THE BOOK--Playing The Percentages In Baseball

David Bell’s clutchiness has been horrible this year, -1.83 wins, but his OPS in close and late situations has been better than Abreu (.862 vs. .731). This use by so many people and writers of ‘close and late’ as the definition of clutch seems a little misguided.

Verducci is a typical sports journalistic dolt, so what would you expect.

Most sports journalists are really an embarrassment to their profession.

That being said, this whole concept of adding an impact player at the trade deadline is ridiculous.  In most cases, an added player, even a very good one like Abreu, is worth less than 1 win, which depending on the team, can increase their chances of making the post-season typically by 1 to 5%, although occasionally up to 10%.

Of course, in retrospect, lots of players acquired at the deadline, or any player at all for that matter, can look like an “impact” player, whatever that means.

Let’s take the Abreu acquisition.  Let’s assume that he is replacing Bernie Williams in RF for the remainder of the season, and that is not even properly accounting for the fact that Sheffield might come back in September, and assuming that he is healthy, that means that the Yankees will have an extra very good player on the bench, unless they play Sheff in LF, I suppose.

Abreu is one of the better players in baseball.  I have him as 3.5 wins above a replacement RF (in Superlwts).  Bernie is one of the worst players in baseball.  I have him as replacement level.  So that is about as big as an upgrade that you can get - Abreu for Bernie.  And even that amounts to about 1.3 wins for the rest of the season.  I don’t consider that a “major impact.” And that is best case scenario by far, as Bernie was not going to play every day, and Sheff may be back in a month or so.

One mistake that people make is comparing the performance this year of the one player who is replacing the other.  In fact, everyone does that.  I don’t have to explain on this blog, the problem with that.  I would be preaching to the choir.

Even Keith Law has described the addition of Lidle, who is a pretty good pitcher, as at least a “one win upgrade (IIRC).” This is ridiculous. I have Lidle as around 1/4 run per 9 innings better than an average starter.  I don’t think that too many people would argue with that.  If anything, they would probably say that was generous.

Whomever he would be replacing in the rotation would be no worse than 3/4 of a run worse than the average pitcher.  Contrary to popular belief, virtually no pitcher is worse than that (again, in talent, not in retrospective performance), Jose Lima excluded of course.

That makes Lidle, at best, one run per 9 innings better than the pitcher he is replacing.  He is expected to make 11 or 12 starts for the Yankees, I think, at around 6.5 IP per start, and that is generous I think, as he is not a durable (per start) pitcher.  That is 75 IP, which is an 8.3 run upgrade, which is around 3/4 of a win in the AL.  And that is very generous. It is probably closer to 1/2 a win.

So basically, the best trade of the season, in fact, a pretty blockbuster one, adds at best 2 wins to the Yankees.

All the other trades that were made, including the Carlos Lee trade (barely an upgrade on Mench, who is a very good player overall), are not worth spit.

As I showed here:
http://www.insidethebook.com/ee/index.php/site/article/62_to_go/

with 62 games to go, in a tight division, every .016 wins added per game is worth .08 in added chances to make the playoffs.  (That is, going from .500 to .516, means you chances would go from, say, .200 to .280).

Since .016 x 62 = 1.0 wins over the remainder of the season, that makes our calculations easier.  If Abreu is a 1.3 win upgrade, that makes it about an extra +.10 chances to make the playoffs.  (Again, this is true only with respect to my illustration of a tight race for one spot.)

Whether that is a “major” or “minor” upgrade, is really irrelevant.  He’s probably adding +.1 of playoff chances.

MGL, You are correct about Keith Law.  He said in his blog this: “Lidle’s best pitch is a splitter, but his fastball is a tick below average so he has to have good command to be effective and keep the ball out of the seats. Since the guys he’s replacing have been so bad, he’s still a one-to-two-win upgrade for the balance of the season, making this one of the biggest impact deals any club will make this month.”

