THE BOOK--Playing The Percentages In Baseball

Heyman tweeted that the Mariners turned down a bigger package from the Yankees in an effort to trade for Felix.

It could be argued that the value of those extra 2ish wins Felix is projected to provide over Pineda next year and 2013 are (in marginal terms) worth more to the Yanks than most other teams.  Also, money is worth less to the Yanks than other teams, since they basically print it.

MGL I agree and I think it’s really that many people don’t think in terms of equity or surplus value.  Tango loves using mortgages here as analogies, but I think many people use the wrong reasoning when they buy their actual houses.

To many people, it seems to me that people have a budget in mind, and they search for the best house under that budget.  So if they can afford $300k for a house, they look for the best house not exceeding $300k.  Now suppose House A costs $295k and gives them $305k of utility for a $10k surplus.  House B costs $275k and gives them $300k of utility for a $25k surplus.  They should buy House B and use the $25k saved for something else, but instead they buy House A because it has the higher raw value despite less surplus.

Now I freely admit this is anecdotal and I lack hard evidence for this, so this sentiment carries little weight.  But whenever I’ve spoken with people about buying a house, they rarely speak about which house-price pair they like.  The price seems like a separate, binary consideration.  If they can “afford” it then the price doesn’t seem to matter.

So if that’s how people think about houses, it’s consistent with the Felix/Pineda reasoning.  The Yankees can “afford” the salary of either player, and Felix is better.  Pineda has more surplus value, but that doesn’t seem to enter the equation at all.  Similarly, I imagine Yankee fans would have been unhappy at a straight-up AROD for Longoria trade shortly after AROD signed his new deal.  Longoria’s contract had a ridiculous amount of surplus value compared to probably 0 or less surplus in AROD’s contract, but surplus value just doesn’t seem to enter into the equation for most people.

I think a better analogy is cars.  You have a Civic that has a street value of 20K, an Accord that has a street value of 30K, and a Lexus with a street value of 40K.

The dealer is offering you a Civic for 10K, an Accord for 25K, and a Lexus for 40K.

Which do you buy if you have 45K in your pocket?

You *could* buy both the Civic and Accord, try to flip it, and then buy the Lexus.  But if you can’t flip them, and you REALLY need the Lexus, then you’ll pay the 40K.

You might be tempted to get the Civic, and then take the other 30K and buy season tickets to all your sports teams, but the wife wouldn’t like either of those two choices.

#2 yeah, yeah, we’ve discussed that a million times before and the argument just doesn’t hold water and it certainly doesnt hold water in this case where the difference in equity between Felix and Pineda is enormous.

If the Yankees would truly pay tens of millions (in the form of players like Montero) for a pitcher like Hernandez AND pay Hernandez 20 mil a year, don’t you think they would be scarfing up free agents like Lee, Wilson, et al for way above market value?

They don’t because they are not stupid (Bill Gates doesn’t walk into the Lexus dealer and offer 200 thousand for a 70 thousand dollar Lexus because he CAN) and because they don’t have to. 

The Yankees would not have offered Montero for Hernandez regardless of how much a marginal win is worth to them. So can we please put to rest that argument?

MGL, I understand what you’re saying and it’s something that had crossed my mind as well. However, teams like the Yankees clearly do not view it the same way.

The Yankees put a GIGANTIC premium on winning in the playoffs, and they see a proven, durable Ace like Felix as their Holy Grail, the type of player who gives them the best chance of winning the World Series. And that’s what really drives them, winning it all. So when they look at Felix, they really don’t care what he’s being paid they simply see a guy you should empty your farm system to get.

Remember, just a year and a half ago the Yankees agreed to trade Jesus Montero plus several other players for 3 months of Cliff Lee. That means that they value a half season of Cliff Lee greater than 5 YEARS (and super cheap years) of Michael Pineda. Lee had some surplus value then, but that wasn’t what made him appealing. It was the idea that he was a sure fire winner in the playoffs, and for that he was worth his weight in gold.

My concern about Felix is that in 3 years (IIRC) he’s a FA. SEA is not going to be contending in the next 3 years (unless something drastic happens).

Had they traded Felix, they could have gotten a handful of prospects to combine with the few prospects they have that are MLB ready (or already played a season in MLB) and have (perhaps) 4-6 3-5 WAR players in 4 years ... and a bunch of money to add FAs to that core.

The concern for me in regards to Pineda is that he’s the next Liriano in that he’ll throw the hell out of the slider, his arm will fall off, they’ll sew it back on, and then he’ll either repeat the pattern or he’ll not throw his best pitch as often and not have a quality backup pitch.

"SEA is not going to be contending in the next 3 years (unless something drastic happens). “

Please, Nostradamus.  What does that mean?  That they have a 0% chance of contending over the next 3 years?  10%?  What exactly does that mean?

I’ve shown that teams that finish in the bottom 3rd one year will finish in the top 3rd at some point over the next 3 years one-third of the time.

That somehow equates to “not”?

In all fairness to Nostradamus, Seattle has a pretty bad team and is now in a tough division. I would guess that they have less than a 5% chance of making the post-season this year and given that they are not going to go out and spend a ton on FA in the next 2 years (like, say, WAS or MIA) and that they don’t have a bunch of prospects in waiting, I would say less than a 10% chance in 2013 and 15% in 2014.

Saying that teams that finish in the bottom 3rd, etc., probably does not apply to the Mariners.  That “stat” includes teams that got really unlucky in that one year by fluctuation or injury and teams that greatly improved their team (like the 2008 Rays). I don’t think the M’s fall into either of those categories.

#6, yes, there are lots of teams (not just the Yankees) that will be willing to pay more than the average team for a FA because they are in the sweet spot of wins or because they have lots of revenue, but....

The Yankees are simply not going to pay what someone like Montero is worth (say, 40 mil over the next 3 years) PLUS another 58 mil for Felix for a 14 WAR player (Felix over the next 3 years). Not even close.  Not even close to close.

That would be the same as paying a FA pitcher of Felix’ caliber, or some combination of players that nets them another 14 wins, 100 million dollars for 14 WAR! That is not going to happen. Why? Because, like Bill Gates and the Lexus, they don’t have to. They can get all the WAR they want for 5 mil a win. They don’t have to pay 7 mil.

Regardless, the notion that the Yankees did not get Felix rather than Pineda because they would have to have given up a lot more players (with equity) is beyond absurd…

Let’s say we go with that, as an illustration only: 95% of not playoffs (90% of not contending?), 90% of not playoffs in year 2 (80% of not contending), 85% of not playoffs in year 3 (70% of not contending?).

So, over the three years: 73% chance of not making the playoffs (.95x.90x.85), and 50% chance of not contending for playoffs.

For discussion’s sake, here’s the “source”:

@JonHeymanCBS
Jon Heyman
#yankees tried for #felix before acquiring pineda. #mariners wouldnt budge. #yanks woulda offered montero/banuelos/betances/etc./etc

It is hard to wrap your head around that one, especially given the Yanks’ luxury tax status that would actually treat Felix like a $30m/yr player.  Not to mention their well-documented incentive to get their payroll under $189m by 2014.

https://twitter.com/#!/JonHeymanCBS/statuses/157986574699724800

Why would anyone believe a story about what didn’t happen?  I believe that SEA is smart enough that they would have jumped at that deal and laughed in Cashman’s face.

