With all the deferments and signing bonuses, I estimate Wells contract would be the equivalent of him having signed next year as 7/125.  That is, they pretty much worked it out that instead of getting 18 MM in each of 2008 through 2014, that it’s the same thing as getting 8.5 MM signing bonus in 2007, 08, 08, salaries of 0.5 in 08, 1.5 in 09, 12.5 in 2010, 23 in 2011, 21 in 2012-2104.

I’m sure if I was a little more careful in my calculations, it would have come out to 7/126.

Suppan: 4/42, paying for 3.1 WAR.

Suppan’s K/W ratio is around league average, his BABIP is aroung league average, his HR/H is league average.  A league average pitcher as a starter would be around +2.3 WAR, so why is he considered +3.1?  His runs allowed, and his win probability are excellent.  His average WPA is +1.8 wins above average per year, over the last 3 years. 

He’s being considered clutch, and I’m ok with that.  They’re not giving him the whole +1.8 as clutch talent, but they’re giving him half or a third of that.

Calling him a +3.0 WAR pitcher is perfectly fine, and his contract matches that.

Good stuff.

Doesn’t look like your (excellent) salary grid adjusts for Net Present Value, or does it?

Thanks -

Whoop.  Read the article… captures both NPV and player decline.  Gracias.

Must be interesting taking the salt spray off the bow, demonstrating why GM’s aren’t as crazy as we think.  Much needed.

Also appreciate your Shandler-ian point that our calculations are only good within a certain tolerance range, and so a ‘thumb on the scale’ for clutch or whatever, isn’t out of line.

Great thread.

Either they are not as crazy as we think, or those that dabble in the free agent market are all equally crazy.

Remember, I’m applying 4MM per win, when the MRP (marginal revenue product) would only sustain a 2MM per win.

In the end, GMs work on the premise that they have a budget to spend, and since they can underallocate for slave and arb players, they overallocate on free agents, so that the players, as a group, get their fair share.

If you ever watched Back To School, it’s the difference between the way the professor was teaching business, and the way Rodney Dangerfield did.

In the end, GMs work on the premise that they have a budget to spend, and since they can underallocate for slave and arb players, they overallocate on free agents, so that the players, as a group, get their fair share.

***

Right, this is an important issue. If you pay free agents what they’re *worth*, you might end up losing the slave/arbitration system. It’s not a perfectly competitive market, so the question is not, “are teams acting efficiently on the free agent market?”, but “are teams spending what they should OVERALL?” And the answer to that question is yes.

And the other question is how to game the system, if you are the Royals, A’s, Twins, so that you don’t dabble in the free agent market (overpay), but can put that money into player development and acquisitions.

The existence of the Yanks, Redsox, Mets, Cubs, et al makes it possible for the A’s, Royals, et al to exploit the system (if they are able to of course).  Gil Meche was not one of the Royals’ finest hour.

***

Since 1992, the teams that spent the lowest on payroll, relative to the league average, was Expos (52%), Pirates (62%), DRays (63%), Brewers (68%), Twins (68%), Marlins (70%), Royals (77%), A’s (82%).

The cumulative win% since 1992 has the A’s as 9th out of 30 (.524).  Of the other teams, the Expos, Twins, Marlins can be considered moderately successful.  That’s 4 out of 8.

On the flip side, the top 8 spending teams are: Yanks, Braves, Redsox, Dodgers, Mets, Orioles, Rangers, DBacks.  The bottom 5 of those 8 are in the A’s-Twins-Expos-Marlins class of moderate success (win% between .480-.520, or 1 World Series).  Yanks, Braves, Redsox merge deep pockets with large brains.

So, it’s very possible to spend 75% of league payroll and compete against teams who spend at 125% of league payroll, if you’ve got brains (and/or luck).  In essence, that difference (50% of league average, or a total of about 40 million$), is what a great management team brings to the table.

Tango, I am trying to figure what Vidro would be worth if the M’s had signed him as a FA to DH instead of trading for him.  Where do you recommend I find averages for DH to go about this task or can you suggest what Vidro’s WAR would be?

Vidro is pretty much a league-average hitter, and a DH as a fielder.  Call him -1.5 wins at best for his fielding/position. 

(Any player who can’t field is -2 wins.  A below average fielding 1B would be -1.5 wins.  An average fielding 1B would be -1 win.)

That makes him a replacement level player, or a touch above. 

