Monday, October 25, 2010

Game 4: The Efficiency of the Buffalo

‘“No” is, generally speaking, a better answer than “Yes.”…’

The real hero of Moneyball is the city of Oakland. Billy Beane has the leading role, of course. He’s The Man Who Brought The Enlightenment to Baseball, and he gets all the glory. He will be played by Brad Pitt in the film, whereas the city of Oakland will be played by the city of Oakland, which is a bit of a step down glamour-wise. If there were justice in the film world, the city of Oakland would be played by the city of Paris- the Angelina Jolie of locations. That’s how important Oakland is to Moneyball. Without Oakland, and it’s unfortunately small market, and it’s necessarily low payroll, Mr. Beane might have languished, unappreciated, managing some wealthy and successful team.

The first pages of the righteous book begin with a simple observation: The Oakland Athletics are more efficient than every other baseball team. This efficiency can be measured in marginal dollars per win. The A’s had been paying about $500,000 per extra win, whereas other teams were paying multi-millions for each additional victory. If this were any other business, the apostle Lewis observes, the Athletics would dominate the market. They would own baseball. Such is the power, in the business world, of efficiency.

The problem is that sports management is not, by its nature, particularly concerned with efficiency. Unlike real businesses, sports teams don’t get anything for payroll money they have and don’t spend. For the sock manufacturers of the world, every dollar saved in sock-production can be reapplied to better equipment, more factories, better advertising, or Thai massages for stockholders, all of which in turn will grow the sock business, which in turn makes more money, that can then be reapplied again, until such heady heights of sock production are achieved that the sock tycoon can lie back by his enormous pool with a large-breasted blonde who’s entire wardrobe is made of extra cash that has been reprocessed into textiles and assembled into haute couture fashions by brilliant, difficult Japanese men with prickly hair and colorful shoes. Any normal business, money saved is money earned, because money is the only currency that matters.

Sports, however, measure value in wins, not cash. All other things being equal, winning is the end all and be all, and money, like the bodies of players, is a tool to achieve winning. At the end of a season, money unspent is worthless. Saving does not make you better.

In order for efficiency to matter in sports, all other things must be unequal. There must be an artificial limit on the money that can be spent. There must be a small market town or a tightwad owner or a salary cap. Otherwise, efficiency is just a neat trick. Like being able to finish the Saturday New York Times crossword in pen, it demonstrates intelligence, but it doesn’t provide any particular advantages in the world. Something external- like the city of Oakland, or the city of Buffalo, or the CBA- has to make money into a finite resource before efficiency becomes interesting.

Given that marginal efficiency is mentioned on page XIII of Moneyball- a page so early it doesn’t even have an Arabic numeral- it’s curious that it has never really been a big thing in hockey stats. The NHL, after all, has a salary cap, which makes efficiency about fifty times more precious in hockey than it is in baseball. Recently, however, The Copper and Blue, SB Nation’s Oilers site, has been hosting a running discussion of marginal cap efficiency as a tool for judging the effectiveness of NHL GMs. The posts are a bit scattered around the site, but you can find the core of the discussion here, here, here, and here.

I’ll just give you a minute to get caught up.

… [twiddle]…

Finished? Great.

Most of what the efficiency rankings indicate is a little bit obvious. Good teams are good, bad teams are bad, and if you’re paying $3.75 million to an alcoholic goalie within spitting distance of middle age, you’re going to look a little silly. The obvious parts stem from the fact that the rankings are driven heavily by points, such that any team that does exceptionally well in a given year is deemed efficient, even those who’ve overflowed the cap and have backup goalies making $5 million. Similarly, the teams at the bottom are overwhelmingly the teams anyone, no matter how drunk and bereft of charts they might be, would guess: teams that make bad bets, carry big contracts and lose anyway. Most of the teams in the NHL cluster in the middle, with results driven by the standings fluctuations from season to season. Montreal’s salary philosophy, if there even is such a thing, hasn’t varied at all since the lockout, but the efficiency hops up and down around the average in sync with their finishes.

But the project does highlight a few Beaneish abberations: Buffalo and Nashville, the teams that do well on a small budget. The truly efficient franchises. Conversely, if you look at the last post, on Marginal Floor Efficiency, we see the equivalently inefficient, the teams that pay a premium for what success they get: Boston, Philly.

Gut reaction: If I am a Flyers-affiliated analyst, fan, or hockey-strategist, does this matter to me? My team is not efficient. Since the lockout, it has not had hardly one single efficient season. Shitty seasons, terrific seasons, all passed in such an orgy of cash-flinging that even the trainers find stray Benjamins clinging to the bottom of their washing machines when they launder their track suits. But does that mean the management decisions have been equally terrible for the past five years? Nope. The team has improved, and generally speaking, the money goes to the useful pieces of the team, if perhaps a bit more emphatically than necessary.

Ah, you say, but Buffalo does almost as well for much less money. If Philadelphia only had their management sophistication, they would be only a little less successful, but they’d have cap room left over to sign a few big pieces that might put them over the top. Spent to the cap as they are now, they’ve done pretty good, but with Buffalo-style savvy, the might be great. Be cheap until the great deal comes along, my sons, that is the way.

But it’s harder to be efficient at the top end. For most hockey teams, it’s not the ridiculous overpayment to an incompetent or minor player that kills the efficiency ratios, it’s the pursuit of the great deal. At the higher end of talent, players have a better sense of what they’re worth, market distortions and all, and are inclined to demand things, either in salary or term, that are far from efficient. The big contracts are the ones that inevitably leave teams wedged against the cap. They’re almost impossible to outplay, and they involve an element of gambling. Any one or two or three players in a year might get hurt or have an off season, any given year might be the one where pain and aging start to force an inexorable collapse in performance. There are a hundred ways any contract in hockey can defy expectations, and at least seventy-three of those can make an otherwise reasonable decision into a ridiculously inefficient one.