So the Mariners get Montero and all those other players PLUS they have 58 million over the next 3 years to spend on other players or player development, and you think they would turn that down?  No chance…

Put another way, rather than having 58 mil to spend to replace Felix PLUS a few extra players, they elected to GIVE the Yankees a 40 or 50 million dollar commodity (Pineda)?  Right.

Rather than give someone my 2 million dollar home with no equity, I decide to give them instead my other free and clear million dollar home, because I like the 2 million dollar home so much…

Empirically, couldn’t we just figure out how much the Yankees pay per WAR on the FA market each year? If it’s consistently more than everyone else, it would suggest money doesn’t matter. It will, of course, be just a bit more, and not much more, because, as per MGL’s Gates lexus point, the Yankees only have to outrun the other teams, not the bear.

IIRC, they were seriously overpaying for FAs (as much as one could) for a few solid years in the mid 2000s, but in the past 3 years have been paying market value on pretty much every deal.
That either points to bad talent eval (unlikely) or:

a lower utility for $/higher utility for wins than any other team.

****

Economic theory totally supports an uneven valuation for every player in terms of the $/win component of a calculation. Why is that so hard to fathom? The Yankees will place a completely different $ value on Felix and Pineda than the Mariners will. It’s just for convenience that we use a fixed $/WAR across all teams because that’s the current *market* price. But for any transaction between two entities, it won’t necessarily be that number - the team valuation is different.
I can go through an example with widgets if that will help.

MGL, again, the Yankees agreed to trade Montero+ for a partial season of Cliff Lee and the Rangers gave up Justin Smoak and three other prospects for him. By your logic those teams would be absolutely crazy to make such deals but it happened! It may not be what you would do but it’s something teams will absolutely do it if they are focused on winning the Series.

Thing is, King Felix is a known commodity and a more complete pitcher than Pineda at this moment.

Pineda had a 4.9 ERA in his last 16 starts, had arm problems in 2009 and appeared to lose a bit of velocity in the 2nd half of 2011.  He is also at an age (23) when pitchers tend to be more of an injury risk while King Felix at age 26 is in his mature phase and has shown he can handle a 200+IP workload .

Some teams (investors) are willing to pay a premium for certainty. Risk is heavily discounted. 

Now it may well be that Pineda stays healthy the next few years and matures as a pitcher and becomes as dominant as King Felix is over a full season (and maybe King Felix breaks down unexpectedly).  If so, he will provide far more value than King Felix.  The devil is in the uncertainty.

If Felix is a 4 WAR player and Pineda a 3 WAR player isn’t also wise to take in to consideration the error bars on these valuations?  In other words the distribution of total likely outcomes.  Kind of along the lines of what the previous post said.  Or is that risk already factored in to the WAR predictions? 
vr, Xei

#16, when did Cliff Lee pitch for the Yankees or is that another report of a deal that didn’t happen?

In any case, getting a player in mid-season is a completely different story. Plus, if the Rangers are giving up Smoak and 3 other prospects plus Lee for Montero, doesn’t that support my point exactly?  That Lee is worth little or nothing in the deal?

#18, WAR is based on mean or median values. Value related to uncertainty, standard error, error bars, or whatever you want to call it, is something that each team has to decide and determine for itself based on its unique circumstance.

In my opinion, certainty is overrated.  How much certainty do you think a pitcher like Felix has as compared to a pitcher with a similar mean WAR but less certainty?  A lot less than most people think. There is obviously no such thing as a “sure thing,” including Felix, for three reasons.  One, the random fluctuation surrounding a player’s true talent is inevitable. Two, we don’t really know Felix’s true talent just like we don’t know Pineda’s, although we can estimate it for Felix with more certainty than for Pineda.  And three, there is a greater than de minimum chance that any pitcher gets injured and does not pitch or pitches badly (because his true talent declines from the injury).

This whole concept of, “You know what you are getting,” or the flip side, “You don’t know what you are getting,” or, the related concept, “Upisde and downside,” are all overrated concepts in terms of team decision making. By far and away the most important thing is a reliable estimate of a player’s true talent…

I can’t help but disagree with a pure surplus calculation as a model for reality. By this logic, any two free agents at market prices are equivalent, but you’d never see a straight trade of a 3 WAR FA for a 1 WAR FA.

Likewise, it pretty much suggests that a team should flip any (fairly paid) FA for prospects with ANY chance of being above replacement in the majors (and thus having a surplus value).

I think I prefer to think of it as teams having fixed budgets and maximising WAR (or playoff expectations), both now and to a lesser extent in the future.

To clarify, I think the surplus method would be perfect if we had an infinite supply of free agents at all WAR levels; as well as having no roster constraints.

As an example, do you trade Pujols on his current contract (assuming it’s fair) for a mop-up middle reliever (say, 0.5 WAR in year 1, then ageing) under team control?

"In any case, getting a player in mid-season is a completely different story. Plus, if the Rangers are giving up Smoak and 3 other prospects plus Lee for Montero, doesn’t that support my point exactly?  That Lee is worth little or nothing in the deal?”

MGL, what are you talking about?

In 2010, The Rangers gave up Smoak + 3 other prospects for a half season of Cliff Lee. This is an fact.

They did this deal after the Mariners turned down an offer for Lee which was Montero + 2 other prospects.  This was widely reported and I don’t remember it being disputed (Yankees never offered Montero, etc.) at the time.

How does this imply that Lee was worth little or nothing?

This was pre-FA for Cliff Lee. So even though it was only for 3 months, he still had significant surplus value. Fangraphs calculated $15-$21 million of equity (including value of compensation draft picks). This is not at all similar to trading for a market value Felix Hernandez.

"This is not at all similar to trading for a market value Felix Hernandez.”

Do you think Felix at 58/3 is market value?  What do you think teams would pay for him on the open market, especially if they didn’t have to guarantee 5-6 years for him?  At the end of the deal, he should be worth draft compensation too, since under the new CBA all a team has to do to get that is offer a 1 year, 12 million dollar deal.

I’ll just use the average of MGL’s “4-5 WAR” from the post.

4.5 2012
4.0 2013
3.5 2014
---

So 12 WAR over the life of his contract would be something like $48 million to $72 million depending on your average win value ($4-$6). Average $60 million.

$58 million is pretty close, but I suppose it depends on your starting point win value and inflation estimate.

Please, Nostradamus.  What does that mean?  That they have a 0% chance of contending over the next 3 years?  10%?  What exactly does that mean?

I’ve shown that teams that finish in the bottom 3rd one year will finish in the top 3rd at some point over the next 3 years one-third of the time.

That somehow equates to “not”?

The Mariners are that shitty, yes. 128 wins over the last two years and they’ve gotten WORSE by trading Pineda and Fister (and Morrow) for essentially Montero, Brandon League and farmhands.

So, a bottom third team has a 33% chance on making the playoffs once within the next 3 years.

It’s not just that the M’s were bottom third, but they traded away valuable pieces on those bottom third teams.