I mean this guy was the Asst. GM to the Blue Jays and he’s projecting Lidle (who is nothing great, especially in the AL) to be a one-to-two win upgrade the rest of the season.  Using Keith’s logic, Lidle would be worth at least 4-5 wins per year, which is outrageous!  This tells me that the Blue Jays and Riccardi couldn’t have been the sabermetric and so-called “smart” team that many people originally thought.  It’s almost rediculous as the guys in the front office for the Padres saying that Mike Cameron is worth 1/2 a run per game on defense - and their organization is supposed to be very sabermetric.

Whether or not Lee is an upgrade over Mench, the Rangers absolutely pick-pocketed the Brewers in that deal.  If Lee leaves via free-agency, the Rangers collect two draft picks.  In fact, if they do sign Lee, I can already see the contract being worse than what the White Sox gave Lee.  Anyway, they acquired Nelson Cruz who seems to be just as good as Mench offensively (I’m not sure how he compares defensively, baserunning, etc.) and costs less and is not even arb-eligible.  Cordero has been a great reliever, and probably will continue to be on in the NL, but relievers are replaceable.  If Nix could hit, he would be a great overall player because his defense is great. /end rant.

I agree about the Toronto organization.  I have always thought that Riccardi was a poser.  I think Law is a good and bona fide sabermetrician though.  I think he just did not think too much about his Lidle comments or do the proper analysis.

SD has some sabermetrically oriented people in the front office.  However, I don’t think they are sabermetric through and through.  No teams are.  Boston and OAK come the closest by far, but even they are lacking.

The Tex/Mil trade is an interesting one.  “Trades” mostly have to be considered in the context of the contracts that come with them, which is the one thing that is mostly and usually overlooked.  If I “trade” for Jeter, A-Rod, Konerko, and Bonds, and give up some young decent players and prospects, have I made a great trade, or have I simply acquired a bunch of very good players with bloated contracts?  If you would not sign a FA for a certain salary/contract, you certainly shouldn’t give up something of value for the opportunity to acquire that player and that contract.  There are very few FA’s who are actually worth their contracts.  Of course it depends on the team and the value of a marginal win for that team at any point in time.  Lee is definitely not one of them (FA’s who are worth their contracts), even though he is underpaid compared to his current reputation.  But I’ll get to him later.

Cordero is NOT a good, or even very good, reliever.  He is a premier/elite reliever, along the lines of Wagner, Rivera, Ryan, and K-Rod.  He is one of the best relievers in baseball. He never was a big name, and fell out of favor with Texas after pitching badly for like one month, typical of a bad organization, which TEX is.  The guy has posted these NERC’s (SSRATES), or normalized component ERA’s, (the average pitcher is defined as 4.00, the average reliever is 3.90 and the average closer is 3.30) over the last 4 years:

2.54
2.27
2.06
2.83

The guy is an absolute stud.  I don’t normally advocate spending a lot of money for a reliever, even a great one like Cordero, but he is worth a lot as a closer and is vastly underpaid, especially by closer standards.  Tex is NOT a sabermetrically oriented team, never has been as far as I can recall, and is and has been generally terrible at evaluating player talent, sabermetrically speaking.  Cordero should not have been traded.  Period.  It is a joke.  Let’s look at the other personnel in this trade.

Nix is in fact a very good defender in CF, which is worth a lot in and of itself.  His hitting is a little below average for a CF’er, which means that he is a little above average overall for a CF’er.  That is worth 5-6 mil in the FA market!  The guy is making 345,000 this year.  He is an absolute steal!  He is one of those good, young players that you want to hold onto until his arb or FA years, assuming that he is going to command big money in those markets.  These kind of players are gold and the only reason to trade them
is to get better or more gold in return (other good, young players or prospects or the very occasional underpaid FA).

Carlos Lee is the classic (way) overrated player.  A big, slow, 30 year-old, below average on defense, corner outfielder, who is having a great current season.  His Superlwts projection is less than 1.5 wins above replacement!  He is making 8.5 mil this year, which is about 5 mil more than he is worth!  Who the hell would want his contract?  I don’t!  Lots of stupid teams would.  He is similar to Konerko.  ALL first baseman and corner outfielders hit well.  If you are a corner outfielder and hit REALLY well, like Lee and Konerko, unless you are good on the bases and very good on defense (which neither one of these guys are), you are simply a win or so above average - maybe.

