THE BOOK--Playing The Percentages In Baseball

Have you seen Joe Sheehan’s take of the CBA?  He devoted an issue of his newsletter to it; his overall take was not positive.  He sees the restrictions on signing money for draft picks and for international signings as ways of narrowing the range of options for teams.  And he predicts (this will be interesting to watch) increasing emphasis (spending) on coaching and on conditioning/medical staffs.

Thanks for the link. I agree that the new CBA is more equitable in terms of who it impacts, but I was hoping for a decrease in the number of compensation picks overall, and it doesn’t seem like the MLBPA actually achieved that despite all the talk surrounding the issue.

Matt, oh I agree, it would have been fantastic to have a true free agent, like in most walks of life, and all the other sports.  But, the MLBPA doesn’t care about that. 

The MLBPA has simply always treated compensation rules as something between the teams.  Papelbon / Madson is a prime example, as they simply see the Phillies giving up one draft pick but gaining another.  And that teams wouldn’t try to arbitrage it.

Even look at minimum salary level.  The NHL has less than half the revenue of MLB, and a roster size that is very similar to MLB (23 in NHL, and 20 play each game, against 25 in MLB, and 10-15 play each game).

Yet the NHL minimum salary is higher than MLB.  The reality is that the minimum salary in MLB should be pushing 1MM$ (or the NHL min salary should be under 250K).  Again, MLBPA reasons that any money saved on the low-end will simply transfer to the free agents, because teams spend to a budget, rather than doing a cost/benefit analysis on a transaction-by-transaction basis.

(I know you may disagree with me on that last part.  Eventually, I’m going to break you and get you on my side!  It’s a shame one of us is wrong for a few years on this, and I refuse to believe you can be more stubborn than I am on this issue.)

Well, I think that the fact the $/WAR spent on players who cost picks is less than the $/WAR spent on players who did not, and that the difference reflects an approximate 10% discount rate gives some credibility to what I’ve been saying-- I think teams are paying players less because of the added cost.

I actually don’t quite think “budgeting” is an inaccurate way to describe what’s going on, but I think you’re thinking about it from a GM-level, rather than a collective team level. I believe owners give GMs budgets, and they do so based on a general cost-benefit analysis, which is largely based on their perceptions of their position on the marginal-value-of-a-win curve plus their understanding of their local markets.

Their “perceptions” of their position becomes their crutch.  The Royals giving out long-term deals for 5MM$ a win self-presumes they are not true 70-win teams, even if they are.  The Rays, A’s, and Twins refusing to give out 5MM$ a win deals to free agents even if they are true 85-win teams.  Or, their evaluation that their market suggesting they can’t support 5MM$ a win deals.

The reality is that the Twins offered Mauer what they did because they had an influx of cash, and they didn’t offer it to Santana and Hunter, even if they all deserved it.

***

As to your original point: I agree that teams ARE arbitraging.  My point was simply that MLBPA has never thought about the arbitrage, or if they did, they didn’t care, because the money would just flow to other players anyway.

On a fundamental level, you’d be hard-pressed to make the case that teams are actually getting 5MM$ a win of benefit, if the average team has to pay for 30-35 wins each year.  A pure free agent team (no player development) would have to pay over 200MM$ a year, for them to be in the sweetspot of talents (40-45 WAR).

So, there’s a huge inefficiency among free agents.

"Thanks for the link. I agree that the new CBA is more equitable in terms of who it impacts, but I was hoping for a decrease in the number of compensation picks overall, and it doesn’t seem like the MLBPA actually achieved that despite all the talk surrounding the issue.”

At first I thought they would greatly decrease free agents costing compensation, but you’re probably right.  I’m looking at the small number of type A’s this year once the relievers are removed.  But I need to consider the current type B’s as well who would get 1 year/ 12 million dollar offers so a team can get picks.

For example, Cubs exercised their 16 million option on Aramis Ramirez, which he then voided.  Obviously under the new system he gets the 1/12 offer.  Plus just about anyone that a team will offer 3-4 years and at least 30 million to, they’d also be willing to accept on a 1 year, 12 million dollar plan.

It will be interesting to see how it affects the trade market.  Say the New York Knights want to add Roy Hobbs, batting .350 with 45 homers through July, to their 200 million dollar team.  Hobbs is the only bright star for the last place Cleveland Spiders.  He’s a free agent in 2 months. 

In the old system the Knights would give up a good prospect package, partly for the 2 month boost and partly for the compensation package.  But in the new system you have to have a player the full year to get a compensation package.  The Knights will reduce their offer, and the Spiders might well just keep Hobbs, prefering the compensation package to middling propects.

It could well push non-contending teams to move quickly on dealing guys like Hobbs, they will get much better prospect offers in March than they will in July, since the Knights will get the compensation if they get Hobbs early enough.

But then again, very few teams will count themselves out as early as March, especially with the extra wild card.

I agree that any good player who warrants a 3+ year deal will get offered a 1/12 deal. 

Right, I agree about the trade market.  No longer will good guys get traded with under a year left in free agency. 

