THE BOOK--Playing The Percentages In Baseball

David Ortiz: 1/12.5MM$, option exercised.

This one pretty much sets the ceiling for DH.  A .380 wOBA last year, probably forecasts close to .355 for 2011. 

It’s pretty much what Vlad or Matsui entered 2010 as.

Compare Matsui’s 2007-2009 to Ortiz’s 2008-2010:
.378 / .380: last year
.348 / .340: T-2
.368 / .372: T-3

And Matsui is one year older, meaning they entered their season the same age.

Really, this is about the best comp you could ever hope for.  And, it’s guys in back-to-back years.

Matsui and Vlad also comp well.  They both entered 2010 with around a .355-.360 Marcel forecast, and, guess what, that’s pretty much exactly what they put up in 2010.  And they signed for 6MM$ each.

This one should have been cut-and-dried.

So, either the Redsox grossly overpaid, or they believe the market is about to ignite.

I’ve been fine with the idea of $12.5M for Ortiz for one season, as it doesn’t tie them to him for more than the one year if he doesn’t hit as expected, and the Sox certainly have the revenue to spend a few million extra on a DH if they have to on a single-year contract.

It won’t keep them from re-signing any of their free agents or those of any other team if they want them, and Ortiz should be useful for 2011 without locking them into 2012 and beyond as he wanted (and which would have been pushed for if they chose negotiations instead of the option).

Marc: I don’t understand that line of reasoning.  If Ortiz, Vlad, and Matsui all forecast the same (and a quick Marcel has them all in a dead heat), why would the Redsox overspend on an Accord if they can buy a Camry or Altima, just because they happen to have double-the-money to spend, and there’s no Lexus out there?

Given that the Rangers declined to spend an extra 8MM$ on Vlad with no bidders, I can’t see how the market would shake out the way the Redsox exercising their option would dictate.

No, either the Redsox have a better forecasting engine than Marcel insofar as mid-30s players with no fielding value are concerned, or they spent 5MM$-6MM$ over what the open market would have dictated for “comfort” reasons.

I’d like to see MGL’s and Oliver’s and PECOTA’s forecasts on these three players for 2011.

They may have a better forecasting engine, in that they know something about Ortiz that we don’t.  Maybe they think he’s a contract year guy.  Maybe they think he’s had significant personal issues the past couple Aprils, and they’re discounting his performance in those months relative to the rest of his performance.  Heck, maybe he has some measurable clubhouse value.

Or they may very well be paying him “irrationally” in the sense of economics and value, and just paying him out of emotion and thanks for his past accomplishments.

The Rangers don’t have the pockets of the Red Sox, and will be spending a large portion of their available free agent money--including the signing bonus from the new TV deal--on attempting to lock up Cliff Lee. That Vlad option was never meant to be picked up in Texas, and shouldn’t be viewed as a sign of the market. I’ll be more interested in whatever number they do assign to Vlad as they attempt to re-sign him (or if they are willing to give him more than one year).

Ortiz is the youngest of these three and had the strongest 2010 even after park-adjusting, so I’m curious about why their 2011 Marcels are that close.

I don’t have the 2011 PECOTA on-hand yet, but PECOTA did like him for 2011 heading into 2010 (though, like you, I would rather look at the 2011 forecast than that approximation).

I think there are also many human issues involved with Ortiz and the Red Sox surrounding this contract issue. Clearly, $12.5M is far too much given the publicly available forecasting systems. However, I’m not sure Ortiz would have signed back up with Boston on the market if his option was declined. Despite his recent decline with age, Ortiz is still a prominent face of the franchise and a fan favorite.

Sabermetrically, this is a terrible move. Perhaps Boston thought the public backlash of releasing Ortiz or the roundabout way of signing him on the FA market would be worse than overpaying.

@3: Maybe that’s the premium the Sox are willing to pay for certainty: There is always the off-chance that they don’t get one of the DHs they target in FA; however unlikely. Perhaps that’s part of the “comfort” aspect?

Also, I know we usually refrain from this perspective, but Ortiz does come with some branding benefits. I highly doubt it’s $5-6M worth of branding benefits, but perhaps enough to mitigate the cost a little.

Marc,
Keep in mind that the Red Sox’s financial advantage is contingent upon them spending money as wisely and as frugally as everyone else. For example, if I go into a grocery store with twice as much money as the average shopper but decide to pay twice sticker price for everything, then I no longer have anny financial advantage. When big market teams think the rules don’t apply equally to them, well, meet the Mets.

@#8

Oh of course--but for one player, and a few million, it won’t be a big deal. Certainly not on a Mets’ scale

Following up on #6, I think the Sox would take a pretty big PR hit if they didn’t exercise the option. I could imagine a nominal decrease in revenue if they didn’t sign him, and among future FAs, a slightly greater reluctance to sign with the Sox based on the sense that they don’t treat their players well. Nothing big, but maybe on the order of a million here or there. So, with a bit of inflation factored in, they’re only wasting a few million by paying $12, instead of wasting ~$6. The Yankees have a similar issue w/ Jeter - they’re in a position of having to pay a bit more to maintain status as a “classy” organization and they can’t really cry poverty as a defense.

How are people forecasting Berkman? And what type of deal do they think he will get?

I feel like for the Cubs Berkman is a better fit than Dunn.

Oliver thinks the Sox overpaid. It doesn’t really like any of the three DHs. Here are their 2011 projections:

Ortiz - .341
Guerrero - .346
Matsui - .332

This is what Oliver has for 2011

       wOBA  BA  OB  SA
Ortiz   341 239 338 442
Matsui  333 253 338 417
Vlad    346 285 334 466

Thanks guys.

Brian, you seem to have an aggressive aging on Matsui.  According to Fangraphs, his career RC+ is 126, his 2010 RC+ was 125 and his 2009 was 131.  A .333 wOBA would essentially be an RC+ of 100.

So, setting age aside, and applying some regression, we’d have an estimated RC+ of something between 120 and 125, closer to 125, for 2011.  But you are giving him 100, which means an aging factor of 20%.