And after this season, in order to get those draft picks (and it is not clear how much they are worth anyway), they have to offer Lee arb of course.  If he takes the arb, he is going to be vastly, vastly overpaid by Tex, making the whole deal even worse than it is!  They’ll probably end up signing him to some bloated long-term contract anyway.  They are a BAD organization!

Mench meanwhile is a BETTER player overall, and a much underrated one.  He does not hit as well, but he is better defensively, has a better arm, runs the bases better, and is 2 years younger!  Add everything up and he is actually worth more (in Superlwts projection for 06 and beyond) than Lee for 1/3 the price!

You are 100% wrong for all the above reasons about this trade.  It was absolutely horrible for the Rangers.  They traded away a better player for a worse one who is making almost 5 mil more per season (and both are FA in 07).  In addition, they gave away a way underrated good, young, cheap player in Nix, and one of the best relievers in baseball who is vastly underpaid.

This is one of the most misunderstood trades in baseball history by almost everyone and I am here to set the record straight!

Yeah, when JP said that measuring defense is most accurate coming from a scout’s notebook, it really threw a question mark up.  I think that he’s given in to the Toronto media and overpaid for players like Glaus, etc.  Why Law left, I don’t know.

MGL, Great post.  I wish I would have spent more time in mine b/c I exaggerated it too much.  Obviously Cordero is going to be a stud in the NL - his “month” problem was basically due to losing control of his fastball (I’ve watched the Rangers quite a bit).  He still has that dominating stuff...and he’s due $5 million in ‘07, which seems to be exactly his value.  Actually, he may be underrated by a $1 million or so…

Other than having a pre-arb/pre-FA player (which as say is “gold"), the Brewers are loaded with OF’s right now, so I’m not sure where Nix fits in.  The problem the Brewers will have with Nix (if they Brewers are a sabermetrically-oriented team which I don’t think they are), is that teams (and even I did admit doing this a little) see his horrible performance offensively at the major league level over the last three years and don’t even bother looking at his defense.  If he can put up numbers like he’s currently doing at AAA, he’ll be one of the better CF’s.

While I now agree that Mench may be better overall (given his other peripherals), Nelson Cruz isn’t that bad of a player either.  Well, I can’t 100% say that because I don’t know his UZR’s or baserunning skills, and those may not even be accurate given his experience.

I agree that the Rangers will give Lee a bloated contract ($4-5 years, $60 million) instead of offer arbitration and letting him walk.  I’d take my two draft picks and call it a day because you know a team is willing to pay for Lee.  I know that the Rangers haven’t been a traditionally sabermetric team, but they just hired a 28 year old kid who seems smart at the beginning of the off-season.  While I liked the Wilkerson trade, he’s really looked horrible this year.  But, it’s one year, so I can’t really put that much into Wilkerson’s off-year (unless scouts somehow notice that he’s a completely different hitter).

Oh, since we’re throwing around stupid organizations, here’s my #1 “pathetic organization:” Pirates.  Oh, and the trade of Wilson for Chacon has to be one of the worst in a long time.  Why Wilson was not traded before the year started or why the Pirates didn’t just collect a draft pick or two for him really amazes me.  I guess they thought he might accept arbitration…

Tango wrote: “he is probably adding .1 of playoff chances”

Interestingly, after the trade the Yankee’s WE for the AL East rose about 8 basis points. Though that also included a loss by the Red Sox. That particular contract may have one or two liquidity issues though

I noticed that Keith Law just posted a few times at BTF, so maybe he’ll come over here and explain his Lidle comments.  I’d love to here what he thinks.

Thanks for your analysis of the Lee trade, MGL; just going off of Clay Davenport’s figures it looked to me as though Texas might be gaining around five runs offensively from the Mench/Lee swap, so it’s interesting to hear that Mench takes the lead once defense and baserunning are taken into account.

MGL, the one part of the trade you didn’t address was Nelson Cruz, and how his value compares to Kevin Mench.  Your points about Cordero and Nix are well taken, and it could be that the community at large has failed to properly value these players and thus the trade a whole.  But Cruz is also pre-arb “gold” and presumably close-ish in value to Kevin Mench.