He and WFB will make a wonderful tag team.

And this from someone who loves Jose Vidro.  Check out the sponsor of this page.

++ If you ever watched Back To School, it’s the difference between the way the professor was teaching business, and the way Rodney Dangerfield did. ++

LOL.  *Precisely* the movie I think of, every time we bloggers start in again on the GM’s for the latest ‘insane’ signing.

What a website.  Keep up da good work.

Speaking of below average fielding 1st basemen, Angels signed Hillenbrand.  He’s about an average hitter, and I think that’s generous, so he’s a replacement level player.

Another righty hitting 1B/DH who’s no worse, Josh Phelps, was available to all, and the Angels might not have room for Robb Quinlan, another equal player, after this move.

So I’m pretty confident to say they just wasted 6.5 million on a replacement level player.

He’s an average hitter, and a below-average fielder for a 1B.  His position is -1 win, and his fielding might be an additional -1 win beyond that, but maybe it’s only -0.5 wins.  So, he’s at best a shade above replacement level.

Oh, and he’s a poor clutch hitter.

Speaking of the DH penalty, because the DH position affords you the luxury of not fielding any balls, you could theoretically put a floor on the poor fielding play of a 1B/DH at -0.5 wins below average.  So, perhaps the floor for a fielder, “true-talent real-life manager-moves” wise is -1.5 wins. 

In any case, Shea is like Jose Vidro really, and a few hundred other similar players.  I’d be surprised if either gets more than 200 career PA from 2009, onwards.

Eh, I guess I should say that I’d be disappointed, not surprised.

Zito: 7/126, paying for +5.0 WAR, if decline is 0.5, and +4.3 WAR if decline is 0.3 wins per year.

I talked about Zito here:
http://www.insidethebook.com/ee/index.php/site/comments/sabermetric_moves_of_the_off_season/#55

Where I concluded:

You can reasonably argue Zito as anything between a 6/81 to 6/113, depending if you think he is currently +4.0 or +4.5 WAR, and if you think his decline is 0.3 or 0.5 wins per year.

If I extended that to a 7yr deal, I would have said: 7/88 to 7/132.  So Giants are betting on both his WAR to be at the higher end, and that his decline will be rather limited.

The range here simply goes to show what kind of risk you are taking here.  Calling someone a true 4 win player or 4.5 win player is not that much of a difference.  And saying his decline will be .3 wins as opposed to .5 wins per year, again, is kinda hard to say.  Couple that with a 10% inflation rate every year, and you have anything between an 88 and 132 million$ deal that can be considered justifiable (under free agent terms, anyway).

As well, I did not build in any extra attrition for pitchers, who are likely to break down at a far higher rate than nonpitchers.

Couldn’t they have kept Jason Schmidt instead?

And I can see why Boras is likely not happy with Matsuzaka.  My guess is that if Matsu were more “American-ized”, he would have realized that he needed to negotiate better, and not make his goodbyes in Japan, forcing his own hand.

He could have stayed in Japan two more years, and waited for free agency.  It will be interesting as we get to know Matsu if he doesn’t care about egos and how he’s getting Gil Meche money, or he will become Americanized, and feel rather stupid that he (effectively through the process) gave Seibu 51 million$ so that he can leave two years early.

I wrote this on USSM
http://ussmariner.com/2006/12/28/2006-free-agent-signings/#comment-161197

===================

You are correct that my chart is based on a linear relationship, and therefore a 2-win player will, for a single year, earn half as much as a 4-win player, who will earn, for a single year, half as much as an 8-win player.

The value in the commodity is in sustaining that ability. A 2-win player will say be 1.5 wins next year, then 1.0 wins, then 0.5 wins, then out of the league, so that over a 5-yr period, he’d be 5 wins total. A 4-win single-year player will be a 15 win player over 5 years, so he’d producing 3 times more value, rather than the initial year double the value.

I believe it is on that basis that the perceived “non-linearity” comes in.

I think if all players signed for exactly one year, the linear relationship would hold (like in fantasy baseball). In any case, I think my model as I’ve presented it would probably hold if we look at it long-term.

As for 10%, this has been the case if you look at most 10-yr periods. I wouldn’t be surprised if in the future it’ll be more than 10%.