Everyone is looking for moneypuck, but hockey has no equivalent yet of what OPS was in the glory days of Beane. Hockey stats, to this point, can reveal the nuances of an offense and evaluate hitherto mysterious facets of defensive play, they can highlight fine-grained differences between players that vision alone misses, and they can raise red flags about guys whose results are likely to be illusory or insubstantial. But there’s no philosopher’s stone stat, nothing that can help you find bottom-of-the-barrel talent and turn it into first line. Hockey offense remains stubbornly obvious, and while individual managers might occasionally be stupid or superstitious or ideological in making certain signings, there’s no League-wide mythology that’s leaving NHL talent languishing in the AHL.

The pursuit of efficiency in hockey creates a culture of no. It’s a deeply conservative worldview, where when in doubt, the best thing to do is nothing at all. If something must be done, it should be the smallest, safest option. Don’t offer a deal that can’t be outplayed. Don’t buy at the deadline. Don’t shop the UFA market. Don’t trade draft picks. The don’ts run several fathoms deeper than the dos, and ultimately offer no very useful recommendations for team building other than the classic suck-hard-for-high-draft-picks method, because baby contracts are very nearly the only way to get elite talent at a bargain price.

There are only so many guys in the NHL at a given time who can provide their particular skills. At the bottom end, there are a lot of guys with similar surface value playing for comparatively cheap, so research and conservatism can be well-rewarded. It’s easy to find a guy with a good underlying skill set who can give three million in value on a two million contract, and chances are, that guy will be available. Move up the ladder, though, and it gets harder. There are only so many guys who can justify a six million dollar contract, and a lot of them area already locked up somewhere else. Those who are available are likely to know what they’re worth and demand exactly that, or even more. An available player who can provide six-seven-eight million in production is a rare creature. One who can do it cost-effectively is a unicorn. You might as well fire your scouting staff and hire a crew of hot virgins to wander the forests of Northern Quebec.

The no method is a long-term waiting game, with extra money on the books, and that’s fine for a team in a rebuilding phase, or one that needs the extra money to survive in a tough market, but if a team feels it can compete and has the disposable income to do it, then there’s no reason to let efficiency guide decision-making. That doesn’t mean making every deal you can, no matter how desperate or improbable. It doesn’t mean throwing cash around like a lottery winner at a strip club. Any GM with any amount of cash should still be doing the homework necessary to make good bets based on substantial evidence. But it means sometimes making the inefficient deal in a real-world market where the perfect thing isn’t always available at the perfect price. It means sometimes giving the extra year or the extra million. The privilege of wealth is being able to overpay to get the guys you need to improve, even if it scuttles the efficiency ratio. A good GM is one who manages contracts elegantly, not just cheaply. If there is a number out there that can accurately reflect the quality of a GM, Marginal Cap/Floor Efficiency isn’t it.


Jeff J said...

Great post. A team-level $/marginal win number loses a lot of its usefulness when your 40th win is worth a shitload more than your 35th.

A couple of points about your observation of this culture of no: 1) I'm not so sure that it is generally true and 2) where it is true, it is specific to the cases that get the most publicity.

Every decision *not* to do something has, at the very least, a positive default option (unless you're Lou Lamoreillo you still have to field a full roster). The default is to fill spots from within or plug holes cheaply. This is not necessarily equivalent to 'suck-hard-for-high-draft-picks.'

"Don’t offer a deal that can’t be outplayed." The decision not to re-sign Michael Ryder opened up a roster spot for Tanguay or Latendresse. The Hamrlik deal may have been a mistake, but not as big as a Souray deal might have been.

"Don't buy at the deadline." I read this one as "*Do* sell at the deadline." Maybe I'm an optimist. See the Rivet deal - one of Gainey's best moves.

"Don't shop the UFA market." I read this as 'Don't shop the UFA market for star players.' As you mention, 'There are only so many guys who can justify a six million dollar contract.' A good GM will come to terms with this fact and fill the spot with a stopgap that allows flexibility until the justifiable $6M contract comes along. A bad one will throw $6M & term at Briere or Ribeiro and hope for the best.

I think it makes sense to shop the UFA market to flesh out a roster. Good role players are always available: Moore, Metropolit, Gill, Halpern.

"Don't trade draft picks." I'm not sure this one would be backed by most of the statzis. If it were you could phrase it as a positive: "Acquire draft picks." In fact, many of them seem to think picks are overvalued unless you have a shot at an early 1st.

Rather than a 'culture of no,' I might call it a culture of 'less wrong.' But like I said, maybe I'm an optimist.

E said...

you have a point. there are probably positive corollaries to a lot of the nos, but i still think the realm of yes, in hockey-moneyball, is a lot less explored than the realm of no. people have lots of great criticisms of moves that are made, but you have to look a lot longer and deeper to find suggestions for better ones, other than 'wait and hope for something perfect.'

ironically, my pet example of a good-but-inefficient move is one i was totally wrong about at the time: gomez. when gainey took him from the rangers, i was positively enraged, exactly because his contract seems like such a waste of cap space. and it's true, he'd could play a hundred seasons and never outplay that deal once. but guy is a player, an extremely useful player, and a far better one for his role than the canadiens have ever been able to get on the free market. his deal is the dysfunction that makes him available, and the canadiens are a team with the ability to overlook that. if gainey had actually negotiated that deal, i'd probably still hate him, but the compromise of taking it to get the player... it was a good choice. but not an efficient one.