Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

Big baseball fans love October; huge baseball fans love November, too. Now, I’m not talking about the occasional World Series that peeks around the corner into November, I’m talking about Hot Stove season. The Hot Stove League is a great circuit for baseball fans, because every team is currently undefeated and nobody other than the most recent draft class is untradeable. Anybody could potentially be out there on the mound pitching your team a shutout on the first Monday of April. Anybody could be hitting a grand slam for your team in the first inning.

Unfortunately, this is when we see some very ridiculous trade rumors come about and people take them seriously. Those among us who have a sense of how baseball economics works can smell these a mile away. In this article, I am going to take a look at a few huge mistakes that people make when thinking about baseball trades. This can serve one of three purposes. First, it can help you figure out whether your GM helped or hurt your team by making a trade when one does happen. Second, it can help you figure out whether a trade rumor being floated makes any sense. Third, it can help you invent a ridiculous rumor to float around on the internet that you can pretend is plausible. Sound like fun? Good. Let’s get started.


1. Salary Matters.
Stop simply comparing players and trying to trade players for others of a similar caliber. You need to take into account the opportunity cost of spending money on a certain player. Roy Halladay is the hot topic this winter, and one season of Roy Halladay will win you about six more games than replacing him with a cheap replacement player. However, Roy Halladay costs $15.75 million for 2010. A team that took that money and spent it on free agents would get about 3.5 wins above replacement. What that means is that Roy Halladay is worth about 2.5 wins for this season if you trade for him. So, he is worth the same as a decent but unspectacular third starter would be for free. Don’t compare him to a young ace. No one will trade him for a young ace who is a sure thing.


2. Years Under Contract Matter.
The most ridiculous trade rumors come when people do not consider how many years are left on a player’s contract. Roy Halladay with all expenses paid is worth six wins. So are two players who are worth three wins. So is one player who is worth three wins for two years in a row. The counterargument to this is that sometimes the extra three wins gets you over the hump but the first three wins do not get you into the playoffs on their own. However, this ignores the fact that the standard deviation of a team’s win total will be about six or seven wins different than their true talent level. There is too much variance to know exactly where you and your divisional rivals will finish. Therefore, the value of two years of a three-win player is very close to the value of one six-win player when we ignore everything else. Trading a young player with half a season under his belt, but who has a pretty well understood skill level is a very valuable thing-players do not reach free agency until they have six full seasons under their belt. That player is going to get the league minimum for three years and three years under salary arbitration where they will get salaries of approximately 44 percent, 61 percent, and 64 percent of their market value (according to John D. Burger and Stephen J. K. Walters of Loyola College in Maryland). Therefore, getting six years of a player only requires “paying” for 1.7 years of their actual value. So, it’s like getting 4.3 years of their value for free. Thus, even a 1.5-win player with six years remaining on his contract can be worth as much as Roy Halladay.


3. Draft Pick Compensation Matters, but the Probability of Signing the Player to an Extension Does Not.
The price of any free agent depends on the opportunity cost of signing them. Teams bid for free agents in an auction setting, which means that teams are going to bid players up to their true value. Of course, teams will have slightly different valuations for individual players, but teams will bid each other up to the point where they are no longer better off if the player says yes. For Type-A free agents, that is the point, where they surrender their first-round draft pick, or if they are the team whom the player played for the previous season, where they forgo the opportunity to get a draft pick from another team and a compensation pick between the first and second rounds. The value of one pick near the end of the first round is about $2.5 million, and the value of two picks is about $5.0 million. Of course, there are other little factors that affect a player’s value, but chances are good that you are just not going to get a bargain when you sign your own free agent-especially if you just acquired him by trade. It’s a good rule of thumb to assume that the benefit of signing a guy who is likely to get Type-A compensation is worth $5 million beyond the net value of holding his contract, or approximately 1.1 wins. If the team re-signs him, they are probably paying a price approximately equal to his value, minus the two draft picks. The general point is that whether your team can sign the player to the extension is not really all that relevant, because their alternative is going onto the free agent market and paying the fair price for talent there instead.