I think it’s important to keep in mind that your chart is in some sense constructed from the team’s point of view. From that perspective, 2yrs/$19m is the same as 5yrs/$34M.  But the player is going to see those very differently.  The second contract essentially insures him against injury and any sudden major drop in ability.  He could probably do as well or better signing a new contract in 2 years, but “probably” is for suckers—better to have a guaranteed $34M now. 

So when Dave (and many others) say “all these long contracts are crazy,” the problem is that desirable players just aren’t going to accept a 2-year deal.  They key for the team is to make sure they properly account for decline in peformance in the out years, as your grid does.

I also think we would have to know something about contract insurance to properly evaluate these deals.  Can teams insure themselves against catastrophic injury at a reasonable price (as the players in effect are with long-term deals), or are they assuming all the risk?

Seems to me that salary/value has to be nonlinear.  It makes too much economic sense, and it also makes sense in the non-economic thinking of baseball.

Not every team’s Marginal Revenue Product is equal.  Teams in major markets bid for premium free agents more than teams in smaller markets.  How often do you hear that such-and-such a team is “out of the bidding?” In those cases, the bidding has gotten to rich for that teams’ MRP.  So the economic basis for bidding is higher for those players.

This is compounded by two other factors that separate team-specific MRP: pennant contention and marketing.  Teams in major markets are more likely to contend for a pennant, which means they are on the steeper portion of the MRP curve (in other words, premium players are worth more to them).  And, you can derive more money from specific player marketing in large markets than you can in smaller markets.

Finally, the richer owners will want the higher profile players for their owner trophycase.  That’s the “non-economic” aspect of it all.

Every study I’ve seen has concluded that the salary/value lines curves upward, plus it makes sheer economic sense, too.

By the way, I should have added that I like your analysis of how a long-term contract plays out for different levels of players.  Great insight.  Even so, I believe that the salary value line is curved.

I was in the curved salary value camp for awhile as well, but I’ve come around to the linear.

On one hand, the superstar should be worth more, as it has been shown that a Koufax type career will add more pennants than a Milt Pappas ( I think that was the comparison Prospectus used a few years a go).

But by putting all your salary eggs in one basket, you are assuming more risk.  You can diversify your risk by getting several good players for the same $.

I think those two ideas sort of cancel each other out, and the result is the linear salary structure actually works.

Rally,

I think the point is (as I read it at least) is that for a team the curve probably is linear, well, more or less linear. You could argue that 1 superstar worth 9 WAR a season will bring more merchandising revenue than, say, 3x 3 WAR players. If this is the case then there will be a slight curve, but it will be slight.

Studes’ point is that different teams can afford to pay different players different amounts. Therefore taking baseball as a whole the curve won’t be linear because the Yankees can afford to spend more on a win than the Royals, or the Pirates, or just about any other team come to think of it.

My own view is that for baseball as a whole there will be a curve of some sort, though it won’t be that great. It is interesting that we can justify most deals with Tango’s handy dandy chart by changing an assumption here or there about decline rate. For me that indicates it is close to linear.

Saying that there are some deals that just can’t be explained: Carlos Lee for example. There is only one word to describe that contract. Egregious.

The issue of the Yanks/Sox/Mets/Dodgers v Royals/Pirates/DRays/Twins may actually be the reason that we see things as non-linear from those studies you quote, as well as the multi-year scenarios.  So, you have two very important considerations in determining if something is linear or not, and I’d bet that none of those studies addressed these two issues.

If you were to focus only on single-year contracts from mid-market teams (and then two-year contracts from mid-market teams, etc), then I would bet you’d find linearity in those particular classes of contracts.

***

Even when I proposed that Pujols would get a 8/300 or 10/400 deal, or whatever it was, people were incredulous.  It’s almost as if people will accept a non-linear ever-increasing $ per win, but then, they want to flatten it out at some point.  So, for say each win, it would be like this:
1st win: 1 MM per win
2nd win: 2 MM per win
3rd win: 3 MM per win
4th win: 4 MM per win
5th win: 4 MM per win
6th win: 3 MM per win
7th win: 2 MM per win
8th and beyond: 1 MM per win

It’s like saying there is a sweet spot of players at the 4 or 5 win level, and then after that “no one can be that good, so let’s not get too crazy paying for it”.

***

Of course, just as teams may pay Picasso prices for a franchise, they may pay Picasso prices for someone’s a$$.

It’s like saying there is a sweet spot of players at the 4 or 5 win level, and then after that “no one can be that good, so let’s not get too crazy paying for it”.