Take, I dunno, Cole Hamels.  He’s coming into his walk year.  How much will a team give up for him for two months, knowing they will get zero picks back?  It’s a huge drop-off for the Phillies, if they trade him one day before the season starts compared to one day after.

Tango/5:
The issue of competitive and non-competitive teams paying the same price per win doesn’t really say anything other way. They are bidding in the same environment, so they are paying for a smaller quantity of FA wins but at the same price, rather than paying for the same quantity at a lower price.

The Twins influx of cash is because they built a stadium which increased their marginal revenue in the future, not their total revenue in the past. The contract for Mauer was signed within a month before the new stadium opened, so it was clearly an anticipation of future revenue. I believe that to be an anticipation of high marginal revenue; you believe that to be an anticipation of high total revenue.

Rally/6:
Yes, I agree with everything you said. Lower surplus from trade => fewer trades. Lower value for contending team => lower prospect price offered by contending team. Discontinuity in value on March 31/April 1 => more deals earlier on.

That’s probably valuable for the league overall, if you think about it. People hate players switching teams. It creates a lot of bitterness to see it happen mid-season.

Matt: I still don’t see why a team like the Royals would pay for Guillen or Meche or any other .450 team paying market rate for free agents.

If they really needed a body, they could take that money, and trade it to a team for a half-decent prospect.

Say a team has a prospect expected to generate 10 wins over his pre-FA years.  Supposedly that generates 50MM$ of value, and it’ll cost them 15MM$ in salary.  Why wouldn’t the Royals send 35MM$ and acquire this prospect?  Heck, if a team discounts future wins at 10%, then they could send them 25MM$ or 30MM$ instead.

But if teams could really sell off their half-decent prospects (remember, just 10 wins over six+ years) for 25MM$ a pop, they’d be lining up to do so.

But they don’t!  Why is that?  Because a win isn’t REALLY worth 5MM$.

Other than with free agents, no one actually behaves as if a win is worth 5MM$.

Now, suppose that a win was ACTUALLY worth 3MM$.  Now, that 10 win guy over six+ years is worth 30MM$, and will cost 15MM$ in cost-controlled pre-free agency.  His net value is 15MM$.  And discounting future years, say 10MM$ today.

Isn’t that more plausible?  Isn’t it more likely that a team can sell a half-decent prospect for 10MM$ than it could for 30MM$?

Remember, we’re just talking about a guy expecting to get 10 wins over 6+ years.  I’m not talking about Bryce Harper or Gerrit Cole.

Teams trade prospects as though they worth $30MM though. That they don’t actually sell them for cash outright is irrelevant. They don’t trade veterans with good deals for cash outright, either.

The average win is not worth $5MM. But that’s not what I’m saying. The average purcashed win IS worth $5MM. The team that signs a free agent is the one with the highest perceived value for that player. They have the combination of highest spot on the marginal-revenue-of-a-win curve, highest value for player at that position, largest market, etc.

If what you were saying was correct, then teams that had large changes in spending & winning at the same time would get massively less profitable as they spent/won more and massively more profitable as they spent/won less. Do the Phillies and Tigers lose a lot of money nowadays as they have risen on the payroll rankings? Do the Mets and Indians suddenly make a lot more as they have fallen on the payroll rankings? If the price of a win was so massively off, then that’s what we’d be observing. I’m not seeing that. There are plenty of ways to justify the value of a win being that high if you consider that the team who just purchased it outbid 29 other teams for that win and that future ticket sales are driven heavily by previous season’s competitiveness.

Teams trade prospects as though they worth $30MM though. That they don’t actually sell them for cash outright is irrelevant. They don’t trade veterans with good deals for cash outright, either.

Your first line is based on the free agent market, isn’t it?  Otherwise, give me an example.

As for the third line: there aren’t any deals like that!  For that to happen, you’d have to have a veteran player who increased his value in the middle of the contract.

***

As for your Phillies/Tigers: that’s a good way to look at it, but I’d have to see the data.  Obviously, we have to look at the whole enterprise value (including prospects, etc), and all 30 teams.

Intriguing, and I’d like to see that data.

The point I was trying to make was that when we do the prospect-for-veteran trade analyses, the math generally works out about right when you treat wins for both the prospects and the veterans as being worth ~$5MM or so.

You’re basically making an argument that teams are overpaying for wins across the board, but you don’t really have any evidence for a very strong statement. My evidence is based on the profitability of teams, and that cheap teams don’t seem to make much money, even after revenue-sharing. The null hypothesis has to be that teams spend on labor based on its marginal revenue, like any other business. There are peculiarities to baseball, for sure, but you need real evidence to reject the null hypothesis, I think.

The math will make sense regardless of whatever number you use, with respect to veterans for prospects. 

I’m not saying they are overpaying across the board.  I’m saying they are overpaying for free agents.

But yes, a team that has a disproportionate number of free agents will make less money than a team that simply ignores free agents.  (All other things equal, obviously.) So, yes, that’s a good way to show it.