The only thing I can see is an elevated K rate, which definitely does look bad for his future.  But, not to the extent you are showing.

Is it possible to show us what your forecast for Matsui is if you change his birth year to 1979 and 1984?

I pulled those numbers from the file I sent to THT last week, but I’ve been tinkering with the code since the end of the season, and although I thought I sent Dave the final version, when I reran numbers just now to check on Matsui for you, I got higher numbers for all three.

       wOBA  BA  OB  SA
Ortiz   355 248 355 473
Matsui  347 268 349 443
Vlad    359 296 344 491

Here are Matsui’s basic, aged and park specific projections

          BA    OB    SA  wOBA   _BH   _HR   _BB   _SO
Basic  0.275 0.361 0.460 0.359 0.292 0.052 0.109 0.151
Aged   0.266 0.348 0.435 0.345 0.286 0.046 0.101 0.153
Parks  0.268 0.349 0.443 0.347 0.286 0.048 0.100 0.151

Brian, excellent.  Those numbers are much closer to what I have.  I was right around .355 or so for all three, so we are in basic agreement.

Ted Lilly re-upped earlier for $33/3 years. Guess that implies 7-8 WAR over the life of the deal?

According to fWAR, Lilly produced 8.9 WAR over the past 3 years, so 7-8 WAR over the next 3 seems a tad optimistic for a 35 year old.

As far as comps-- Randy Wolf (3 years/$29.75M + option buyout) and Ryan Dempster (4 years/$52M) are the most recent decent comps who were recently FAs. Lilly’s a been a bit better than Wolf and a bit worse than Dempster, and a year older than both of them.

Seems like his new contract is pretty much in line with past markets.

We have our first, uh, controversial move of the off-season (I think...)

Marlins trade Dan Uggla for Omar Infante and sign John Buck to 3/18.  Buck projects to be...what, probably 1.5 WAR or so?

Not sure what is controversial about it?  Uggla will make at least 7M more than Infante this season.  Throughout 150 games Uggla is roughly 20 runs better on offense, Infante probably around 7 runs better on defense.  Difference of 1.3 WAR and 1.3 WAR is less than 7M.  Give Uggla a bump up because he always plays more than Infante and looks pretty even.

And if Buck projects as 1.5 WAR like you said than 3/18 makes perfect sense…

Cards sign Westbrook, 2/16.5 with a mutual option that could bring the deal to 3/25 or a 1 million dollar buyout.

@Sean: Don’t forget that the Marlins are passing on Type A arbitration that was essentially guaranteed. In return, they’re getting Infante, who should receive Type B but might be more willing to accept arbitration.

"Cards sign Westbrook, 2/16.5 with a mutual option that could bring the deal to 3/25 or a 1 million dollar buyout.”

I think I’ve seen this movie before.

Year 1: 1.3 WAR. Disappointing.
Year 2: 3.1 WAR. He’s finally regained his form.

Option picked up.

Year 3: Out for Season - Injury.

... or ...

Years 1 and 2: Very good.

Sign with another team for more money, production goes way down minus Duncan.

I like the signing, and I liked last year’s deal for Penny. Paying a guy as if he’s going to be “league average” when he could be slightly below to slightly above, that’s a decent deal.

#3, Just looking at my end-of-season projections for Matsui, Vlad, and Papi, I have Matsui and Vlad around the same, 5-8 runs per 500 PA, and Ortiz a little higher, at 12-16.  So, .4 wins better or around 2 mil.  That is a rate and does not include chance of injury or expected playing time.

I may have undervalued Vlad because I don’t usually do any tweaking for injury plagued seasons and the consensus, I think, is that Vlad was injured last year and that is why he played so badly.

P.S. I was just looking at the BJ 2011 projection for him.  He’s got a heck of an optimistic one for Vlad - 135 RC+ after posting 142, 131, 110, 124 over the last 4 years.  Even if we “adjust” 2009 for injury, a 135 projection with that past performance seems ridiculous for a 36 year old player with a history of injuries.

I was pretty surprised to see the rockies extend Tulowitzki through 2020: 7 yrs/$134 million starting in 2014.

I have had a man crush on Tulo since his rookie season, and I probably overrate every aspect of his game.  But I just don’t understand this from the Rockies’ point of view.  He’s under contract for three more years already, at a massive discount. 

I realize that I realize that by 2020, $20 M per season (or whatever it is at that point) may be a massive bargain.  We have know way of knowing what $/WAR will be when the contract ends.  And it is, in fact, a good deal now, given that he’s something like a 5-6 win player (or better) as long as he can be mostly healthy.  But he is a player with a history of injury, and was already under contract for three more seasons so there was no rush to get this done (or, at least, there shouldn’t have been).  He’s now signed for an entire decade into the future.  What if he gets hurt next year?  Or any time in the next 5 years?

Convince me I’m wrong!

And what if he’s better than we think?

6/119 is average.  If you don’t think it’s average, then TELL ME what is average?  What should have been a fair deal?

Show your work…

Uh, I just realized that I posted this in this thread, and not the Tulo thread.  Justin: did you see the Tulo thread?  Please post there.

Nope, sorry.  Was doing this too quick in a break and didn’t check.  My bad…

I use a $3.5M/win for 1 year contracts, $4.5M/win for 2-3 year deals, and $6M/win for longer than that (based on a study I saw from Tango). Add in a 3% inflation rate and a 10% discount for +3 year deals, my model likes these two deals.

Juan Uribe signs a 3 year $21MM deal with the Dodgers. I have him at a 2.2 WAR player with 70% playing time, falling to 1.9 next year at age 32 and less half a win per year after that. Assuming teams wouldnt pay him after he falls below a 1 WAR player, I get a 3 year deal for $21M - wow that worked out well.