If the percent of competing over the next 3 years is 33%, that’s not high enough for me to keep Felix, if I can get a good prospect package for him (AND the money to sign FA).

1. I’ll have a bunch of prospects under team-control that should be doing well in MLB or MLB ready at the same time. IN other words, I have a very cheap core.

2. I also have money free for acquisitions to bolster the good young core. Especially when the Figgins and Ichiro contracts are gone.

Rambling aside, I’m fine with saying the mariners are really shitty, not getting much better (have actually gotten worse since the Fister and Pineda trades), and won’t compete for the next 3 years.

I have the same thought as C.J. in posts 20/21.

First get the WAR difference, assign a dollar value to 1 WAR, equity gain should be equal to the the difference.

#22
Exactly. The Rangers traded 4 near MLB-ready prospects, one of whom had a huge amount of value, for about 20 starts from Cliff Lee. You can not use a cookie cutter $/WAR model and get that to balance. You have to factor in how much he improved the Rangers chances of winning the World Series AND how much the Rangers valued that Championship.

The same thing goes for when the Angels traded for Mark Teixeira late in the season even though he was going to be a free agent and they were locks to win their division.

#23
If he had $18 million in surplus value that year, then he had about $8 million at the time he was traded. Justin Smoak alone would have far exceeded that.

#25
1) What matters is how well the Yankees and other teams think he will perform and there is no way that they view him as a 4.5 win guy in decline. They see him as a Cy Young caliber Ace who will give them at least 15 WAR over the next three years in the regular season alone.

2) That math completely ignores his potential contribution in the playoffs, and that is what is really driving the interest in him from the Yankees and other franchises. They see a Cy Young starter as the surest way get them a World Series ring, and that is far more than just a glittery piece of jewelry. It’s what consumes them. If the Yankees don’t win the World Series, then they consider their entire season to have been a failure.

"MGL, what are you talking about?”

I completely misread Aaron’s post in #16, so ignore what I said…

"and won’t compete for the next 3 years.”

“So, a bottom third team has a 33% chance on making the playoffs once within the next 3 years. “

Look, all I asked you is what does it mean that they “won’t compete”.  You say they won’t compete and you agree about the “rule of 3” I noted.

What exactly is the point of saying that you agree to the rule of 3, and then saying “they won’t compete for the next 3 years”?

It’s a terrible way to take a numeric and clear description, and then turn it into the english sentence you did, which seems very unambiguous “won’t compete”.  And the reality is, you don’t mean it as 0% chance of competing.

Your two statement contradict themselves, unless “won’t compete” can be interpreted that they have a 33% chance of making the playoffs over the next 3 years.

All I’m saying is to be clear.

By the way, I will say zero about the Mariners specifically (other than to speak with illustrations) so if I ignore you, don’t take a slight to it.

Right, turning numerical answers into vague English answers is a dangerous way to make a point and an even more dangerous way to make a decision, yet it happens all the time.

Anyway, my main point is that it is ludicrous to discuss these trades, as they do on radio and TV, without addressing the contract issues, regardless of the unique value of player A or player B to team X.

If it makes sense for the Yankees to trade Montero straight up (or more) for Felix, which I still think it doesn’t, would it make sense if Felix were getting 80 mil over the next 3 years rather than 58?  What about 100 mil?  What if Montero, rather than the MLB minimum, was signed to a 3 year 15 mil contract?

It is just insane to discuss these trades without discussing and analyzing contracts. Yet the media, even the so-called serious analysts, do it all the time.

I mean it is a pretty simple formula: What is the value of that commodity to me, how much do I have to pay for it, and what are my alternatives?  The latter is very important for teams like the Yankees. A marginal win may be worth 15 mil to them in some circumstances, yet they are not going to ever pay 14 mil for a win, because they don’t have to (there are generally other, better alternatives).

Midseason trades are fairly different from offseason trades/signings, since you *can’t* otherwise improve your team. Going from 0 -> 4 WAR in one position can make a pretty sizeable difference to your chances of winning the WS in some cases.

OTOH, I think these trades usually end up being terrible. I was pretty unhappy when the Giants traded for Beltran =\.

To build on #6 and #20:

For teams like the Yankees and Redsox that think they have a high probability of reaching the postseason, aren’t they factoring in playoff WAR?

Particularly for an ace pitcher, the playoff WAR contribution would presumably be amplified.  A 4 WAR pitcher making 32 starts (.125 WAR/start) will start at least 1 and possibly up to 7 more games—generating between .125 and .875 WAR just in the postseason.

How much more is 1 postseason WAR worth relative to 1 regular season WAR?

I imagine WAR in playoffs is simultaneously worth more, because the standard of play is higher; and less, because a playoff team probably doesn’t have many replacement-level players.

So, in absolute terms, WAR is probably more valuable. But it’d have to be a upgrade above and beyond a likely non-replacement player, which is likely to come at a premium and be more expensive and thus undesirable.

You could also make an argument that being X WAR better in the regular season is dramatically more likely to get you into the postseason than the X WAR in the postseason is to get you to a WS ring. I don’t know those odds, however.

Tom, looking back through the thread, here’s my exact statement ...

SEA is not going to be contending in the next 3 years (unless something drastic happens).

I was also considering that the LAA have gotten better and should remain really good over the next 3 years, and I don’t see any reason why TEX won’t be good over the next 3 years.

Given that the M’s are in the bottom of the bottom third (including 2012), and the two teams they are trying to leapfrog are in the top 3rd, it would take something drastic for them to compete.

The perspective I was commenting from is “will they compete enough to NOT trade Felix”, and my opinion was “no way”.

If they were in the ALC, NLC, or NLW, it might be different.

I also don’t think the Orioles have a 33% chance at competing in the ALE in the next 3 years. If that’s erroneous, it might be a good idea to bet $1K on the Orioles making the playoffs each year from now until 2014. That bet could be profitable.

I assumed that the 33% figure included teams from all divisions, even thought the divisions are not equal in team quality.

I am looking at it solely through the lens of “not trading Felix” because they may compete during his contract. If I were a GM, the odds wouldn’t be good enough to “keep him” if I could trade him for a very good package of prospects (and as others have said, keep a slightly lesser pitcher for far less money and more years of team control.

I’m not trying to further an argument but to clarify the perspective I was commenting from.

"The same thing goes for when the Angels traded for Mark Teixeira late in the season even though he was going to be a free agent and they were locks to win their division.”

That trade had two benefits to the Angels, since there was no competition that year for the division:
1. Increase chances in playoffs
2. Draft pick compensation if he left

Angels came out pretty good in that exchange.  Mike Trout >>>> Casey Kotchman

"I assumed that the 33% figure included teams from all divisions, even thought the divisions are not equal in team quality.”

Any such rule will need to be revised this year.  Third place teams now can earn a playoff spot.  That doesn’t completely eliminate the disadvantage of being in a tough division (playing more games against good team) but it does lessen it.

... and probably another team in their division.

One other way in which people tend to ignore the contract is in terms of the risk of the deal. A number of Yankees sites I read have bemoaned the trade because the Yankees failed to receive a “sure thing” in exchange for Montero (like Felix or Lincecum for example) and thus the trade is too risky.