Your point about 4-5 win players being the sweet spot is right on, but you didn’t really characterize my opinion on a huge deal for Pujols.  It’s not that I feel “no one can be that good.” It’s just that I don’t think MLB teams will pay that much money.  They can get away with overpaying for talent at lower levels of talent, but they can’t when the total contract comes out to half of the entire franchise value.  It’s psychological.

What’s worse?  Tim Purpura actually overpaying Carlos Lee by 50 million$, or Tim Purpura spoofed as overpaying Joel Pineiro by 50 million$?

http://www.baseballthinkfactory.org/files/newsstand/discussion/seattle_pi_moore2/

Thanks to comment #4 in the above link for making the point.

Minky: 1/1.5, paying for 0.4 WAR.

Average hitter (0 wins), above average glove (+1), but at 1B (-1).  So, he’s average, or +2.0 WAR in full-time play. Assuming he’ll plays half a season, he’s +1.0 WAR.

http://ndai.blogspot.com/2007/01/mientkiewicz-or-why-steven-goldman-is.html

Has a different take, as it look at the average hitting of a 1B, which means a 1B gets a bigger penalty.  I’ve never agreed with the “average 1B = average 2B”, and that’s where that kind of analysis leads.

Definitely a great deal for the Yanks.  I’d also use him as my backup 2B.

I kind of like the Doug deal too.

I think plugging him into the Yankee defense knocked about 5 points of Wang’s ERA in my latest projections.

Say what you will about this year’s contracts, but at least nobody got 5/$250m…

Utley: 7/85, paying for 4.5 WAR.

Is Utley crazy?  What and incredibly stupid signing by Utley, and a brilliant signing by the Phillies.

Where to begin?  Utley is an excellent hitter.  Not the top end, but the set right below that.  He’s +3 wins above average.

As for his fielding, the Fans love him, Dewan loves him.  He’s got to be +1 win above the average 2B. 

I’m also harsh on 2B compared to others, and I treat it as a neutral position (like 3B).  No bonus points for being able to play 2B.

Overall, he’s +4 wins above average, meaning +6 wins above replacement.  The Phillies paid him like he’s +4.5 WAR (or +2.5 wins above average).

***

Even the arbitration figures were ridiculously low.  He had three arbitration years left, meaning he’s entitled to around 40% of the free agent price.  I’ve got him at 6 WAR, meaning if he were a free agent, he’d deserved 24 million$.  So, 40% of that means his arbitration value was 9.6 million$.  He asked for 6.25, and the Phillies offered 4.5! 

Divide those two figures by 40%, and you get between 12 and 15 million$ as what the Phillies and Utley thought of Utley.  Divide by 4 million$ per win, and you get back between 3 and 4 WAR. 

This is so ridiculous as to be incomprehensible.

Not to mention he’s clutch too, and I didn’t even talk about his baserunning.

Chase Utley deserved a 7/135 deal, including an arbitration discount for the 1st three years.  (Otherwise, he would have warranted a 7/164 deal.) He accepted 85 million, meaning he left 50 million$ on the table.  Which Carlos Lee somehow managed to take.

While we’re are, overall, in equilibrium, these signings make no sense.

Pat Gillick is a genius.

I wrote this on USSM, but I’ll repeat it here:

===========================================
As for Beltre, he’s got a, what, 3/40 left on his contract? He’s at worst an average hitter, and probably a +1 wins above average hitter. He’s a great fielder, at worst +1 wins above average. Let’s call him +1 win above average, at worst, or +3 wins above replacement. Such a player would warrant a 3/33 deal as a free agent, today.

A 3/40 deal implies a +3.5 wins above replacement (+1.5 wins above average).

Beltre’s deal, the three years remaining, is fair for both sides.

===========================================

Erstad: 1/?, paying for one-fourth of ?.

Whatever he signs for, Erstad will be worth it.  He is, at worst, an average-fielding CF, and likely still above average.  He is a below-average hitter.  He plays a premium position (CF).  Overall, he’s average, or a bit over.

If he plays full-time, he’s worth a 1/4 deal.  Add in the injury factor, and the injury potential because of his style of play, and a 1/2 deal would be fair. 

He’ll likely sign for 1/1.5, or 1/0.750.  He’ll be a bargain, and a worthwhile signing.

He’s also a clutch player, if there is such a thing.