Victor Martinez signs a $50M / 4 year deal with the Tigers ($12.5/year). At 32 and assuming a .5 win loss in each year to 34 and then .7 after that, I have a 3.5 WAR player falling to 3.2 next season and 1.7 by the 4th year of the contract. Using $6M/win for a 4 year contract with a 10% LT discount, I get a $57M deal at $14.2/year. Looks like not a bad deal here for the Tigers. However I have him with 80% playing time as a catcher over that period. Should he get shifted to 1B or DH perhaps this falls into the players favour.

Many teams are literally using sabermetric models to figure their contracts.  It is no great coincidence that we are seeing evidence of that in the contracts themselves.  And of course even in not all teams are familiar with how to properly evaluate a player, if enough teams are, then that essentially sets thee market more or less…

Bronson Arroyo 3 years 35 million.  As a Cardinals fan, this pleases me greatly.

Cardinals sign Lance Berkman to a 1 year 8 million dollar deal, no draft picks given up.  Berkman’s likely a 3 WAR player next year in the outfield.  Fantastic deal, with the way that dollars are being doled out this offseason.

"Berkman’s likely a 3 WAR player next year in the outfield.”

That seems awfully high.  I would guess he is a -10 defender in the corner OF.  In order to be a 3 WAR player, he would have to be 25-30 runs or so on offense and base running.  I don’t have him anywhere close to that.  I have him like +10 runs in offense, which would make him a 1 to 1.5 win player.

I also would think he’s a -10 run defender in the corner outfield, but I would have him at around +30 runs per 600 at bats.  That translates to around a .385 wOBA, which is about .020 points below his career average, and right about what he’s done the past three seasons.  I suppose aging brings that down a bit, so maybe 2.5 WAR.

+10 runs on offense seems very pessimistic to me.  Last year, with a deflated BABIP, he was roughly a 10 run hitter, and that was by far his worst season ever.

Contracts
-------------
Matt Holliday: 120M/7y
Jayson Werth: 120M/7y

WAR - 4 years prior to contract
----------------------------------
Matt Holliday: 22.4
Jayson Werth: 18.2

Age at time of Contract Signing
--------------------------------
Matt Holliday: 29
Jayson Werth: 31

Err, Werth 126M/7y

Werth: 7/126, paying for 4.32 wins at a 0.5 win drop a year, with 5MM$ a win in 2011, and increasing by 8% after that.  This is ONE schedule:

I have seven nonpitchers forecasted at over 4.32 wins, and Werth is not one of them.

Werth is 32, and that means he should on an accelerated aging curve, not the standard -0.5 from above.

As much as I love Werth (as a ballplayer), this is a huge overpay.  The above schedule is not justifiable.  A 4/56 deal would have been much more reasonable.

Either that, or Werth is going to be ageless like Ichiro or Rivera.

Werth is 32, and that means he should on an accelerated aging curve, not the standard -0.5 from above.

Numerically, could you describe what this curve should be like, or point to a link? Are we talking -0.75 now, some other type of aging curve, etc.?

Check the archives from November.  Look for WAR aging curves for great players.

Geez, that thread was only a week ago and I’d already forgotten reading it. Thanks.

Tango, when you talk about -.5 or -.75 wins per year decline, I assume you are talking about rate PLUS playing time (both decline)?

MGL: yes, definitely.

The PA drop is roughly about 5% per year (less when you are younger, more when you are older).

So, let’s try some examples.  You are a true 4 WAR per 162 G player.  You lose 5% of your PA, which means you lose 0.2 wins just on playing time.  When you are older, it could be even 10% or even more in year-over-year.  So, you could lose say 0.5 wins just on playing time.

For performance, you lose say about 0.3 wins per 162 G, a bit less when you are younger, and a bit more when you are older.

Something like that.

I can try to come up with two separate functions, one to age by playing time and one by performance, and do that on an age basis, and maybe also by performance level.

It’s unofficial, but Jack Cust with the Mariners for $2.5 Million. Looks like another low-cost, medium upside player for the M’s, and he instantly becomes the best hitter on their roster.

Bill James projects a rather optimistic .359 wOBA and an increase in ISO, which I don’t understand. Cust’s power has been steadily declining each of the past 3 years. That BABIP’s coming down, so I’d expect more like a .340 wOBA. What does Marcel/Oliver think?

ZiPS projects .230/.362/.411.

Carl Crawford got 7 years, $142 mil from the Red Sox. Busy week for Boston. Sick lineup.

I’m more interested in where Ellsbury gets dealt to. They could hold on to him for after Cameron is gone but they moved him off CF indicating they may not value him as above-average defender there. So who knows how highly they rate him.

#43

BJ from FG - 242/377/428
Oliver     - 241/364/430

(His signing hasn’t shown up in MLB’s transactions yet, so Oliver above is still a FA’s neutral park projection.)

Still only makes him around a 1 WAR player, as a just about average DH who will cost runs when he does play the field.

CHONE’s last update had Cust at .241/.372/.432 for the season (in 562 PA). That’s about +20 for hitting. I believe we have been roughly using -22 for DHs/very poor defensive corner OFs. From there, I give 20 runs for AL replacement level given 80% playing time (25 *.8). So, that comes out to 1.8 WAR or so, which makes this a fine deal for the M’s.

Crawford: 7/142MM$, paying for this schedule:

$perW going up at 8.5%, starting at 5MM$ per win.  WAR starts at 4.8, dropping at 0.5.

I have Crawford as a 4 WAR player.  I think this is an overpay.  How would a more typical team see him?  Well, they’ll be conservative on the growth of salaries, say setting it at 7%.  The $ per win would start at 4.5MM$.  If we do that, this is what they’d see:

That’s 102MM$ for 7 seasons.  Or, for the first five seasons, it would be 84MM$.  That’s more in-line with Torii Hunter and Ichiro from a few years ago.

***

As you can see, it highly depends on the rise in baseball salaries.  The Redsox are betting big.  Not only that, but they have him valued at 4.8 WAR, which is excessive.  Maybe they think he will have a better aging curve.  Let’s try another schedule:

I started him at 4 wins in 2011, but this time dropping him by only 0.25 wins a year.  Inflation set at 8.1%.  In this case, inflation and aging cancel out (as you can see by the constant 20MM$/21MM$ payout each year).  That gives you 7/142.