In my mind this ignores that the trade impacts the team risk not only through the acquired player. While the error bars on Pineda’s production are much larger than those on Felix’s, acquiring Pineda instead leaves you with 20M extra dollars to spend which is used to reduce the risk associated with the rest of the team. The Yankees already added Kuroda with that money, and they will presumably add a hitter as well. That means that they’ve spread that money among 3 players instead of 1, and the diversification reduces risk. I’m not sure how the average and standard deviation of the production from Felix+RL 5th starter+RL DH compares to Pineda+Kuroda+Mystery DH, but I do think that being able to get the other two guys makes the production look a lot less risky.

Any such rule will need to be revised this year.  Third place teams now can earn a playoff spot.  That doesn’t completely eliminate the disadvantage of being in a tough division (playing more games against good team) but it does lessen it.

Right.  The second wild card changes all these odds. somewhat.  By my estimate right now the threshold for the first wild card in the AL is about 93 wins on average and for the second wild card it’s about 90. 

In about 1 out of every 1000 simulations of 2012 that I ran the second wild card went to an 81 win team.

It’s hard to envision Seattle beating out Texas and LA of A over the next three years, but it’s not hard to see them squeaking past one of them, particularly if the season series between Texas and LA of A swing things in a certain way.

Not likely, but not impossible.

#42, that is actually a very good point.  If I can spend 20 on a 5 WAR player with a track record versus spending .4 mil on a 3 WAR player who has a large uncertainty surrounding that 3 WAR, I still have 19.6 mil to spend on other players which, as you said, greatly reduces the overall uncertainty of my added wins. One thing: Keep in mind that the difference in uncertainty between a player like Pineda and Felix is NOT that large. Pitchers like Felix get hurt all the time and a significant part of the “uncertainty” is random fluctuation, which everyone has (to the exact same degree, more or less)…

That model I find more palatable than a “pure” surplus argument. Consider the trade, then get the team with surplus cash to purchase a free agent worth the difference, THEN compare.

So Felix might be better than Pineda, but with the money you can buy Kuroda/Beltran, or maybe Edwin Jackson. I don’t know.

This stops the model breaking down at the extremes, I feel (Pujols for mid-relief).

Your logic makes complete sense, but you are implying that baseball management is completely logical.  The difference between Felix and Pineda is pretty huge.

Felix is a the youngest established ace in Major Legue baseball.  At 25 years old he has 6 straight season producing at over a 3.5 WAR and the two prior to this last season were over 6.  He was given the Cy Young award at the age of 24.  You honestly estimate that he will be worth 4.5, 4, and 3.5 for the next 3 seasons after posting 3 straight at 5.5 or higher?  Money aside, Felix Hernandez could be the most valuable pitcher in baseball.

Michael Pineda is a 22 year old kid with a dominant fastball, that makes his other pitches better.  He has the potential to develop his secondary pitches, but he still has some trouble against left handed hitters.  While in the minor leagues Pineda had elbow problems that made him lose a season which adds a little more of an injury risk to his resume. 

The Yankees aren’t trying to position themselves to make a run at the playoffs, they are trying to win once they get there.  They would give up the house for Felix because Felix is everything you could want in a pitcher.  Logically does that make sense, maybe not but you also can’t pretend that you have everything you need to predict how a player is going to do in the future.

Seconds thoughts while running through the comments. 

The Mariners are a lot better organization that you guys realize.  Please actually take a second to research a team before you say they aren’t getting better.  Yes The Mariners have lost a lot of games over the last three seasons, but they do have a lot of talent coming up through the system and are starting to develop a solid base or young very talented players.  Felix Hernandez, Dustin Ackley, Justin Smoak, Kyle Seager, & Mike Carp are under 25, have lost their prospect status, and appear to be very solid players.  Likely joining them this season will be Jesus Montero, Danny Hultzen, James Paxton, Vincent Catricala, & Stephen Pryor.  Even after this season there are impact prospect like Taijuan Walker, Francisco Martinez, Nick Franklin, & Guillermo Pimentel.  The Mariners still have around $15mil in budget for this season and have Ichiro’s salary coming off the books next year.  With money to spend, young talent and an extra wildcard spot next year, it is not a very far fetched idea to think that the Mariners could be in the playoffs in the next 3 seasons.

Yes I am a Mariners fan, so I can understand that you guys might think I am just a homer.  Do a little research though.  You will see things like how Hultzen, Paxton, and Walker are all better SP prospects than Betances or Banuelos.  If you disagree, do research and use stats to back up your argument.

"You honestly estimate that he will be worth 4.5, 4, and 3.5 for the next 3 seasons after posting 3 straight at 5.5 or higher?”

I don’t know about those exact numbers (4.5, etc.), but you obviously know little about how projections work, particularly for pitchers. Pitchers, even those with track records of good health, are always a significant risk for break down and not pitch or pitch fewer innings than in the past, and they are always a risk for their true talent to decline for whatever reason.  So yes, if a pitcher has posted several seasons of 5+ WAR, his projection going forward is going to be considerably less than that, due to statistical regression toward the mean and due to the fact that almost all pitchers get worse with age.

If you doubt or don’t believe that, these types of questions can always be answered quite simply. Look at all similar pitchers in history. Make sure you make the requirements broad enough so that the sample size of players and IP is not small. Once you set the requirements (say all 23-27 year old pitchers with an average of at least 5 WAR per season for 3 straight seasons), look at those same pitchers’ WAR’s in the next 3 seasons. I don’t know the answer and I could be wrong, but I would suspect that it is on the order of 11-13.

“ They would give up the house for Felix because Felix is everything you could want in a pitcher.”

That is ridiculous.  Every player has a value at any point in time for any particular team. If the price for the Yankees exceeds that value then the price is too high.  Plus, wins are fungible.  If they Yankees think that Felix will provide 5 WAR on the average, then they would prefer Pineda plus Kuroda, for example, if they think that the two of them combined will provide 6 WAR, especially if the price for those two were the same or less. There are many FA and trades available to all teams. If the Yankees would trade or pay the house for Felix, why didn’t they sign CJ Wilson and Prince Fielder. The Yankees are smart enough to realize that Fielder at DH plus CJ Wilson is worth a heck of a lot more than Felix alone. It is not even close.

Your premises are typical fanboy B.S.  Ridiculous assertions with no data, no research, no evidence, and no proven facts or logic to support them…

I wonder if there’s any significant discount for future wins over current ones; and I don’t necessarily mean deadline deals.

It would seem there would have to be, because the “surplus value” argument would say that you trade a 5m asset (a 3 WAR player earning $10m) for $5m cash now; but not for $5m cash six years from now.

Silly me. Not even a surplus value argument. Any economic (or common sense) thinking shows that you’d be indifferent to the first trade (if I abstractly define the player as a 5m net asset) and dislike the second one.

MGL,
I agree with your logic, and you are correct in that I really have done very little research in the regression of pitchers.  I do understand that typically a starting pitcher loses about 1 mph per year on their fastball velocity.  I would assume that this information means a pitcher who relies heavily on the velocity of their pitches will age worse than someone who relies heavily on pitch movement or accuracy. 