The Yankees obviously have not “other things equal”. But, Phillies, Tigers, Dodgers, etc and other big players in the FA market, yes, those are good teams to look at.  If you can find other teams in mid-to-large market that bypasses free agents, then we’d have a good comparison group.

It’s hard to find an apples-to-apples comparison, and you’d need to also find the revenue sharing redistribution numbers because those are lump-sum after-the-fact transfers you’d need to subtract out, and then you’d need to look at a team over a period of years. So basically you’d need to find teams from big markets that don’t spend over a period of years to see what happens, and then contrast it with teams that do. Or you could look at the profit of teams that changed their minds about spending, like the Phillies. I think the Phillies are a perfect example of a team that went from low-spending to high-spending and hasn’t been losing money. The Tigers also don’t seem to be losing much money for changing their ways. The Indians don’t seem to be making much money for changing their ways either. I’d need actual numbers to make a better case, but teams that shift their payrolls relative to league average by $40MM or somthing like that should be raking in huge profits if you’re right.

When one of the guys you are spending big on is perhaps the best player in the league, and who signed away his FA years (Chase Utley), that’s a huge leg up already.  Then you have perhaps the best pitcher of his generation who has never signed a FA market-level contract (Roy Halladay).  Add in Carlos Ruiz who basically had his fate sealed in Philadelphia, the incredibly inefficient contract given to Placido Polanco, and you have a team well-positioned to earn money.

Exactly who did the Phillies sign to an FA deal?  Cliff Lee?  Is there anyone else?

How that team also signed Ryan Howard to that horrible contract is hard to fathom.  (And, given that that deal doesn’t take effect for another year, it doesn’t even count for what we’re talking about.)

Raul Ibanez, Brad Lidge.

Yeah, that pretty much shows we’re not talking about alot here.  We have one year of Cliff Lee to evaluate (and, yes, he would be someone who is overpaid based on what I am talking about).

We also have them NOT signing Werth.  Basically, trading Werth for Cliff Lee.

And same for Ibanez, no?  Wasn’t that the same year they let Burrell go?

Tigers sound more like what we are talking about.

There are plenty of Phillies that were FA-eligible, though a lot of them were bought out FA years. I’ve covered that I think pretty extensively in the OPP stuff in this thread: http://www.insidethebook.com/ee/index.php/site/article/other_peoples_players/

The Phillies were a great example of a team that did better re-signing their own guys, but in terms of major players with 6+ years of service time, it’s an old team. In the recent spending spree years:

1B- Thome’s 6-year deal
2B- Utley (bought out FA years)
SS- Rollins (bought out FA years)
3B- 4-year deal for Bell, mini 2-year deals for Helms & Nunez, 2-year deal for Feliz, 3-year for Polanco
LF- Burrell (bought out FA years), 3 year-deal for Ibanez
CF- Victorino (bought out 1 FA year)
RF- Abreu (bought out FA years), 2-year deal for Jenkins, Werth (bought out 1 FA year)

SP- 3-year deal Eaton, two 2-year deals for Moyer, traded for last 1.5 years of Oswalt, 3-year for Halladay, 5-year for Lee, 3-year for Jon Lieber, bought out a FA year of Myers, bought out 2 years of Blanton

RP- 3-year deal for Gordon, bought out a FA year of Madson, 3-year deal for Lidge, and like a whole bunch of 2-year deals for various relievers that I’m struggling to remember.

They’re spending like $170MM now, up more than double from several years ago. A lot of that is on FA or FA-eligible players with six years service time.

And they certainly have made a lot of money from me at least, haha.

Thanks for the list. 

The Thome one is a perfect example of what we’re talking about.

Brand-new player added, so that the expectation is increasing payroll, adding player, to get more wins.

Signing Utley is NOT that.  The expectation of wins stays the same.  We’re not even adding players.

Lee for Werth is also NOT that.  Expectation for wins stays roughly the same for roughly the same cost.

Polanco is a good example too, but, man he signed a really low deal.

Doc Halladay is a bit of a weird one, because he is signed for a fair price, but it cost them prospects.

So, Thome and his ilk is what we are really talking about.

Well, I don’t really agree that you should exclude extensions. That’s my other point: re-signing your own free agents is a good deal and other people’s players are a bad deal. You can’t turn around and tell me to argue against that! The Utley thing is a good point. He’s the kind of player that the Phillies used to let go once he hit free agency. He was Schilling/Rolen, largely home-grown or home-blossomed superstar, except they ponied up and brought him back instead.

But even without that, you still have Thome, Bell, Polanco, Ibanez, Lieber, Gordon, Eaton, Halladay, Lee, Papelbon, and a ton of two-year deals. So they have plenty of OPP and plenty of extensions all together payroll quadrupled from $41MM in 2001 to $166 in 2011.

I’m not arguing about OPP.

We were talking about something else: that free agents are way overpaid per win.  And you had the insight to look at those teams that spent big on free agents.  If they lost money, that means they overpaid.  If they made money, that means they underpaid.

Luck notwithstanding.

So, I love the idea, and now we just need data.