Anyway you cut it, this is an optimistic or aggressive forecast.

***

Werth was a huge overpay, and this one was also a big overpay.

Obviously, the biggest assumptions having the greatest effect on final valuation is how he will age and his initial WAR. The difference btw .5/yr and .25/year is huge and similarly starting at 5WAR vs 4WAR.
If you think CC is 5WAR, aging at .25WAR, you can use a very very conservative $4.5m, 4% inflation and get $150m in value. So how is that an overpay?

Also: How many contracts are needed to reset intuitions on fair $/W market value? If we split the diff and say CC is 4.5WAR, .375 aging, pretty conservative inflation (5%) while give you 142 if present value is up to $5.25/win, so in some sense, is that not the market now, especially with the werth deal?

In order to treat your numbers as reasonable, you would have to apply them to *everyone*.  Not only to Crawford and Werth, but to Beltre, Lee, Dunn, et al.

And we should see it, to some degree, for extensions, like Verlander, Felix, Ryan Howard, and so on.  (Though they pose additional problems because of arb status.)

Basically, you can pretty much justify any contract by manipulating the numbers to make it fit. 

The question is: are you being reasonable, and are you being consistent?

I now like to use Matt’s thesis about switching teams (or not) in order to assume “good” or “bad” aging.  I think that is a reasonable rule of thumb for long term contracts.

Use .25 if the old team re-signs, and .75 if they don’t…

It’d be great to see a running total of cumulative $/win valuation during the offseason!

But my second point was more of a digression - what was important is that in these calculations, it’s almost all about the WAR and aging and has very little to do with differences in opinion on $/win or inflation.

If you take fangraphs WAR and Marcel-project it, even standard aging (.5) and medium inflation (6%), you get bargain basement $4m/win.
Fangraphs wisdom-of-crowd WAR is even higher - 3% inflation does the trick.

So, you’re really saying you think there’s an error in fangraph fielding valuation, because that’s all that’s needed to account for the discrepancy between an overpay or a pretty good deal.

I also can’t help but think that there is a supply and demand force at work with some high value FA…

51/MGL

I’m saying you have to be way more principled in these aging numbers (and initial WAR valuations). You can’t be so flip as to just guess out a 3-fold difference in aging standards (.25~.75). By comparison, that’s the equivalent of a difference in assuming 3% inflation and 22% inflation for the CC contract.
This is the danger of guesstimating. While you can do so reasonably for some numbers (e.g., inflation) your results are way too sensitive to changes in others (aging, initial WAR) to just throw out, “oh, .25, .5, .75 - whatever” To be rational, you’d need a much narrower band of variance - say, .45-.55- which might not be possible given the data.

My Q doesn’t really belong in any particular thread, but wasn’t sure where to ask it. As this is more of ‘just’ an announcement thread, I figured I’d ask it here.

1. Does the mailbag submission link at the top of this page get responses? I just submitted one. It seems like I recall you (Tango) saying that a better way to contact you would be to email you directly. Which brings me to my second question:

2. What’s your email address? Again, as I recall, you’ve given it out in the comments here before. Something like tom(at)tango(dot)net. But I don’t recall for sure.

ANYHOO, the questions I would like to ask you is: Does Carl Crawford’s skillset age well?

Dave thinks so: http://www.fangraphs.com/blogs/index.php/boston-signs-carl-crawford/

But Craig is not convinced:
http://hardballtalk.nbcsports.com/2010/12/09/so-is-the-carl-crawford-deal-actually-you-know-a-good-one/

To summarize on the differing opinions: Dave seems to intimate that a guys like Crawford are naturally athletic and avoid injuries, hence they ‘age well’. Whereas Craig says guys like Crawford rely on speed, which is one of the first things to go with age (as opposed to, say, HR power), hence they do not ‘age well’.

tom~tangotiger~net

The email link at the top right of every page goes right to the above email, so either way is fine.

Tom, what does WOWY think of JJ Hardy’s defense.  UZR rates him as a plus defender, but the Twins seem down on his defense in their interviews. Thanks.

JK/57: I find it interesting they’d be down on it. There’s an interview with a Twins blog with their assistant GM about how they go about things and he mentioned the Hardy trade specifically as him getting glowing reports on his defense.

JJ: right around average

Mike Rogers/58:  Gardy said, “We like Hardy a lot. He’s a great guy, great teammate, and we believe when he’s healthy he’s solid at shortstop”

That doesn’t seem like, at least the manager, thinks he’s the plus defender his UZR suggests.  I’m also inferring that the FO doesn’t value him as a plus defender based on the trade return.

JK/60: Here’s an interview with the Assistant GM from March of 2010 talking about Hardy in particular: http://overthebaggy.blogspot.com/2010/03/twins-and-statistical-analysis.html

“Question: Defensive stats such as UZR have been made sexy by Seattle’s front office recently. Did the Twins look those kind of statistics when targeting players, like Orlando Hudson or JJ Hardy?

Answer: When you look at Orlando Hudson, you can look at the UZRs and all that stuff and that’s all fine. That’s a piece. They say range is declining. Okay, what do our three scouting reports say? He still gets the ball – he’s never been fast, he’s never been a fast player – but he has all the quickness. He still has first-step quickness and he makes all the plays. So we will put more stock in that.

Just like JJ Hardy. He may not have the range he did when he was 24 at 27 now. He makes all the plays, he has enough arm and gets the ball. Our theory has always been: Make the routine plays – don’t beat yourself - and if you don’t get to a ball or two that is offset when you make all the plays. There are guys that may have 15-20 errors, a lot of them are routine errors. We’d rather have the guy that when there is a groundball to short, that’s an out. That type of thing. We scout that way, we look at it that way and we make our decisions that way.

There might be some out there that go jeez, they just traded for Hardy and signed Hudson. Our stuff says that if you look at it purely from statistical analysis and Zone Ratings and all that other stuff, defensively they’re not what they were. We’ve got people watching, saying that we know what this guy can do. That’s what we’re looking for.”