4. Differing Estimates of Player’s Skills are Not Necessary for a Trade to Happen.
This is related to J.C. Bradbury’s “Hot Stove Myths” from last week, in which he explains that there does not need to be a winner and a loser in every trade. Although sometimes GMs will trade two similar players because they both feel that the other’s player is better, these trades are far less common than a situation in which two teams have comparative advantages in different things and swap with the intention of helping each other out. The most common trade is a “win-now” team trading prospects to a “rebuilding” team for their current star. This is a situation where the relative marginal value of a win is higher for one team now and the other later. These are obviously common at the trade deadline when certain teams no longer have a chance of winning. Sometimes teams trade when they both have a surplus at a position that the other team has a deficiency-again this is an example of comparative advantage. The most illogical fake trade rumors are the ones in which one team is not getting any obvious value in an area where they would need it. In both of these common types of trades, the teams probably have very similar expectations with respect to the players dealt, but they do not value this production equally because of differing player alternatives available to them, or different years in which they plan to be more competitive.


5. Quantity Does Not Make Up for Lack of Quality.
The most obvious fake trade rumors are the ones where one team gets one superstar player for fifty of the other team’s players. There is such a thing as replacement level, and trading fifty random guys who may or may not land on a big-league bench for an ace does not give the other team any production that they could not replicate from players on the waiver wire or in the high minors of their system.


6. You Cannot Apply These Rules to Deadline Deals Without Adjustments.
The methodology specified above where you consider the number of years that the players have left on their contracts, the free agents that would be available for the salary spent on those players, and the draft picks that a team would net by acquiring a player who will soon reach free agency-these are all good rules for the offseason. However, they make no sense during the regular season. The dollar value of a win is based on the probability that that win determines whether a team makes the playoffs. We have seen the graph of the marginal value of a win before, which is largest at the point where the win is likely to determine whether a team makes the playoffs. Clearly, most free agents are acquired by teams who are near that breaking point. However, with only 60 games left in a season instead of 162, the odds of one win being the determining factor can be much higher. At the beginning of the season, there is a strong possibility that a team will fall well behind the pack or get out to an insurmountable lead, but in the middle of the season a team with a slight division lead or deficit has a lot to gain by adding a star player because they know that the odds of those wins affecting whether they make the playoffs are higher. A good example of this is my article on Roy Halladay this past July.


7. The Owner’s Bank Account Does Not Matter.
The question of whether a team will sign a big free-agent contract or trade of an expensive player is frequently characterized as whether the owner is “willing to pay for a winner” or “can afford it.” These are silly red herrings in the discussion. Although team owners probably prefer winning to losing, they spend money in accordance with whether it will add to their profit. The owners do not spend money because you are having a hard time at work or because your significant other just left you, they spend money to make more money than they spend most of the time. If they think that adding a free agent will bring fans to their games and playoff games with even more fans, they will do it. Similarly, the state of the owner’s finances rarely has anything to do with baseball. The amount of money that they just made or lost in the market almost definitely has nothing to do with whether their team will spend money on a free agent that will paid out over time as the players brings money in. Again, that comes down to whether doing so is profitable. Owners are neither being cheap or nice-they do not spend money to fix your life; they spend money to get your money.

These seven rules should provide a nice framework for readers to analyze the trade market or to make up rumors of their own. The important thing to remember is that the quality of talent involved is only a small portion of the issue at hand. An excellent source on the economics of trades and free agency can be found here, and J.C. Bradbury’s article cited above provides some other important myths to keep in mind when looking at the market. Remembering the way that dollars relate to wins is essential in understanding why these transactions take place, and should make the Hot Stove season all the more exciting as you plan what you think your team should do.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
saucyjack88
11/17
About #5: Someone should have told that to Cleveland GM Mark Shapiro BEFORE he traded Cliff Lee to Philly.....
Mowstangs
11/17
but, but... Carlos Carrasco!!!
irablum
11/17
about #5. quantity with quality can overcome more quality.

