Say, for instance, we gave Garza 20-20-7-6 instead of 13M per. The first two seasons are basically at market value, but you lose maybe 1M in inflation. The subsequent seasons are closer to 1.75M. So you lose 4.5M over the course of his contract in terms of inflation value. But what you gain in 2016 and 2017 is
A) Potential trade value (low pay for high rewards)
B) Significant advantage in payroll flexibility
C) Maximizing that you are paying the player the most money when they are most likely to contribute.
And here is where I take my biggest issue: It's a bit disingenuous to say it's "understood" that it has to be that way. It's only "understood" to pay more on the back end because that's the environment players have demanded.
As the original poster in this discussion mentioned - the Twins are precisely in Scenario B. (And I would suggest the idea we're the first to ever be in this scenario to be a bit unlikely. It happens, and not rarely) We have a rough idea of when the kids are arriving. We have payroll flexibility now and can anticipate wanting more later. Could we lose a million or two in inflation dollars to front-load Drew? Yes, but I guarantee a savvy GM would see that as minimal risk for the idea of having 5-7M more in payroll flexibility and the extremely high value of perhaps having a pretty good SS making 3-4M when you are looking to add more to your roster. (He has high trade value now and he's giving you huge bang for your buk when you need it most)
Teams like Oakland, St. Louis, Tampa, etc. where they are very savvy and buyout arb. years to extend the life of retaining players could certainly utilize this strategy with some regularity. Yet....it doesn't happen.
And the reason for that is it is a threat to player negotiating power and the union discourages it. I have no doubt that for teams this often works well for them too, but the buck ultimately stops with those with the most to lose and those most in control of the terms. In free agency that's the player, not the team.