I looked up the defensive number for Hardy. GBH (ground ball hits to outfield) are estimated based on the vector of the batted ball, and the base/out state and batter hand for each GB. Infield hits and reached on errors are simply counted. Raw numbers, no adjustments.

DER is pct of all GB’s that get an out. Range is pct of all GB’s kept in infield. Hands is pct of infield GB’s that get an out. Rank is fro 1 to 30 (top 30 players in opps each season, then ranked by category).

Except for 2008, Hardy’s range numbers have been virtually identical in his four other qualifying seasons, ranking him 20-25. His sure handedness ranks better, in the 5-10 range the past three years.

Over the entire six year period 2005-2009, Hardy is 19th in DP’s started with 57.3%, 25 in DP pivots with 56.7% (although his pivots may likely be pulled down by Fielder being 30th in DP starts at 1b and Weeks 28th at 2b. Need WOWY to help straighten out the combined efforts.

Using WOWY methods, Hardy’s total fielding runs as reported at THT Forecasts were -8, +11, -11, -11 the past four years.

YEAR  DER Rnk  Rng Rnk Hands Rnk Opp GBH IFH ROE
2005 0.649 16 0.754 21 0.871 18  365  93  26   9
2006 0.719    0.781    0.920      96  21   5   1
2007 0.629 19 0.754 20 0.882 11  568 163  36  12
2008 0.690  2 0.807  4 0.903  5  551 130  28  13
2009 0.647 16 0.762 25 0.894 10  416 115  25   7
2010 0.634 19 0.756 25 0.900  6  413 122  23   6

Very interesting, Brian. Thanks. I find it very intriguing how the Twins found Hardy to be still a good defender (good enough to acquire and start him) last winter and now have determined his fielding is not good enough to keep around (like Nick Punto).

I bit surprised to see no mention of the Jay Bruce extension.  Waiting for the specifics or just not noteworthy?

Cardinals traded Brenden Ryan to Mariners for pitching prospect Maikel Cleto.  I still do not see why the Cards think that Theriot is an upgrade over Ryan, in the short or long term, especially with the groundball staff they have put together.  Sounds like LaRussa has gained power over Mozeliak (and Luhnow).

Lee gets 5/120 (or 6/135 if the option vests).

Looks like a great deal for the Phillies.

If we assume $5M per WAR, 5% yearly inflation, and a 0.5 yearly drop for aging, they’re paying for a 5.5 WAR pitcher.

Lee averaged 7 fWAR the last three years.

If Lee signed for 5/100, it’s this:

If he signed for 5/120, it’s this:

I kept both at 8% increase, and 5MM$ per win. 

I have him forecasted for 4.6 WAR.  The Phillies are in the playoff sweetspot, so if he signed for 120M$$, maybe 10-15MM$ is a playoff bonus.

I should probably show the WAR curve for great pitchers next…

I’m curious what the actual gain is to the Phillies in the playoffs. Obviously this makes them better in theregular season because of the amount of starts Lee should get. But if he’s only replacing the innings of Blanton in the playoffs, how much of a gain is that? Is Lee minus Blanton greater than not having Werth anymore?

Don’t forget Domonic Brown.

And factoring in the regressions and aging for their current players.

In a vaccuum, if you remove Werth from the Phillies before the playoffs started and add Cliff Lee, would their chances be higher/lower/the same?

tango, is lee really projected to drop two and a half wins? that seems quite excessive, though i don’t doubt you have good reasons.

Observations is not equal to true talent.  You don’t “drop” from observations to true talent,

Does anyone have any thoughts on Overbay’s 1/5 deal with the Pirates? There’s a pretty furious back and forth about it over at BBTF. Misuse of resources (with Pearce and Bowker already on the roster and bigger holes at SS and SP) or reasonable deal to obtain some reliability for a team that probably desperately wants to avoid losing 95+ games again?

If you treat Overbay like he has no fielding value, then of course it’s going to look bad.  If you treat him like a gold glove-type, then it’s going to look much different.

Without first agreeing on his fielding value, then the discussion is pointless.

So, first even decide: do you want to have a discussion with a point or not?

Hmmmm. I guess I’d like to know what people’s thoughts are on the Overbay signing. If that’s not enough of a point, then you don’t have to weigh in.

I have no idea what Overbay’s fielding value is, myself. He’s probably not below average at first, but beyond that I don’t know - recent discussions on this blog have made me much less confident in trusting the publicly-available fielding stats.

But the only way to have an opinion on the Overbay signing is to have an opinion on his fielding.  If he’s a Frank Thomas fielder, then you have a league average hitter who’s really a DH, and therefore, has no value.  Encarnacion or Aybar, etc.  If he is the best fielding 1B in baseball, then he’s got tremendous value.

To have an opinion for the sake of getting opinions but not understanding the REASON for the opinion is… well, useless to me.

I guess there’s some miscommunication here. I’m not asking for mindless, knee-jerk opinions. I didn’t think I had to explicitly articulate that, but I guess I should have. I understand that Overbay’s defense is important, as are a number of other factors. What I was interested in is hearing people’s well-and-thoroughly-reasoned opinions about the signing, although I must confess that being lectured about it kind of sapped my interest. Sorry to have wasted your time with my useless questions.

Well, you’ll get a lecture here from time to time.  Nature of the beast (or the benefit of being here, depending on your perspective).

There aren’t really “a number of other factors”.  He has over 4000 MLB plate appearances, he’s going to enter the 2011 season as a 34 year old player, and his career RC+ (or OPS+) is around 110.  We’ve seen this story a hundred times, haven’t we?

If he has no fielding value, that would make him just above a replacement-level DH.  If he is a gold glover, that would make him a somewhat above-league average 1B.

The major opinion to have here is how good is his fielding.  A minor opinion is if his hitting is going to follow a normal aging pattern or not.  (Obviously, I can’t offer my opinion.)