“almost all pitchers get worse with age”

As a whole, on average pitchers do decline as they age, but there is very little public knowledge as to what causes the decline and what pitching attributes are more likely to sustain.  My guess is that a pitcher like Pineda who relies heavily on his fastball speed will age worse than a pitcher who can reply more heavily on his secondary pitches and location.

“If they Yankees think that Felix will provide 5 WAR on the average, then they would prefer Pineda plus Kuroda, for example, if they think that the two of them combined will provide 6 WAR”

Not necessarily, because you are forgetting to take into account the subtraction of the replacement player.  So a Yankees rotation before any of the additions in question would include Sabathia, Nova, Burnett, Hughes, Garcia.  For example purposes, lets say that Burnett, Hughes, and Garcia are all worth 1.5 wins.  The addition of Felix’s 5 wins would also include the subtraction of Garcia’s 1.5 wins and would be a 3.5 win improvement.  The addition of Pineda and Kuroda’s 6 wins would include the subtraction of Hughes and Garcia’s 3 wins for only a 3 win improvement.

“Your premises are typical fanboy B.S.  Ridiculous assertions with no data, no research, no evidence, and no proven facts or logic to support them”

You are right that I have no data, research, or evidence.  I do however believe there is logic behind my words and you aren’t exactly throwing data, research, and evidence back at me.  I also don’t expect you to because I agree with you.

I do believe that you are missing one attribute to comparing the value of Felix vs. Pineda though.  Your logic is based on creating a team to make the playoffs rather than competing once you get there.  The Yankees are obsessed with winning the world series and will take nothing less.  A three man playoff rotation lead by Sabathia and Hernandez is a lot more intimidating than one with Pineda as the no. 2.  Facing the likes of Halladay, Lee, Hamels or Weaver, Haren, Wilson is not easy to do.

"You will see things like how Hultzen, Paxton, and Walker are all better SP prospects than Betances or Banuelos.  If you disagree, do research and use stats to back up your argument.”

Sorry, but I can’t let that stand.  You’re challenging someone to use stats to back up an argument you start with opinion and no facts.  What stats are you talking about anyway?  Hultzen hasn’t thrown a professional pitch yet.

I’m not offering any opinion on which of these pitching prospects are best.  But you opinion without facts is as useless as that of a Yankee fan.  Using stats, you’re bound to get a wide range of opinion on how to best balance performance in AA, rookie ball, Arizone Fall League, College, or whatever stats you have available for this group. Once Baseball America comes out with their top 100 list we’ll have something that may not be 100% right, but at least can be considered somewhat objective.

P.S. the captcha is “fear12” so it’s obligatory to mention Jim Rice.

I’d like to echo CJ/37’s point - the Yankees may be as (or more) concerned with “WAG” (wins above good) than WAR. Or, pushing that to the logical extreme, “WOA” - wins above some team-specific alternative.

It is interesting to see the Yankees do this trade after Cashman was quoted in December as saying they wouldn’t trade Montero unless it were for a talent like Cliff Lee. MGL did a thread about it back then. Maybe they think Pineda is a Lee-level talent? Or (more likely) the discussion of Lee (or Felix Hernandez) is more for public consumption. The Yankees and Mariners likely both value players similarly to how MGL describes, and are well aware that the costs of long term contracts in large part offset the value the players produce on the field.

The interesting thing about the consummated trade is that it’s effectively two young, high ceiling prospects for each other. Rather than a contender trading prospects for an “established” star, each team trades away a potential star in one area for a star in another, because they perceive their needs are greater in the other area.

The Yankees likely think they have a higher replacement level for Montero than for Pineda (they’re seemingly doing all they can to ensure they don’t need to rely on A.J. Burnett), while the Mariners see the opposite chance.

Please actually take a second to research a team before you say they aren’t getting better.  Yes The Mariners have lost a lot of games over the last three seasons, but they do have a lot of talent coming up through the system and are starting to develop a solid base or young very talented players.

1. The M’s had win totals in the 60s in ‘10 and ‘11.

2. They have traded away 7 WAR from the 2011 team, in just 2 pitchers (Fister/Pineda).

3. They’ve received less WAR than that in return. Montero is projected as a 2-3 WAR DH (he’s nowhere close to a 5 WAR DH, and there isn’t one in baseball at the moment).

The Mariners have to GET BETTER just to be AS BAD as they were in 2011 ... as unbelievable as that sounds.

- Ackley is 4-5 WAR full-time, +2 over 2011.
- Montero projects as a 2-3 WAR DH
- Wells is perhaps a 3 WAR guy in full season
- Carp is a 1 WAR player (or so).
- Justin Smoak, 886 career MLB PA, 0.2 WAR. I don;t want to hear anymore about Justin Smoak at least until he reaches the “Mitch Moreland” level of performance (1 - 1.5 WAR 1B).
- Doubling Carp’s 2011 performance (which we don’t do), he’s 1 WAR.
- B. Ryan maintains or regresses.
- K. Seager, 1 - 1.5 WAR full season.

And on and on ...

Their 2012 projections don’t look that good.

I guess I have to concede that they may be better in 2012 than they were in 2011, but to the point that’s like arguing which twin is uglier.

45/

“Consider the trade, then get the team with surplus cash to purchase a free agent worth the difference, THEN compare.”

To reiterate, and echo 54/ (which may be saying it better than I am), I think this is pretty easily solved by each team having it’s own
1) utility (value) from money
2) utility for wins
It seems these points are pretty intuitive and #2 is supported, at the least, but the research on marginal value of wins. #1 may be extraneous - I’m not sure.

This will leading to a different $/WAR for each team.

#52- lookingback over that statement, you are correct.  I do find the lack of control from Banuelos and Betances to be concerning, but there is definitely not enogh information out there on Hultzen and both Walker and Paxton will need more upper level minor league experience before I can substantiate that claim.  I should have just remained quiet and optimistic.

#54 -
“They’ve received less WAR than that in return. Montero is projected as a 2-3 WAR DH (he’s nowhere close to a 5 WAR DH, and there isn’t one in baseball at the moment).”

Incorrect- I agree with your Montero estimate, but he is not the only player that the Mariners acquired for Fister, Pineda, and Campos.

Jesus Montero, Hector Noesi, Casper Wells, Chance Ruffin, Charlie Furbush, and Francisco Martinez are all of the players that head to Seattle in these two deals.  You already meantion that Montero is worth 2-3 wins and Wells could be three.  Noesi could easily be worth 2 wins of value as a starter in Safeco field and the other two MLB arms should be able to have some out of the bullpen.

“The Mariners have to GET BETTER just to be AS BAD as they were in 2011 ... as unbelievable as that sounds.”

At least half of the 2011 Mariners season was spent with Milton Bradley in LF, Jack Cust and Adam Kennedy at DH, Chone figgins at 3rd base, Jack Wilson at 2nd base, and Franklin Gutierrez playing after losing 20lbs that he didn’t have due to a stomach virus.

Offensively you can honestly go position by position and realistically expect improvement at every one aside from SS.