my example: Teixeira. (as Brave fans the world over cringe).
mhmosher
11/17
No kidding...they got positively rear-ended on that one.
buffum
11/17
I like the way you looked at WAR in the first two items, but I am not convinced that a lot of GMs share your perspective on those. I can count a number of trades that were made that do not adhere to those principles, so I'm not sure if they apply as universally as you seem to imply.
swartzm
11/17
Certainly there are trades where mistakes are made, but in general these rules are more or less true. If somebody makes a mistake or if a GM tries to save his job with a move, then there exceptions but there aren't great examples of this clearly being violated. In general, it is true.
pjbenedict
11/17
I love this article. It presents information that I haven't seen summarized this way before. Thanks.
ddanyc
11/17
How does item 7 square up with the often reported November story along the lines of "Mr. Owner has given Mr. GM a payroll budget of $X million"? Pre-determined payroll caps do not suggest that there is a separate ROI calculation made for each potential free agent signing or trade acquisition.
hessshaun
11/17
Because there is always a "market", for anything. You can assume that if they don't want a 3B, OF, or RP, their budget will be pretty close to zero. I am not disputing what a team needs or does not need, but the owner has the pocket book and the ability to analyze, rationally or irrationally, where his money goes into the market.

And cmon, seriously, when you have Dusty Baker as a manager, that's easily 8 to 9 wins right there.
swartzm
11/17
You're mistaking payrolls set based on profitability from payrolls set arbitrarily. I would suspect that in my circumstances, the owner gives the GM a budget based on their discussion of what kind of win totals are possible with different budget levels, and then decides how much that is worth. The effect would be the same.
Oleoay
11/17
About #2, while it is true that "Roy Halladay with all expenses paid is worth six wins. So are two players who are worth three wins.", there's still something to be said for roster optimization. If there aren't two slots in the starting lineup/rotation that are replacement level, and thus, can be upgraded with two players worth three wins, then it makes sense to put that entire investment in one slot. In addition, it's easier to find players worth two to three wins inexpensively, so ideally, you shouldn't have to go the free agent route to acquire them... but you would have to go the free agent or trade route to get a six win player.
mglick0718
11/17
Similar to preceding post, in reference to #1 on the list: you say "Halladay is worth about 2.5 wins for this season if you trade for him. So, he is worth the same as a decent but unspectacular third starter would be for free." this seems like a false comparison -- why can't I have Halladay and the free unspectacular third starer? Halladay is replacing my otherwise even-less-spectacular fifth starter, not that guy.
mbodell
11/17
I agree with both of the above, roster slots have value. In addition, many teams end up using starting pitchers who are at or below replacement level. Thus an ace #1 starter usually has less of a chaining effect (he displaces marginal talent), where as a star position player will often deplace an ok better than replacement player. Thus a true ace like Halladay is worth a lot, even for only 1 year.
swartzm
11/17
What you're missing is that in the middle of the offseason, you can trade a position player frequently if he has value. And if you don't have a valid trade alternative, what you are doing making the trade in the first place? We could evaluate the idea of the Marlins acquiring a shortstop, but they wouldn't be in that trade market unless they had a plan for Hanley Ramirez. The point of this is to analyze trades that are filling holes that currently exist or holes that will exist when you trade players. Your point is well taken for in-season trades, and that's precisely why I said above that these rules do not hold for in-season trades because a roster is not as flexible, but you can certainly trade a position player or move a 1B or LF to DH in the AL to make a trade work. The roster spots have value hypothesis only makes sense for situations where you cannot trade players who have value, which is usually not true.
bsolow
11/17
This is a misunderstanding of Matt's argument. The argument he's making is: assume you have a full roster for now, where the 5th starter ~ replacement level. You can acquire a "third starter" for free (this is clearly not true in the free agent market, but let's assume it is for now) or Roy Halladay for $15.75 mil. Which do you prefer?

The correct answer is that you shouldn't care, as long as you can spend the 15.75 million you save from not acquiring Halladay on upgrading your other players to the tune of 3.5 wins. In other words, if you have a prospect who's not terribly exciting but could be a decent mid-rotation starter this year, i.e. Michael Bowden or maybe Tim Alderson for lack of a better example, trading him for 1 year of Roy Halladay at 15.75 is a wash, provided you can reallocate that 15.75 into 3.5 wins, which is what usually holds on the free agent market.

