rtallia posted this question in "Hey, Bill". I machine-gun-typed an answer, but then I liked the answer enough that I decided to expand the answer and post it here, as an article:
A question on your "relievers throughout history" article. Since the top reliever workload is now exactly half of what it was 30-40 years ago, presumably those next 70 innings are now being pitched by each team's SECOND best reliever. The question is...is that 2nd reliever, who gets a bunch of "Holds" to his credit and maybe 4-6 saves throughout the season, almost as good (at this point) as the closer himself? Do you think that the combination of the 1 +2 revlievers, especially over the past 5 years, is equal in effectiveness to that 140 innings from the top reliever from 40 years ago?
My answer, somewhat edited:
I don't have a good understanding of that, nor, I think, does anyone else.
In the evolution of strategy, when the managers/executives by groupthink focus on an advantage within a strategy, they are working blind to the fact that there is also a disadvantage within that strategy. Eventually the strategy will expand usage until the point is reached at which the disadvantage becomes self-evident. But to calculate the costs and benefits of a strategy is in almost all cases such a complex mathematical problem as to be almost impossible to resolve.
For every strategy, there is a point in its expansion at which the costs of strategy will outweigh the benefits. Stealing bases, for example. If you only have one base stealer and he can steal 80% of the time, no doubt it is a good strategy. If you start stealing bases with your second-best base stealer, your third-best, your fourth-best. ....
If you start stealing bases in the optimal base-running situation. . .let's say two out, singles hitter at the plate. . ..if you start stealing bases in that situation, OK, but then you start stealing in the less-optimal situation, the next less-optimal, until eventually you are trying to steal bases down 4 runs in the 9th inning. Which I have seen, I think; I have seen players do that.
If you start stealing bases when you have a pitcher on the mound who is slow to the plate and a catcher who doesn't have a great arm, that may be a great strategy. The pitcher is slow to the plate and the catcher has a "D" arm; GO. But then you start stealing when the catcher has a C arm, and when the catcher has a B arm, and when the catcher has an A arm. As you expand the use of the strategy, you move into the area in which it is no longer optimal strategy.
Let us say that a leadoff man who has a .350 on base percentage but can steal 100 bases a year is more valuable than a leadoff man who has a .360 on base percentage but can't get out of his own way. But as you expand the strategy, the on-base percentage of the players you choose to put on your roster gets lower and lower, until eventually you have Omar Moreno in your lineup every day. The costs of the strategy expand as its usage expands.
What I am trying to get to. . .EVERY strategy is bound in this way not by a single cost, but rather, every strategy is surrounded on all sides by a variety of costs, all of which increase as the usage of the strategy increases. It is impossible to calculate what all of these costs are at each point. Baseball people see the benefit in a strategy and adopt it and use it more and more and more, until it becomes apparent that its cost are outweighing its benefits. But no one ever knows what that point is until it becomes apparent; therefore, strategy changes endlessly, always shifting, always trying to find the balance points at which each strategy is at maximum value in comparison to other strategies.
And this, of course, is not unique to baseball or to sports. Every business, every endeavor at all points in time is relying on strategies that have expanded to the point at which they have become self-destructive. Television commercials. Television stations rely on commercial revenue, so they try to sell more commercials and more commercials and more commercials. But as the strategy expands, the costs of the strategy expand. You put enough commercials in the middle of a sitcom, people start watching baseball games; you put enough commercials in the middle of a baseball game, people start watching soccer, or they go out bird-watching or mountain-climbing or kayaking or something. Then somebody starts something like Netflix; they can make sitcoms just as good as the TV station can, collect a few dollars a month from a vast number of people, and all of a sudden the network which relies on selling commercials finds its revenue plummeting. It works until it doesn’t work anymore.
Or negative advertising/opposition research in political campaigning. The assholes who promote this stuff have studies that prove that it works. They can prove the benefits—but the costs are enormous. You start broadcasting negative stuff about your political opponents; eventually people start to believe that it is true. They start to believe that is true about both sides. After a while you have Senate confirmation hearings in which every Senator makes an ass of himself trying to throw mud the other Senators. Eventually you have riots in the streets, with the people who believe that the Republicans are venal rioting against the strongholds of the Republicans, and the people who believe that the Democrats are venal rioting against the strongholds of the Democrats.
It is the same thing; it is strategies expanding until the costs of them become much higher than the benefits. We are surrounded by this on all sides; the undocumented, unmeasured costs of old strategies wash up against the sides of our house. This is why societies need value systems. You need a value system that says "attacking your political opponents is wrong" which is stronger than the research of political operatives which says "NEGATIVE ADVERTISING WORKS."
In baseball, too, you have to be guided by Values, rather than strictly by math and strategies. Many of you will remember that I have written many times about Red Rolfe in Detroit in 1951-1952 platooning Vic Wertz. This fascinates me because it is one of the clearest examples of this principle that I am trying to explain here. Casey Stengel’s brilliant use of platooning made platooning a hot strategy. Detroit manager Red Rolfe started platooning Vic Wertz, who had driven in 133 and 123 runs the two previous seasons. It pissed off Wertz, destroyed Rolfe’s managerial career, and the Tigers went from 95 wins to 104 losses in two years. VALUES. You don’t believe in the strategy more than you believe in the player. You don’t believe in the commercial revenue more than you believe in producing a good television show.