Here is a reasonable expectation from Dave Cameron prior to the Pineda & Campos / For Montero & Noesi deal. 

http://www.ussmariner.com/2012/01/08/what-now-2/

“Their 2012 projections don’t look that good.”

I am not expecting this team to compete in 2012, but saying that they

“are really shitty, not getting much better (have actually gotten worse since the Fister and Pineda trades), and won’t compete for the next 3 years.”

The Mariners already have both of your requirements listed above:

1. A bunch of prospects under team-control that should be doing well in MLB or MLB ready at the same time. IN other words, a very cheap core.

2. Money available for acquisitions to bolster the good young core.

1. A bunch of prospects under team-control that should be doing well in MLB or MLB ready at the same time. IN other words, a very cheap core.

For my preference I just see too many 1 WAR types that probably aren’t “prospects” but just guys that haven’t played in MLB yet. Especially if we would consider them a core to build around.

I am interested in Wells. Zips projects him similarly as they do Milton Bradley. I think Wells could be better than that.

------------------------------

I enjoy Dave’s writing and find his projections for the M’s to be reasonable.

Dave projects them at 75 Wins, which sounds better than the 60s wins they’ve finished with the last 2 seasons.

Question: If we “project” the entire AL teams, how many “projected wins” do we get for each teams? How many total?

I’m wondering if we get more “projected wins” than wins that are mathematically possible?

In other words, would that 75 wins need to go up or down to scale a 162g season when compared to the projected WAR of the other teams. I don’t know if it’s actually 65, 70, 75, 80?

At this point I’m less concerned about being “right” or “wrong” and more interested in the process and being accurate.

If I need to change my opinion to match the data or reality, I certainly will.

"As a whole, on average pitchers do decline as they age, but there is very little public knowledge as to what causes the decline and what pitching attributes are more likely to sustain.  My guess is that a pitcher like Pineda who relies heavily on his fastball speed will age worse than a pitcher who can reply more heavily on his secondary pitches and location.”

I doubt that is true, but it could be.  Without doing any research, you have no way of knowing whatsoever. I could easily be the opposite.  If I throw 95 I can easily afford to lose 1 or 2 mph.  If I throw 88, I cannot.  My logic is just as good if not better than yours.  Neither one of use has any idea what we are talking about though (although keep in mind that I have been doing rigorous baseball research and analysis for over 20 years - when I “guess” that something is wrong or right without doing the research, I am much more likely to be correct than someone who does not do baseball analysis nearly ever day for over 20 years!).

And yes, we all know that teams trying to make the playoffs and win the WS do not necessarily have utilities that in a linear relationship with WAR. 

As far as who a certain player replaces, when I talk about a team adding a WAR, I mean a marginal win, which obviously is a net value. In fact, teams don’t add or subtract WAR.  They add or subtract wins.  WAR is simply a player’s context-neutral value as compared to some arbitrary standard (we use the replacement level standard only because that defines how much we pay him).

@55 - Yes, in theory. But I think the surplus model breaks down when you trade wildly different assets.

Say you pay a 10 WAR player $50 million, at $5 mil per win, and someone wants to trade for him for a mop up 0.5 WAR middle reliever with no upside at league minimum.

Leaving aside that nobody would ever do that (or the existence of a player that good, or a player that good being part market rate), I understand the surplus value point of view. You take your “free” 0.5 WAR and buy free agents with the additional $50 million.

However, one of two things:

a) Due to the supply of good players, it’s unlikely you can go buy another 10 WAR player for $50 million. This is what I mean by “go and look at what you can buy”: there may simply not be enough good options to justify the deal. This is especially true if teams aren’t trying to minimise the price of wins as a goal in of itself, but simply trying to optimise the acquisition of a set number of wins (usually playoff number - get to 90+ as cheap as possible).

b) The other alternative is to purchase a series of “worse” players - couple of 1 WARs here, 2 WARs there. The problem with this is that you’re inherently limited by the fact that you have only 25 roster slots, only nine batters, etc.

In either case, I simply feel the model ignores two constraints - players and roster spots. I’m sure that when it comes to trading similarly-valued assets it’s fine.

"If I need to change my opinion to match the data or reality, I certainly will. “

The ideal poster of this blog.

Anyone else who does not subscribe to this viewpoint should simply leave this blog and never come back.

Question: If we “project” the entire AL teams, how many “projected wins” do we get for each teams? How many total?

I’m wondering if we get more “projected wins” than wins that are mathematically possible?

In other words, would that 75 wins need to go up or down to scale a 162g season when compared to the projected WAR of the other teams. I don’t know if it’s actually 65, 70, 75, 80?

It really depends on where you set replacement level.  If you set it too low you will have a bunch of teams that look better than they will be in reality.

It’s probably a good sanity check to add up the league WAR and see if your replacement level makes sense, but I think the exercise of projecting a team by WAR isn’t necessarily done to be precise.  It’s more of a quick and easy way to ballpark a team’s talent level. 

Since teams don’t necessarily perform at their talent level anyway, you can’t get super accurate.  You can probably get into a range of +/- 5 wins using WAR, which is probably good enough.

The idea that you can rule out a team being able to compete 2 and 3 years into the future is preposterous.

Here is the opening day lineup for the Tampa Bay Devil Rays in 2006:

Julio Lugo
Carl Crawford
Jorge Cantu
Aubrey Huff
Jonny Gomes
Travis Lee
Toby Hall
Damon Hollins
Joey Gathright

Here are their top starting pitchers that year (in order of IP):

Scott Kazmir
Casey Fossum
James Shields
Tim Corcoran
Jae Weong Seo
Mark Hendrickson

That team was absolutely terrible and went on to lose 97 games. Not only that, they had practically no young, high-end talent to build around. If I asked you then how that franchise would fair in 2008, you would say that they would have absolutely no chance at contending. Not only were they a “shitty” team, but they had a low payroll and are in the same division as the Yankees and Red Sox. If Toronto can’t go to the play offs, then there was absolutely no way that Tampa Bay could do it. Well they not only won 97 games in 2008, but they won their division.

Seattle is far better positioned now than the DRays were in 2006 so writing them off for years to come makes no sense. Ackley, Smoak and Montero are all far better players than any of Tampa’s hitters except for Crawford, and Felix is better than Shields or Kazmir. The M’s also have far more money to spend, and for as much dough as the Angels and Rangers have thrown around they still aren’t in the same class as the Red Sox and Yankees. You can dismiss Seattle’s chances in 2012, but anything after that is up in the air.

#44/MGL: could you clarify this?

Keep in mind that the difference in uncertainty between a player like Pineda and Felix is NOT that large. Pitchers like Felix get hurt all the time and a significant part of the “uncertainty” is random fluctuation, which everyone has (to the exact same degree, more or less)…

Do you see Pineda as less of, the same as, or more of an injury risk than Hernandez? Could you quantify that risk differential? And setting that differential aside, could you quantify the difference in random fluctuations between the two—or show evidence that the random fluctuations in pitchers with Pineda’s profile are more or less the same as the random fluctuation’s in pitchers with Hernandez’s profile?

