Tuesday, November 06, 2007

"Homegrown players" -- a viable strategy?

In 2007, the four teams in the league championship series had 187 of their wins (as measured by Win Shares) contributed by players "homegrown" by the respective team. That's up 68% from last year, and 43% more than the "recent average."

These numbers come from a recent Wall Street Journal article by Russell Adams, "
Baseball Promotes From Within."

Adams doesn't make an explicit argument, but the implication is that teams are focusing on player development, rather than on signing free agents (who are getting very expensive), and that the strategy is working.

I'd argue that the strategy is not so much to concentrate on homegrown players, but perhaps to concentrate on *cheaper* players. After all, if you have a star in his "slave" years earning only $380,000, it doesn't matter whether he came from your farm system or someone else's. Either way, he's going to help you win equally.

The Indians, Diamondbacks and Rockies were all in the bottom eight
payrolls for 2007. They won because their low-priced players performed well, not necessarily because their homegrown players performed well.

However, there is an argument that homegrown players are a better investment:

"Executives say promoting your own players makes sense not only because they are familiar, but because everyone in the organization knows how they've been trained. Instructors in the Phillies' farm system, for instance, follow a manual that describes the "Phillies' way" of doing everything from warming up a pitcher's arm to defending a bunt. Promoting from within is "a safer way to go," says the team's assistant general manager Mike Arbuckle."
Even if you don't accept that the "Phillies' way" is better than the "Brewers' way" or the "White Sox' way," it's still possible that bringing the player up yourself can benefit the team. You'd expect that the team that knows the player best would be the best judge of his major-league expectation. By watching the player carefully, perhaps the Phillies can avoid the mistake of bringing a player up too early. But if they got the guy in trade from the Astros, they might not know enough to make a proper judgment. (I don't know of any evidence either way.)

But still, there's nothing to stop other teams, even free-spending ones, from also developing homegrown players. Even high-spending teams have a budget. If the Red Sox, for instance, find a gem in their minor-league system, they can trade away the expensive free agent at his position, and use the money for someone else.

For teams with little money to spend on salaries, there is an obvious strategy, one that's also used in Rotisserie. You trade your expensive players for young minor-league talent. Eventually the acquired players are ready for the big leagues, and you get three years of free service out of them (and a couple of still reasonably-priced arbitration years). If that's what these teams are doing, then, again, it's not the "homegrown" factor at work – it's the "cheap" factor.

Labels: ,


At Tuesday, November 06, 2007 10:54:00 PM, Blogger Brian Burke said...

This strikes me as very similar to conclusions of the Massey-Thaler study about the NFL. Their paper estimated the equivalent cost of a draft pick and free-agent veteran of equal expected performance levels. They found that the draft pick is a lot more cost efficient relative to the free-agent, leaving more cap room for additional players or greater depth.

I wouldn't be surprised if it were the same case in MLB. It basically boils down to the idea that free-agents are overvalued. My guess is that GMs don't understand that a guy who hit .260, .250, .260, then .290 is probably not going to hit .290 next year.

At Tuesday, November 06, 2007 11:55:00 PM, Blogger Phil Birnbaum said...

Aren't draft picks a lot cheaper in MLB than in the NFL?

In football, the players are pretty much ready for the NFL right away (they'd better be, since there are no minor leagues). But in baseball, drafted players need time in the minors to develop. That would explain the difference, if indeed there is one.

So that would explain why slaves are so cheap compared to free agents. Even if free agents were correctly valued, that would be the case.

At Wednesday, November 07, 2007 9:29:00 AM, Anonymous Anonymous said...

Dave Studeman has done some nice work showing that pre-arb players are producing a larger share of player value: http://www.hardballtimes.com/main/article/2007-net-win-shares-value/.

Basically, in 2006 pre-arb players contributed 32% of total value, vs. 44% from FAs; but in 2007 each group contributed 39%. That's quite a shift in one year. So it's not just the top 4 teams that have more home-grown talent, it's all of MLB. But it could also be that playoff teams have had a disproportionate gain in young talent. Probably need more than one year to figure that out.

At Wednesday, November 07, 2007 9:42:00 AM, Anonymous Anonymous said...

It seems the more diversity used to build a team the better. It's true that the Red Sox do not have an infinite budget, but it does seem that their budget exceeds the current market for free-agency (Drew and Lugo for $100 million and it was actually worth it?!?!?). The limiting factor is that their are only so many free-agents each year, so they have to look elsewhere: draft-picks, draft-picks, Japenese imports

In comparing MLB with NFL, also remember the number of roster spots and playing opportunities. A 21-year-old outfielder can't help out on special teams no matter how close he is to being ready for the Majors

At Wednesday, November 07, 2007 9:43:00 AM, Anonymous Anonymous said...

that should have been draft-picks, trades, Japanense imports. . .

At Saturday, November 10, 2007 2:02:00 PM, Blogger Brian Burke said...

I didn't intend to say that baseball and football draft values and systems are equivalent. They are different animals. The point I was making was that FAs are overvalued relative to the the same expected performance from rookies. The Massey-Thaler study suggested NFL teams can get the most bang for their buck from the draft (particularly at the end of the 1st round and beginning of the 2nd) rather than from FAs. The 'Moneyball' A's pretty much demonstrated the same effect in baseball.

At Saturday, November 10, 2007 8:52:00 PM, Blogger Phil Birnbaum said...

Oh, okay, sorry. What you're saying sounds very plausible. And you'd think the effect would be stronger in baseball than football, since in baseball, it takes longer for players to develop, so what you think you see today may not be what you get tomorrow.

At Monday, November 12, 2007 4:05:00 PM, Blogger Bob Timmermann said...

Not all NFL draft picks are ready to go. Only a few positions have players who can make an impact right away and these are usually running backs and receivers. Quarterbacks in the NFL take a couple of years to develop for the most part (unless they're named Dan Marino).

The offenses and defenses in the NFL are so much more complex than college football that it takes quite a while for players on the offensive line to become effective.

At Monday, November 12, 2007 5:25:00 PM, Blogger Phil Birnbaum said...

Hi, Bob,

But the drafted players are on the roster and in uniform, right? And they will eventually play after learning the system.

So isn't it reasonable to assume that the level of ability is fairly accurately known for football draftees? That is, they don't need any more playing time to reach the NFL level abilitywise.

At Wednesday, November 14, 2007 6:33:00 PM, Blogger Bob Timmermann said...

I've always thought that projecting which players will do well in the NFL after playing college football is one of the dark arts.

Some QBs in the NFL are guys I never thought would be any good. And some guys still keep playing.

Some players seem to learn NFL systems faster than others, which must account for the difference. But with the injury rates in the NFL, you don't have time to waste with some players, especially running backs.


Post a Comment

<< Home