I’m providing (lecturing?) you a framework for you to think of this deal in clear and fairly certain terms. I’m giving you a productive way to think about this.  Who the f-ck cares if I sound professorial about it?  If it makes you feel better: “pretty please, with a cherry on top?” (*)

If you want to think about this deal in some other terms, as it seems you want, then there’s plenty of other places where you can spin your wheels and not be better off.

(*) Harvey Keitel, Pulp Fiction

If that’s what you want, I’d suggest giving us your own well-and thoroughly-reasoned opinion on it.  And if others take interest, they can respond to that.  I don’t do well-and thoroughly-reasoned on request.

Oh, and often I have said any single person’s opinion (mine included) is almost worthless.

Based on the framework I’ve provided, the whole thing hinges on his fielding.  You’ve got the Fans Scouting Report and the 30 or 50 or whatever fans’ opinions, and you have UZR, and DRS, and TZ, and nFRAA.  The tools are there for you to combine them as you see fit, in a consistent manner.

My best guess:

600 PAs = 20 RAR
.340 wOBA = +7.5 RAA
1B position = -12.5 RAA
field 1B = +5 RAA

Total = 0 RAA, 2 WAR, ~$10M are current FA rates

Oliver has Overbay’s DER at 1b ranking 13, 7, 6, 3, 3, 2 that past five years, +3 runs in each of the last three.

But, Overbay has had park adjusted wOBA’s the past three years of .343, .365, .329 where .355 is average at 1b.

Projection (in Pgh) for 2011 is 252/343/410 +3 with glove for 0.7 WAR.

Garrett Jones is projected 259/320/445, -4 fielding for -0.2 WAR.

Overbay looks to me as Jones with a glove but 10 fewer HR’s.

Pirates bring in Overbay and Diaz where they already had Jones/Bowker/Pearce/Clement for two positins (rf/1b), and Bowker may be the best of the six (and that’s not saying much).

e poc, you have unwittingly violated the posting policy here.  Your posts must follow this format:

Pirates sign Overbay, 1 yr/$5 million.  Projects as about 15 RAR as a hitter, no better than average in the field in his age 34 season.  Not really a significant overpay, buy why?

Something like that will show that you’ve looked at Marcel or Zips or some other projection sheet, that you’ve thought about the player’s age, and that you are obviously smarter ("but why?") than the Pirates’ GM.  You do not want to belabor the point; that would tend to undermine the implication that it is “obvious” that your little back-of-the-envelope arithmetic is far superior to whatever faith-based formula the Pirates GM is using.  This will garner you praise, and you will be accepted into the community.

Seriously, when did Tango become the insult comic?  Didn’t that used to be MGL?  He seems downright cheery these days.  And I’m old enough to remember when we thought of Canadians as really nice people!  Nothing like an author hawking a book (and I bought it; quite insightful once you get past the rather ponderous writing) insulting his audience.  How dare you ask “Whaddya think about the Overbay signing?” without prefacing it with the approved preamble!  Tango just doesn’t have time for you!

I’m not “hawking” a book.  That’s bullsh!t.

The rest of your post: funny stuff!

Brian, I’m not following how you get .7 WAR from the info listed.  Would you mind breaking out hitting, fielding, position, and replacement runs for me?

Sky: league average hitter (0), +3 for fielding, -12.5 (or whatever Brian uses for 1B), +20 (or whatever) for replacement.  That’s about +10 runs or +1 win.  Playing time brings him down a bit.

Tango - I understand the concept of true talent level, I guess the better question is how the heck do you determine’s Lee true talent level to be around 4.5 WAR, 2.5 WAR lower than his average of the last 3 years (at least according to Fangraphs)?

I have him as around a .625 pitcher with 21 complete games.  Hmmm… that comes out to 5.1 WAR.  If I treat him as a .600 pitcher, then it’s 4.6.  Not sure why I said he was 4.6… must have made a mistake somewhere… I might have quoted you his numbers from last year maybe?

Jenks -> R.S., $12m/2yr.
Last year’s high ERA was probably anomalous because of a .368 BABIP; based on last 3 seasons, he looks projected to be around 1 WAR, so seems like a minor overpay, but it’s only 2 years in the face of all the sucky 3 year reliever contracts going out the door.

Discuss.*

* - Except for Tango because I don’t want him to get hulk-angry at me for exhorting him to discuss.

#64/RMR, I’m interested in talking about the Bruce extension, so here goes.

Hitting: After being average his first two years, Bruce’s amazing surge over the last two months put him at +18 RAA last year (both b-ref and FG agree).  ZiPS has him average again for 2011, while Oliver has him with a .354 wOBA (basically same as last year).  Splitting the difference and going the full season = +12 RAA

Position: RF, so -7.5 RAA

Fielding: He’s clearly very good.  Fans rated him as 2nd-best RF behind Ichiro last year, +9 RAA (+11 full season).  His career average UZR/DRS in RF is +15 RAA/season.  Colin had him at 54 +- 12 RAA last year, which even if we just go with the low end of that range is an absurd +42 RAA.  I dunno, is a +20 RAA too optimistic?  +15 maybe?

Replacement: NL so +20 RAA (though NL has been getting better...)

Playing time: Bruce’s has an injury history the past two seasons.  ZiPS projects 73% PT, Oliver 76%.  I’m going to go with 75%

So that give us: +12 - 7.5 + 15 + 20 = +39.5 RAA * 75% = ~3 WAR player?

He signed for 6 yrs/$51M, with an option for another year at $13 M.  He was a super-2 this offseason.  Tango-esque schedule (sorry about formatting, can’t get html to work right):

Using .3/yr aging mostly due to injury risk--he’s still very young and probably will get better. 

Looks pretty much dead on to me.  And I’m biased, but I think I was being pretty conservative in the projection.  I think he’ll beat this level of performance.

Thoughts?  I’m not sure if this is right, but I think it is pretty close...?
-j

Justin: excellent.  You broke it down by components, so now, if there’s a disagreement, the disagreement will be on the particular components.