If there’s research out there on this, I’d love links to it; I don’t know that I’ve seen this sort of analysis before, but I’ve been thinking that it’d be nice to see more written about WAR variance and its sources.

Thanks in advance…

(Also, with a “like25” captcha, I feel obligated as a Rangers fan to state that I do not just like, but love Mike Napoli.)

@ Aaron

SEA is not going to be contending in the next 3 years (unless something drastic happens)

1. I think the words within the parentheses allows for drastic improvement or roster turnover or injuries to rivals, and things of that nature.

2. Again, I was discussing the aspect from the perspective of whether to trade Felix for prospects or not (assuming he walks at the end of 2014).

Having a chance and having a reasonably good chance are too different things.

when we talk about a good young player and we say “he has a chance to be something really special”, we aren’t exactly saying that with “1 in a million chance” scenario in mind.

Likewise in common speak when I say Seattle is ‘poopy’ and won’t contend within the next 3 years, I’m really not saying with certainty that they have a 0.0% chance at contending in any of the next 3 years ... only that the % chance of competing is too small to “bank on it”, or in this case to keep Felix when he could have been traded for a collection of good prospects (and kept their young pitcher with years of team control).

I will backtrack from the “SEA got worse” part. I think the case can be made that the 2012 Mariners should be better than the 2011 version.

I hope we’re not really using TB ‘08 as an example of what could be expected of any team.

I might as well use Brady Anderson as an example of how Carlos Beltran could hit 50 bombs this year. He could.

I do agree that projecting more than a season or two in the future is crazy. 3 years was included in this scenario because that’s how many years Felix has left on his current contract. Lesson learned.

I think the words within the parentheses allows for drastic improvement or roster turnover or injuries to rivals, and things of that nature.

But those should be expected! That’s like saying “Bryce Harper has no chance of being a good major leaguer (unless something drastic happens, like getting better with age)”.

If we were to assume that the M’s were going to carry the exact roster the next 3 years as they did in 2011, then of course they would have no chance. But what kind of assumption is that?

#64, there are 3 elements to the uncertainty of a player’s projections, assuming that we know his role and how much playing time a manager is going to give him (if we don’t assume that, then there are 4):

1) His actual true talent. When we project Pujols to have a .390 wOBA in 2012 or Felix to have a 2.90 FIP, or Pineda to be worth 3 WAR (based on his FIP or ERA and his playing time), we are really estimating true talent (and role and playing time).

We are not certain about the true talent for anyone. The uncertainty around our estimate of a player’s true talent is largely a function of the amount (and quality) of data that goes into our projection system. A pitcher with 3000 prior MLB TBF (or IP) should have around the same uncertainty surrounding his true talent estimate as another pitcher with the same number of historical MLB TBF.  There are other considerations such as, was there an injury history, do we think the player’s true talent changed recently due to learning a new pitch, a change in velocity, repertoire, mechanics, etc., or perhaps even the fact that pitcher A had a narrow range of historical performance and pitcher B had a wide range.  But, by and large, the error bars around our estimate of a player’s true talent is a function of the amount of historical (and other) data we have on him.  For example, if pitcher A were estimated at 3 WAR with a .5 standard error (we think there is a 95% chance that his true talent is 2-4 WAR), if pitcher B had the same number of historical chances, I would not think that any other factor would make his standard error larger than .7 or smaller than .3.

2) Then there is the chance of injury or anything else that changes his true talent next year. This is different than our estimate of his true talent, #1, because that is really an estimate of his true talent now, as opposed to next year. There is an overlap with 1 and 2 and the line between them is blurred, but I am trying to make a distinction. For pitchers, not only is the estimate in #1 above harder than for batters, but #2 is a lot harder. Pitchers change their true talent much more often and more dramatically than do batters (we think, at least).

3) Finally, even if we know a pitcher’s true talent at the exact moment that he pitches each and every game, his performance obviously will not necessarily exactly reflect that true talent.  The distribution of his expected performance will be centered around that true talent, but it will randomly fluctuate around it as well.  It is the same thing as if we flip a coin 10, 100, or 1000 times.  The coin’s “true talent” will be 50/50 heads/tails but of course the actual results will deviate around that randomly (assuming that the “flip” is unbiased of course). Just as we can mathematically describe exactly what the distribution of those coin flips will look like, we can also pretty much exactly describe what the distribution of performance around a pitcher’s true talent will look like, given X number of batters faced.

If we knew Felix’ true talent (say, 2.90 FIP) and Pineda’s true talent (say, 3.20), the distribution of performance around those numbers would pretty much be the same for both pitchers.  This is the random fluctuation that I mentioned and you asked about. It is pretty much identical for all players with the same number of performance opportunities. So the “error bars” based on this #3 are almost exactly the same for Felix and Pineda given the same number of IP or TBF in 2012.

Now, let’s look at #1 and #2 for Felix and Pineda. 

#1, Felix has many more historical TBF than Pineda, so there will be a smaller uncertainty for him on our estimate of his current true talent, strictly based on his past performance (and “scouting,” if we want to use that).

For #2, I don’t know who would have less of a chance of being injured.  Felix has a track record of being healthy, but Pineda is younger (I assume younger players have less of a chance of being injured). Pineda, though, probably has more of a chance of changing his true talent through learning, new pitches (like he needs to work on his change-up or develop another pitch).

So we have #1, less uncertainty for Felix, #2, probably less for Felix, but maybe not, and #3, pretty much the same for both.

In order to get the total uncertainty, you add up the variances of course.  I am suggesting the variance for #3 is the highest, by far, of the two (depending on how much they pitch in 2012), so when you add them up, I don’t think you are going to get a gigantic difference between the two.

Of course, that depends on your numerical definition of large, small, gigantic, etc. Which is why it is generally fruitless to have (what should be) a quantitative discussion using qualitative descriptions.

So, Pineda might be projected at 3 WAR plus or minus 2 WAR (at 2 sigma - 95% confidence interval) and Felix might be 4.5 WAR plus or minus 1.5. I am taking wild guesses at these numbers, although we could estimate them by looking at historical data.

It is left to the reader to decide whether the error bars of one is much larger, larger, or a little larger than the other, and what that means in terms of decision-making by a team.

To me, error bars for one player do not mean much at all.  That is because the only thing that counts are the error bars around my team’s expected w/l record and no one player’s uncertainty is going to effect that very much, especially if I spread out my WAR among as many players as possible - another good reason not to play a lot of money for a superstar player.  If you do that, you automatically increase your team’s w/l uncertainty…

#67/MGL

Thanks for the detailed explanation; much appreciated.

Here’s an (the?) issue I’m trying to think through. You write,

The uncertainty around our estimate of a player’s true talent is largely a function of the amount (and quality) of data that goes into our projection system.... For example, if pitcher A were estimated at 3 WAR with a .5 standard error (we think there is a 95% chance that his true talent is 2-4 WAR), if pitcher B had the same number of historical chances, I would not think that any other factor would make his standard error larger than .7 or smaller than .3.

I am trying to suss out whether there’s data supporting that last sentence. In other words, is there evidence that a pitcher like Hernandez, in his prime, does not have a narrower, higher-peaked WAR probability distribution that is possibly even skewed right, relative to lesser pitchers who might have wider, less peaked, and more normal distributions?