Roster construction is an important constraint, though, especially for teams with players/contracts locked into positions. In some instances, Halladay's 6 wins at 15.75 are more valuable because it's not possible to reallocate that money into 3.5 wins elsewhere on the roster.
bsolow
11/17
Actually, I understated the case -- the first year of that deal is a wash according to Matt's argument in #1, the next 5 years favor the Blue Jays according to #2.
swartzm
11/17
Mglick0718-- The idea isn't that it's either/or. The idea is that the values are equivalent, i.e. if someone said "I'll trade you this unspectacular third starter for one year and I'll cover his contract, or alternatively Roy Halladay for one year but you have to pay him" then you are indifferent.
Oleoay
11/18
Indifferent, in some sense, but does the equation change at all for a playoff team? Is it more valuable, as a playoff team, to pay to have a Roy Halladay or to have a free third starter?
swartzm
11/18
The standard deviation of win total versus true talent level is about 6.4 games, so with the season not yet started, there's just not going to be much difference in the marginal return to wins when you add the first three wins versus the second three wins. For example, if you have an 86-win team (based on talent level) that you could turn into an 89-win team with a three-win player and a 92-win team with a six-win player, you would think that going from 89 to 92 is more valuable than 86 to 89. The problem is that you're really considering taking a team that could win anywhere from 74-98 games to a team that could win 77-101 games or 80-104 games. The confidence interval is just too big to be informative.
Oleoay
11/18
With that kind of spread of 74-98 games, are you suggesting that if you were a GM, you would almost never do major trades or acquire major free agents in the offseason annd instead, wait until the middle of the season to see where in that spread your team was performing? In other words, the confidence interval is too broad to do any expensive moves (in terms of dollars or talent) in the offseason?
swartzm
11/19
The marginal value of a win is still positive when you jump from a range of 74-98 up to 77-101. Keep in mind that the middle part of those ranges are more plausible. If you increase your talent level by one win, you probably increase your chance of making the playoffs by 6% if you were a 86-win team. If you increase your talent level by three wins to an 89-win team, you probably increase your chance of making the playoffs by about 18%. If you increase your talent level by six wins to a 92-win team, you probably increase your chance of making the playoffs by about 36%. These are me rounding numbers, but the point is that every time you add one win worth of talent, you increase your chance of making the playoffs by about 6%.
Oleoay
11/19
I understand that concept, but there's still a threshhold of wins where the additional win doesn't make much of a difference in playoff chance. Most likely, the jump from 80 to 81 wins won't increase playoff percentage. You might have a 90 win team, but that'd put you in third place in the AL East.

So, with that in mind, does it make more sense to acquire players midseason after the division has stabilized some than in the offseason? The added advantage of that kind of move is that talent can be acquired for cheaper in terms of salary and talent since you're not trading for a full season's worth of a player.

Or, to look at it another way, should the Phillies have acquired Cliff Lee before last season instead of partway through the season, and if they did, would it have cost more or less in terms of talent/dollars?
swartzm
11/19
The jump from 80 to 81 wins in talent doesn't really hit the thick part of the distribution, so it's only something like 2-3% rather than 6%. But being a 90-win team in the AL East probably increases their probability of making the playoffs by 4% or even more.

For the in-season acquisition, keep in mind that the price moves with the value. If there are more buyers than sellers, the cost could pretty much break even. My Roy Halladay articles pretty much showed that you trade for a star at mid-season, he can be worth 2/3 of his revenue value with 1/3 of the season left, and for a guy like Halladay with his salary, the value before the season should be about equal to the value in season. And then that's basically saying the price before the season should be equal to the value in season, or at least close to it.
Oleoay
11/19
So, if the value and price midseason is similar to the price and value in the offseason, then perhaps a good GM plan would be to acquire the three-win type players during the offseason (when there should be more of a flood of freely available talent), then see how the season turns out and ramp up to a superstar once the team's playoff chase chances sort themselves out... and if the team doesn't make the playoffs, invest that unused salary in player development/scouting/dominican academies, etc.