I forgot about Cruz.  I don’t know anything about him, but his MLE’s over the last 3 years are excellent, especially for a CF’er.  If he is of average defense in CF, or even a little worse, he is in fact gold.  If he is a corner OF’er, then he is not worth all that much given his MLE’s and at his age.  But you are right.  He must be included in the equation.

My main point was the assumption that Lee was much better than Mench, which is patently false, and that people seem to forget or ignore how great Cordero is and has been.  Thinking Lee is so good is a perfect microcosm for what is wrong with player evaluation in baseball.

Could Law have meant 1-2 wins over the below replacement level druck the Yankees had been sending out there every fifth day?  While Lidle is no great shakes he may be a win better than Wilson/Ponson/Chacon/Small, who were dreadful.

Wow, I should have read everything before I posted, sorry about that.

It’s good people realize that so-called baseball writers (eg Verducci)are a joke (or something like that), or smooth, agenda-driven phonies, but who’s
going to complain about this other than me?

jschmeagol, your point is well-taken.  Many writers compare one player (such as an acquisition like Lidle or Abreu) to the actual performance of the player or players who are being replaced to “determine” the value of the upgrade.

Of course that is the wrong way to do it.  You have to compare the expected performance of all players going forward (which, for the player or players being replaced, is generally going to be a lot higher than their recent past performance) to estimate the value of the upgrade.  You would expect Law to know that and I think he does.  I think he just did an off-the-cuff estimate.  Maybe not though.  Maybe he thinks that those numbers are right with regard to Lidle.

One thing I like about BPro is that they often (not always unfortunately), in evaluating players, teams, trades, playoff chances, etc., use current Pecota projections rather than the ubiquitous and STUPID, “What they have done so far this year.”

interesting

and here, my stros got yelled at by most of the fans for not trading

roy oswalt, mo ensberg and adam everett (and maybe willy taveras

for either carlos lee, abreu or tejada

- what do yall think of THOSE trades?

lisa

If you have a house worth 5 million$, and you’ve only got a 2 million$ mortgage on it, and you have another house worth 15 million$, with a 20 million$ mortgage on it, would anyone, other than a baseball owner, trade those even-up?

Unless he has some inside info that the 15 million$ house is sitting on oil, keep the 5 million$ house, and go buy a second house.

That is a great analogy!  Baseball owners do that all the time.  In case anyone does not know what Tango means, he means that they trade worse players who are undervalued in salary (the worse house with lots of equity) for the better, overvalued player with a bloated contract (the better house with “reverse” equity).

As I always say, and I have gotten Neyer to say, you don’t trade players like you trade bicycles or baseball cards.  You trades players and their salaries (like houses and mortgages) and the talent of the players being traded is completely irrelevant except as they relate to their contracts.

When you (Lisa) ask about Ensberg, Oswalt, and Everett, the first thing I have to do is look at their contracts.  Lee’s salary is not all that bad, but he is only 1.5 wins above replacement, so you are paying like 5 mil per marginal win, Abreu’s salary is too high at around 4 mil per marginal win, and Tejada is not too bad at around 3 mil per marginal win.

Oswalt is around 4 wins above a replacement starter at 11 mil, so he is almost 3 mil per marginal win.  He is worth more to a playoff team though as he is a #1 or #2 playoff pitcher.

Ensberg and Everett are steals.  They are both getting paid around 1 to 1.5 mil per marginal win, assuming that Ensberg is healthy.  Of course all of Everett’s value is in his defense.  While he is known as a great defensive player, I don’t think too many GM’s understand how much his defense is worth.

So, no, you don’t trade Ensberg or Everett unless you are getting some darn good good, young players or prospects in return.  You certainly don’t trade them for overvalued FA’s with bloated contracts.

Of course if your goal is to upgrade the team, no matter how much you have to pay, that’s a different story, although it would be hard to upgrade the team by getting rid of Everett and Ensberg, two of your best players.

Taveras on the other hand, sucks. He is an average defensive CF’er who can’t hit a lick.  There are a ton of those.  Not quite replacement, but close.  You trade him for something of value, if a team is willing, or you wait until his arb year and then you get rid of him, assuming that he is going to command a mil or 2 in arb and does not improve his hitting.

