Large Free Agent Position Player Contracts: Updated and Comprehensive

I believe this response has points more pertaining to the article generally than to this comment- so I will respond on that basis and please forgive if I’m misrepresenting what you were trying to say.

First point – I’m not sure exactly what you mean (is there a word missing?) – I’m having trouble with “the effects of year contract is given are not necessarily a wash”. Are you commenting on a specific graphs, or on the methodology generally – because I’m not sure I assumed anthing was a wash? Salary inflation is something teams have to think about when giving contracts. In fact, I wonder if they don’t have some sort of recency bias – when revenues growth rises for a couple of years, salary inflation seems to take off, the contracts go nuts, and look really dumb when revenue growth returns to trend. Likewise, when it slows down, the contracts are more reasonable – Tejada and Vladdy signed reasonable contract around 2004-05, when it appeared that revenue growth was moderating, though it quickly returned to trend. If it’s about one of the graphs, the whole point was to basically show that there isn’t really an observed relationship between those selcted variables. They were included for completeness, since I had done the same thing in the pitching article, and I thought there conceivably could be some logical link.

Second point – Is this referring to the above graph? Otherwise, I’m not sure what the bias is, generally. Yes, younger players are almost all better, so when they hit free agency, all else equal, we expect they get more. But the point is, we’re just looking at what the market deemed to be premium players according to what the players were awarded, and how those contracts performed. In theory, an efficient market should account for age and skills such that players were properly paid in the context of these factors. The purpose wasn’t to compare players of equal skill (say, who each put up 25 WAR in their 6 years of control, and then look at the difference in contract – then age would definitely matter). Guys who reach the bigs are less likely to get megacontracts for those reasons – first, they’re not generally as good, and second even if they suddenly put it together and became awesome, they’d only hit free agency well after their peak so their future expected production is smaller.

Third point – I removed the word overpaid from the above – it was a mindless word choice, and I missed it reading over. Throughout, I’ve taken pains to avoid using the word overpay, for precisely one reason you point out – the signing teams are likely probably higher on the win curve and it’s basically impossible to objective say they did or did not overpay. We don’t their marginal revenue curves, etc and even if we did there’s too variables to fairly treat a contract. So my bad for sure on that.

This grew out of a conversation about Prince Fielder, SuckaMD put together a list of players who signed contract over $100M and looked at the production, I initially was only going to formalize it, then I expanded the list a bit, and then began thinking about comparing to players who signed extensions covered FA years. And I think that last point is the key one – the comparison. In aggregate, the extension contracts are much better for teams – why? Maybe it’s a selection bias whereby players who sign extensions give a hometeam discount, but on the other hand most of them are close to free agency and so they the leverage of that to still get a good price. And as a group, they end up near fair value according to prevailing free agent $/WAR (which Fangraphs estimates for each using the actual contracts). It could be that the team knows mroe about the players, and tey select which ones they want to keep, but I find that more persuasive for pitchers, where arm problems can tank a contract – and the difference in value is much larger. But most of those players left for purely financial reasons. You introduced one I hadn’t really thought about – moving from teams with lower marginal win values to higher marginal win values, but players will probably only be extended if the marginal win value is pretty high (for example, Wells and the Jays was right then they were a mid-80s win team, which is the sweet spot for the marginal value of a win). So I doubt that explains much. I think most of the difference is due to the winner’s curse of auction bidding, and that’s the major takeaway. It is better value to extend your own stars than to get another team’s. Once I finish the second part, I want to look at some of the implications of this for team building.

All this said (and apologies for going on this long), I do not agree with a few of the observations you made. First, The evidence suggest at this point is that $/WAR is linear, whether for 2, 4 or 6 WAR players. I’ll admit I found this inconceivable, but on the other hand, the better players get paid a lot more and there’s a ton of idiosyncratic risk (you can get an Albert Belle you has a degenerative injury and has to retire). So youo have to insure them, which is part of the cost to the team, and takes away from what you can pay them (it’s essentially a wedge between what is paid and earned). Second, I do not agree that having a larger payroll means you pay more $/WAR. It means you buy more wins in the free agent market, but it doesn;t mean you should pay more than anyone else for a given win (though the effect of more teams wanting to buy more wins will drvie the market to higher equilbrium and force smaller budgets to buy fewer wins, which means they have to rely on their system for more wins, or is the system doesn;t produce, then to suck).

Thanks for the feedback, and sorry if I’ve misinterpreted anything