Is that a potentially good GM strategy, particularly for small markets with less room for error on their payroll?
swartzm
11/17
Frequently there are two spots where there is a replacement level player. Generally, there is some spot on the diamond where you know you would get replacement level production if you did nothing all off-season and there usually aren't 5 starters that are considering above replacement level on any team, certainly before the off-season starts. My point is that two players who are worth three wins should each be half as expensive as one six win player-- if not, then you're replacement level is too low. I would actually suspect that fewer six win players reach free agency than three win players by a very large margin.
Oleoay
11/18
I agree that two players worth three wins should be less expensive than one player at six wins (with the exception of those GMs who overpay for mediocrity). I also agree that there are fewer six win players that hit free agency than six win players, just from the sheer fact that there are less six win players than three win players.

Yet, I also imagine there are many three win players who have a bad year and regress to replacement level. That might not happen as often for a six win player, and there is some value to that as well.

Besides, if a team has too many holes, there's little reason to get a six win player anyway if that player is not enough to push the team into contention.
swartzm
11/18
A three-win player means that on average he would be worth three wins. It takes into account the probability of becoming a replacement level player. The price of two three-win players and of one six-win player should be equivalent because of the response I wrote above.
tradeatape
11/17
Matt--

Nice article--thank you! If you have time, I was wondering vis-a-vis no. 7: one of the things I hear all the time about the Detroit Tigers is that the economy is suffering in MI, and so they need to unload some payroll (I'm sure many fans of other teams hear the same thing about their towns). This is despite the fact that the Tigers had the fourth best attendance record in the AL in 2009, so clearly Illich must be making some money. So, does the overall economy of the city/state the team plays in matter for trades, or is it a red herring too?

Also, does the amount of 'dead weight' a team is carrying (e.g., the aforesaid Tigers have a ton in Dontrelle Willis et al.) matter?

Thanks!
swartzm
11/17
Thanks. It's a tough question. It should come down to the margin value of a win-- how much more fans does a win add than before? Chances are that goes down when the economy gets worse, so in that case, it would drive down the payroll that maximizes profit. For the Tigers, their attendance may be high but how are their ticket prices? Do they send out a lot of e-deals where they offer people specials? In modern stadiums, I would guess that it's usually smart to price tickets in a way that you sell out most games, but that's very different when you're selling out with expensive prices or low prices with lots of emailed deals that lower the price further for games with little appeal.

The dead weight like Dontrelle Willis would not necessarily matter per se, but if the team is trying to avoid losing money rather than maximizing profit without concern with risk, then the story changes. They might prefer to play it safe and lower payroll even if the dead weight doesn't change the expected effect of other spending decisions. It would pretty much come down to how the team is equipped to deal with losses.
BeplerP
11/18
Matt: Many thanks, you have a lot of thought-provoking things to say on the topic I find most intriguing now that no one's playing ball in my time zone until February...Have you considered doing a "scorecard" on trades on a seasonal-retrospective basis? Would this isolate astute, or merely lucky, managements? I know we get a great deal of this piecemeal during the season, on BP and elsewhere, but it's on the fly, usually on a too-limited time frame to be meaningful, and a drawing together of deals and your metrics annually would be fascinating. Or am I missing some place in the baseball blogosphere where this is already being done? In any event, thanks again.
swartzm
11/19
It's an interesting idea. The question is whether you can evaluate trades after the fact. Is the Victor Zambrano for Scott Kazmir trade worse because he became good? Better because he struggled after he was good? It's tough to say. I think evaluating the trades at the time has flaws too-- maybe a team has a beat on a certain player that everyone else is underrating. It's tough to decide how is best to do it. You can certainly look at a GM's aggregate sum of trades retrospectively and decide whether he makes good trades in general, I guess.
Oleoay
11/19
Analysis like that gets further complicated by players like Carlos Gonzalez who was seen as valuable, but didn't play well for his original team (DBacks), nor the team who traded for him then traded him away (A's). Do you dock the DBacks points for giving him up (though for a quality player like Haren), do you credit the A's for obtaining him, but then dock them for trading him away?