Without doing the breakdown as you did, and just coming up with a summary conclusion, we would just be cherry picking data left and right to support us and challenge others.  This way a 50-post thread can be collapsed into a 5-post thread, because we’ve all agreed to the debate format.

***

mettle/90: “Tango laughed.  Tango not mad.”

To get the formatting in Justin/91, use [ and ] not < and > .

With the word
code
in the middle of the two square brackets.

I created a Yoda/Hulk thread…

Giants offer Renteria $1M:

http://www.sfgate.com/cgi-bin/blogs/giants/detail?entry_id=79295&tsp=1

Renteria seems to project as a league-average hitter and fielder.  We’re looking at 2 WAR from him in 2011.

The Giants are trying to get a real deal on this one.  “Total disrespect” may be a little harsh, but it’s not clear why they’d low-ball.

#91/jinaz used the 2010 ZiPS instead of 2011 ZiPS.  Turns out Oliver and ZiPS agree on Bruce, and put him around +15 RAA/yr as a hitter.  ::sigh::

Anyway, that bumps Bruce’s projection up by a half-win, which slants the deal in the Reds favor.  And that makes more sense, as it’s supposed to favor the team when you make an early contract extension like this because the player trades potential earnings for lifetime financial security.  Here’s a schedule based on a 3.5 WAR Jay Bruce.

-j

Hawerchuck,

With the Tejada signing and a couple other moves the Giants have made Renteria would be a bench player.  He just wouldn’t get that much playing time unless someone got injured. 

But I do agree with Renteria, if you are going to offer him 1mm tell him the only offer you can make is for a bench role.  Give him realistic expectations so he has an opportunity to say don’t bother.

Justin/97

Oliver has Bruce a park-adjusted 3.3 in 2010 (half of that due to +16 defense), projected 2.1 for 2011.

But...the dude’s only going to be 25, so don’t be knocking so much off a year, he’s still likely pre-preak. Oliver says 2.1, 2.1, 2.0, 1.8, 1.6, 1.3

How you’ve laid out the schedule, even with the flatter aging curve, they would have to believe Bruce’s true current value is much closer to his 2010 performance (will +16 fielding carry forward?) than his three year weighted mean projection.

Brian,

Can you give components behind his 2010 3.3 WAR?  Or the 2.1 WAR projection for 2011?

For example, for 2010 FanGraphs has:
Bat: +18
Field: +20
Repl: +19
Pos: -6
Total: +51

BRef has:
Bat: +17
Field: +15
Repl: +16
Pos: -7
Total: +42

So 3.3 WAR seems really low.  Your fielding is right there with B-Ref, so it seems like your batting line must be way below these other ones.  It’s that, or you’re doing something really different with position adjustments or replacement level.

Using an average of the Oliver and ZiPS projections, plus a best guess on fielding (including consideration of Colin’s +52 rating by nFRAA last year), I’m getting Bruce for 2011:
Bat: +14
Field: +15
Pos: -7.5
Repl: +20
Playing time: 81%
14+15-7.5+20 = 41.5 * 81% = 33 RAR

I can see dropping this by 5 runs if you think the fielding projection is too strong, but all the way to 2.1 WAR?  It would take a much more aggressive park factor that I’m using.  I’m only using a runs-based park factor (1.03 for CIN, applied on an additive basis), not components, so maybe that’s where it is?

As for the aging--what you’re doing makes sense, I’ve just seen Tango do a -0.3 aging curve some time ago and went with that.  I’m happy to not assume aging for a few years as you have done.  I haven’t really investigated aging very well to this point personally.
-j

@Steve

The Giants are doing some weird things - behaving as though players should want to play for them because they’re going to win the WS again. 

They should not have offered Renteria a contract.

Jay Bruce 2010

     wOBA   BA   OB   SA
Raw  .365 .282 .353 .494
Adj  .357 .277 .351 .475

Gets knocked down some for playing in a hitter’s park

.357 wOBA,
100% in RF
.322 replacement
572 PA

(.357-.322)*572/1.15 = 17.5 BRAR
16.3 FRAA
-0.9 SBR

32.9 RAR/10 = 3.3 WAR

Brian/102,

Ok, it looks like our difference is baseline.  You’re using .322 wOBA replacement.  I don’t usually calculate it this way, but I think I’m using .308 wOBA replacement level:

(.308-.330)/1.15*700=-13.4 RAA

(-20 replacement + 7.5 position = -13.5 runs)

Whereas yours is:

(.322-.330)/1.15*700=-5 runs replacement

I’m not as well read on this stuff as I was a few years ago, but yours seems like a pretty darn high replacement level baseline to me.  Everyone in your system should be about a win lower than the fangraphs system, for example.

I’m using what fangraphs uses, afaik, -13.5 runs for RF.  B-Ref is closer to -10 in RF, I think.
-j

Justin - I am using .322 as the replacement level for a RF, where average is ~ .347, not for everyone. It’s approx 90% of average, varies slightly by position

If you do replacement level VORP-style, where you set offensive replacement level as a function of position, the Fangraphs or Rally model would put rep level hitting at about 80% of league average.  (FG slightly lower, Rally slightly higher, I think.) Oliver’s got it at 90%, which is the highest I’ve seen.

Whether it’s right or wrong, keep in mind that a 2.5 WAR player in that model is better than a 2.5 WAR player in a FG or Rally WAR model (and way better than a 2.5 WARP player, at least prior to the Colin revolution.)

80% is runs created per PA.

wOBA is not that.  It would be closer to 90% for its equivalence.

Justin, I’m surprised to see the yearly WAR decline applied to a 24 year old - even if halved.  If the expected WAR progression always has a negative slope, than how do we end up with an observed performance peak around 27-29?  What does the standard WAR curve/curves look like.  Is the expected decline due to playing time attrition?

MGL’s Component Aging Curves google spreadsheet (assuming I understand it properly), suggests a 24 year old at a given level of performance would produce something like 2 additional wins of offense through age 30 above a flat baseline.  Is this increase simply offset and surprised by defensive decline and decreased playing time?  Or am I totally misunderstanding aging curves?