***

Here is an Edmonds / Erstad story:
http://findarticles.com/p/articles/mi_m1208/is_30_224/ai_63859077/pg_4

Adam Everett, the poster boy of sabermetrics:
1/2.8, paying for 2.3 WAR.

(I’m assuming Everett has 2 years before free agency.)

Fangraphs now has forecasts from four systems, and let’s take a look at Everett:
http://www.fangraphs.com/statss.aspx?playerid=1430

He’s a terrible hitter, a -2 wins relative to the average hitter.  He’s also a wonderful fielder, a +2 wins relative to the average SS.  SS also gets a +0.5 win positional adjustment.  Overall, he’s +0.5 wins above average, or +2.5 WAR.

Bingo.

Thank you Tim Purpura for valuing Adam Everett properly.

***

http://www.insidethebook.com/ee/index.php/site/comments/uzr_2000_2005_infielders_and_outfielders/

As you can see here, based on UZR 2000-2005, Everett was +19 runs, per 600 BIP.  For a SS (800 BIP), that’d be +25 runs per 162 GP.  His 2006 UZR was likely even better than his career average, likely making him even better than +25 runs (maybe +27 or +28?).  And that comes out to +2.5 wins.

Houston fans are witnessing one of the greatest fielders of all-time.  And luckily for them, because he’s such a bad hitter, they get to pay him a small salary.

Hey Tom,

Minor (but useful, IMO) request: For threads that go over a page, could you add a link before the comments to go to the next page. It’s annoying to have to scroll down 100 posts just to get to something new.

I know, it bothers me as well.  I looked at the documentation for a few minutes, and I don’t know how it’s done, yet.  I’ll see if I can take a look at it next week.

Ok, I just bumped up the max number of comments per page from 100 to 200.  That solves it.

Thanks! Now we just have to make sure this site doesn’t get popular enough to generate 200+ comment threads!

Jeff Weaver: 1/8.4, paying for 2.1 WAR.

Weaver puts up league average numbers (1.8 WAR with 180 IP).  But, for a starter, that’s better than average (2.4 WAR with 180 IP). 

Everyone knows about him, and he got what the market dictated.

Morneau: 1/4.5, paying for +5.6 WAR.

Morneau is a +3 win above average hitter, and average fielder at 1B, making him +2 wins above average player, or +4 WAR.

If he were a free agent, that’d be worth 16 million$.  In his first year of arbitration, he gets 20 cents on the dollar, or 3.2 million$.  He’s getting 4.5 million$.

He definitely got an MVP/HR bonus.

Whatever he signs for, I’m sure he’ll be overpaid.  Twins ought to wait a year before signing him.

I’m curious what you think of the Bill Hall signing.  Here are the terms:

2007 $3.50 MIL
2008 $4.80 MIL
2009 $6.80 MIL
2010 $8.40 MIL
2011 $9.25 MIL (Club Option)

2007 would have been Hall’s first year of arby eligibility.  The Brewers offered $3 mil and Hall asked for $4.1, so they basically met half way. Assuming Hall can even get close to his combined performance from 05’ and 06’ (.280/.344/.525/.869), he would have surely gotten $4.8 and $6.8 in arbitration for 08’ and 09’.  The present value of the 2010 and 2011 salaries(10% inflation) is $6.3 mil for both years.  Isn’t this one of the best signings of the off season? 

Marcels projects .276/.338/.498/.836 for his age 27 season.  He’s moving from SS to CF this year, which hurts his value a bit.  He was an average defensive SS but seems to have the tools to be an above average CFer (good arm, OK speed). 

As a Brewer fan, it’s hard for me to not get very excited.

Hall: 5/33, backloaded, paying for +11.5 WAR total.

Definitely an interesting case. 

His 5-yr contract works out to the equivalent of 5/32, if paid the same amount each year.  So, no big difference.  He is being paid for +2.8 WAR this year, and I gave him a small decline of 0.25 wins each year.

He is an above average hitter, a +1 win above average.  Dewan loves him as a fielder, while Fans think he’s an average fielding SS.  Let’s call him a +0.5 wins as a SS.  SS also gets a +0.5 positional value, making his fielding +1.0 wins.  Overall, he’s +2 wins above average, making him a +4 WAR. 

As a CF, he’ll probably be an average fielding CF, plus he gets the +0.5 positional adjustment, for +0.5 wins.  Total WAR as a CF is +3.5.

Considering his arbitration-