This is taking injuries out of the picture, and assuming an equal number of historical chances.

Apologies if this doesn’t make sense; it’s late, and has been a bit of a long day, and I appreciate your patience…

I mean left-skewed, of course.

All pitchers’ distributions (of likely true talent) are skewed left. I don’t know what there would be about a pitcher like Felix that would narrow that distribution as compared to a lesser pitcher. And I don’t know what you mean by a “higher peak.” Dont all distributions that are narrower have a higher peak by definition?

Upon further consideration, you might be correct. The better a pitcher is the more left skewed that estimated talent distribution may be. I’m not sure what that would do to the width of the curve or the standard error of the mean estimate.

@ Aaron

But those should be expected! That’s like saying “Bryce Harper has no chance of being a good major leaguer (unless something drastic happens, like getting better with age)”.

Does anyone consider improvement due to increased experience/maturity as being a “drastic event”?

I would consider some of the Mariners 1-WAR 25yo players putting up 3-4 WAR to be drastic.

Dustin Ackley putting up a 4-05 WAR season would not be drastic, but closer to expectation/projection.

----------------------------------------

Does everyone interpret my comments as meaning that the mariners are going to have the same roster from 2012-2014?

My comments were based on the idea that they are going to have to bring in A LOT more talent to compete in/by 2014. .. well, or have the Rangers/Angels just experience horrific luck with injuries/performance.

If someone could show me that it was actually more common than not for teams within similar performance in consecutive seasons to get significantly better, I would find that convincing.

But, until seeing something like that I’ll stick with the idea that bad teams are more like to stay bad than become contenders. Drastic things that are positive don’t seem to happen a lot.

Can I/we just say “The mariners have a lot of improving to do over the next two years to be able to contend in 2014”, and leave it at that?

I would have to say this is the only time I have ever gotten into a discussion about what a team might do 3 years from present, and the only reason I did was because 2014 will be Felix’s walk year.

Getting back to the big point here, though, I don’t think I’ve heard MGL’s opinion and I’m curious what he’d say:

If you were GM and the current $/WAR was $5m and you were offered up a true-talent 0.5WAR back-end starter signed to $500k/1yr deal (surplus $2m) for your true-talent 5 WAR all-star signed to a $24/1yr deal (surplus $1m), would you do it? It seems that you have to say yes.

If so, that seems like a pretty radical conclusion, and one we will probably never see happen in real life. But, I can’t see any other way out of this conclusion (unless some of the other ideas that have been discussed are added into contracts).

ou were GM and the current $/WAR was $5m and you were offered up a true-talent 0.5WAR back-end starter signed to $500k/1yr deal (surplus $2m) for your true-talent 5 WAR all-star signed to a $24/1yr deal (surplus $1m), would you do it? It seems that you have to say yes.

While that is a very good question for discussion, it cannot be answered with a yes or no.

It is a very complicated issue. I’ll try and address a few of them.

First, as a GM or team, I can’t be known as someone (or some team) who is going to trade a star or superstar (or any player for that matter) at the drop of a hat. If I do, then no one is going to want to sign with my team, at least not without a no-trade clause.

Second, and similarly, is a half win surplus worth the wrath I might get from the fans, media, players, etc.? And perhaps that wrath turns into fewer fans at the gate.

More importantly, however, the $5 mil per win figure is an arbitrary number and means almost nothing to a team in and of itself. It represents a very approximate value for how much I CAN purchase a win for on the FA market. In practice, that exact number rarely enters into the decision-making process.

What matters in a personnel/contract/trade/signing decision is how much a marginal win is worth to my team in extra revenue (given my team’s exact situation), now and in the future, and how much I can pay for a marginal win either in the FA market or otherwise.  Those numbers of course are rarely going to be exactly $5mil, especially the former. Sometimes not even close.

Another thing is that my job as a GM and my “job” as an owner who is strictly interested in the bottom line are completely different.  As a GM, my job is to give my team the best chance to make the post-season and win the WS given the budget I am alloted, again, both now and in the future. As an owner, presumably it is to make as much money as possible, now and in the future.

So, for example, in your question, from a GM’s perspective, I would have to ask, “can I go out and purchase another 4.5 wins (not a 4.5 win player) for 22.5, in which case, I am better off (other than the issue of wrath from the fans and being known as a “nit” of a team). If I can’t do that, then I have to re-assess that trade.

From an owner’s perspective, what if a marginal win is worth less 5 mil to me in revenue? Then, I would give away that 5 WAR player for free and I would certainly make that trade (again, PR and similar issues aside). What if a marginal win is worth 10 mil to me?  Well, then I would want to stock as many wins as possible (although, obviously that win value is not going to be linear) within reason.  So the question would be, as for the GM, can I go out and repurchase a similar player or players to get back 4.5 or 5 wins?

There are more issues, which I don’t have the time to discuss right now…

From now on, let’s all disavow ourselves of the notion that a marginal win is worth $5 (or any particular amount). As I said, that number is rarely relevant in a discussion about a specific trade or acquisition.

As I said, all it represents is the approximate average cost of a marginal win on the FA market. That may or may not be particularly relevant to a discussion of a trade…

In what sense, then, does a player have surplus value? Revenue less cost?

To clarify, should we consider a player’s surplus value as some product of revenue per win and wins above team replacement (ie. wins at the margin), less contract cost?

Of course, a team with revenue per win twice that of every other team in the league probably won’t pay for “surplus production” at that rate, maybe just 1.1x the other team’s rates or so. I think.

CJ, yes, and yes.  Exactly the same way any other company figures surplus value…

From an article on BP by Jay Jaffe:

“Now that the Yankees have dealt Jesus Montero to the Mariners, they’re in search of DH options.”

So if the Yankees would have paid just about anything (read many of the posts above) for Felix’ 4 or 5 wins, including giving up a 40 million dollar asset in Montero AND paying Felix 19 mil a year, why wouldn’t they sign Fielder as their DH for 40 million a year?  Surely he wouldn’t turn that down…

Does anyone have any sort of data as to the effect of team wins on revenue? I’m asking if such a study exists at all.

So, by trading for Pineda, which of the following do you think is the most likely case?

a) Wins: the Yankees think Pineda adds more wins over their #5/#6 starter (teams usually get lots of starts from sixth/seventh guys, if I remember right) than Montero does over their DH;
b) Cost: the Yankees think Pineda will be cheaper than Montero, for whatever reason, for the production he produces (this would be projected salary in the prospect-for-prospect case); or
c) Rev/win: Pineda is so much better than Montero that the additional wins increases the Yankees’ revenue per win (assuming revenue per win increases as wins increases).

I never thought of (2), but maybe Montero, as an all-bat player, might be overvalued in arbitration relative to other players. Of course, I really doubt anyone thinks this, plus Montero might be docked hard for being a straight DH. In any case, if you’re trying to game the arb system you probably don’t trade for a young strikeout pitcher.

#79

Because Fielder isn’t a Hall of Fame-caliber starting pitcher. Why that matters so much to a team like the Yankees has already been explained.