Everything hinges on a team being able to do two things:  One, forecast a player’s overall (offensive, defensive, and baserunning) value, to a reasonable degree of accuracy.  Two, understanding how that forecast relates to marginal win value and hence salary.  (Three is understanding the fungible nature of offense, defense, pitching, and everything else that goes into player value.) There are only a handful of teams that are even remotely capable of doing both (or all three) of these things.  Houston is not one of them, I don’t think.

tango and MGL,

thank you very much for your comments. I am going to post them in my blog.

incredible that peter angelos turned down roy oswalt + adam everett + mo ensberg for tejada. i personally am grateful that he is such a dumbass.

unfortunately, there are too many astros fans who do not understand how much adam everett’s defense is worth, neither. ALL they look at is BA.

lisa

A prefect example of the Tango/Lichtman construct (trading houses with mortgages/contracts, etc., rather than players, per se) would be Everett straight up for Tejada. Most people would think that would be a joke for Baltimore (and a steal for Houston), when in fact it would be a steal for Batlimore (and a joke for BAL).

An even worse deal would be trading Everett straight up for Jeter.  That would be one of the worst deals (for HOU) of all time.

I think you can reasonably argue that wins have an increasing marginal value on a player-specific basis.  Both Nate Silver and Vince Gennaro have shown why that would be so in their work—wins have an increasing value up to around 100 wins or so in Vince’s work—and the “market” for players bears this out because top free agents are paid on an incrementally higher basis.

There is an important difference between where teams are on their win curve.  I think Tejada is worth more to the Astros than to the Orioles because Houston is in contention and Baltimore isn’t.  In that case, trading Tejada for Everett might make sense for both teams.

MGL, if Jeter was traded straight up for Everett, you know the Yankees would be recquird to pay some of Jeter’s salary.  How much, I don’t know, but if it’s a lot then the Astros might want to do the deal.

I don’t know how much they’re making (mortgage), but let’s say that the Everett property valuation is +6 wins above replacement over the next three years (15 million$), and the mortgage (salary) over that time is going to be 7 million$.

Let’s say Tejada’s property value is +12 wins above replacement (30 million$), and the mortgage is 36 million$.

Normally, you *don’t* trade it even up.  But, let’s say once you surround the Tejada property with Chez Clemens, Oswalt Enterprises, and Biggio Bungalows.  Everett Condos don’t help there too much, but Tejada Train Station helps alot.  Now, this property valuation of Tejada goes from 30 million$ to 44 million$.  In this case, it would be a fair trade, even-up, for the Astros.

Even so, they should still consider the possibility of getting an even better value, say someone who is worth 28 million$, with a mortgage of 29 million$.

On the other hand, if Tejada’s property valuation doesn’t improve as much as one would need, it’s a silly trade to make, even up.

Adding Oswalt and Ensberg too?  Yeesh.

Tango, that all makes sense.  I like the “Biggio Bungalows” name..haha

The only thing that would top this all off would be to have the financials for every team so we could properly determine how much a marginal win is worth specifically to “X” team.  I just feel uncomfortable using the $2 mil per win for some teams.  Also, for all players that have signed free-agent contracts at any point in their career, shouldn’t we be using the $3-4 million per win when determine their “worth/value” in dollars per year?  Sometimes I see people using the $2 mil per win when a player is in the second or third year of his FA contract.  Using this logic, the “worth” of all players that have been FA will automatically be higher in dollars.

Mike

You can get some financial data here: http://www.businessofbaseball.com/data.htm

It may be a year or so out of date. Forbes publish an annual run-down of the finances, but I am not sure how accurate it is. The data from the business of baseball website is from Forbes I think.

Mike,

In terms of what an FA gives you, most are overpriced.  Sticking with the property value scenario, being a free agent is like moving your house from Newark to Short Hills.  Location, location, location.  The intrinsic value of the house hasn’t changed, but all of a sudden you are on valuable land.  But, a baseball team only needs a good house.  It doesn’t need to buy the land too.