Rick,

As I said above to Brian, I’m perfectly happy to be wrong about the aging.  It is not something I’ve researched much, if at all.  I was just going by something I’ve seen Tango do in the past (at least as memory serves--maybe with the Cano extension) for similarly young players.  It makes sense to me to at least see flat aging, given the likelihood of improvement and the chance of injury.
-j

I think we can see someone young go from being a 2.5 WAR/162G to a 2.7 WAR/162G.

I think we can also see someone go from playing 155 games to 140 games year over year.

So, 2.5/162*155=2.4
2.7/162*140=2.3

The playing time is a killer. 

I just made sh!t up here, but it’s critical to separate the rate portion from the playing time portion.

Steve

“The Giants are doing some weird things - behaving as though players should want to play for them because they’re going to win the WS again. 

They should not have offered Renteria a contract.”

I think the SFG are playing it very smart. When I read about the Renteria deal, I was stunned. Knowing Sabaen’s perceived reputation, I thought Renteria would get a “WS Champ contract” where he’d be overpaid and for too many years.

Rather than throwing out long-term to contracts assumming they’ll repeat or start a dynasty, they’re signing guys to short-term contracts that will allow them to see where they’ll be in 2011 and 2012, and then decide who to give long-term (if any) deals to.

I plan to give an attempt at calculating whether StL bringing Renteria back to play SS, bumping The Riot to 2B (replacing Skip), would be a good move, worth a 2-year contract, and for how much money.

When you consider the bump just from getting Skip out of the everyday lineup at 2B, TR provides an improvement in the field and at the plate. Likewise, ER may be better at both at SS than RT is. I’m interested in whether it would be a win-win for StL, and seeing that he is available, and was already pretty good in StL (likely popular move to bring him back).

I just wanted to cut and paste something from a ‘Nats blog, where the guy who blasted my suggestion that Laroche was a useless expenditure of two years and 8 million per was roundly lauded for a post including claims that his power and fielding were hardly “meh” (my phrase) and that, regardless, the value he would provide would be meaningful for the franchise. But I’ll just include my response to the first part:

“Parsing data doesn’t help Laroche’s case. Many times on this board I’ve changed my mind, but interestingly, the one time it seems absolutely no one has an issue with a player — not really — I just can’t get swayed.

Suggestion for when you evaluate players in the future: don’t use one year of data for UZR or, for that matter, use just UZR. Total Zone / Dewan’s +/- have him weaker than most, but even still, the WIDELY accepted requirment to include three year averages when using UZR has him inthe bottom third of all qualifying first basemen, and even if your 2.0 figure weren’t clearly wrong, that is BAD for such an easily fielded position.

POWER: use ISO, not HR… of which his .215 3-yr. avg is in the bottom half of all qual. 1Bs (with the rest ranging much higher, from .223 – .273 … NOT including Pujols, to remain fair).

BATTING EYE: guess who is THE WORST in BB/K %? Laroche. ’nuff said.

OTHER: bottom half SPD score, bottom half avg. (the only guys with worse contact % are at least known for power to make up for it: Howard, Pena, Dunn, Fielder and one rookie, Davis), worse than average hitting FBs, worse than EVERYONE hitting sliders, worse than three-quarters hitting curves (and only average with the rest), and if you believe in “clutch” (I don’t) he is one of the worst there too.

So, again, looking at ACTUAL numbers, why do I want over 10% of the portion of my ticket price allocated to player salaries to go to Adam Laroche?”

For those economic-minded minds here, is there something I’m missing?

Right....so there is perfect knowledge of his future offensive ability?  No room for “opinions” of his hitting performance this season?  Instead of telling people what opinions they are permitted to have, sit back and listen.  You may learn something.

“But the only way to have an opinion on the Overbay signing is to have an opinion on his fielding.  If he’s a Frank Thomas fielder, then you have a league average hitter who’s really a DH, and therefore, has no value.  Encarnacion or Aybar, etc.  If he is the best fielding 1B in baseball, then he’s got tremendous value.

To have an opinion for the sake of getting opinions but not understanding the REASON for the opinion is… well, useless to me. “

It’s been a while since we’ve had a red herring around here.  So, you figure that to make your argument, you have to presume I implied “perfect future knowledge”.  This entire blog is predicated on the uncertainty of sample data.

If you want to talk about Overbay’s offense, then describe his true talent level, and describe your uncertainty level of this estimate of yours.  Bring something to the table rather than sticking your head in the room and saying the party is lame.  And when you do so, let’s hope that your estimate of his offense is different from the rest of us.

Otherwise, my point stands as valid and correct.

Jays trade Vernon Wells to the Angels for Napoli and Rivera, Jays not eating any of the contract. I don’t think we need to bother analyzing this one…

Quick calcs.  At $5M per win, Napoli earning $6M and $9M in arb (conservatively high) and worth 2.5/2 WAR, Rivera worth .5 WAR, and Wells worth 4/3.5/3/2.5 WAR (conservatively high), this is a $25M gift from the Angels to the Jays.

If you change Wells’ WAR projections to 2/1.5/1/.5 (August CHONE has Wells at 1.7 WAR/150), it’s a $65M gift.

According to Cots, the Jays have $44M committed for 2010, pending the arb outcomes for Bautista, Napoli, Frasor, and Litsch.  And then a bunch of near-league-minimum guys to officially sign.

Wow, desperation must have been high in the Angels camp after not signing any of the big free agents. I would have never thought the Jays could get rid of the horrable Wells contract without eating a lot of money of it…

FWIW, the guy who inputs data for Stats Inc. in Toronto says that Overbay does very well at scoopage, which some (most?) of the defensive metrics do not measure.

Napoli for Francisco plus $?  WAR for relievers is tricky, as accounting for leverage is pretty much a necessity and pretty much impossible to predict in advance.

I’ll be unstickying.

Any nominations for worst signing?  Jeter the winner?