The State of Parity Today

“Tanking” has become a word people are applying to baseball, with various agendas being pushed in advance of the CBA negotiations. No matter where you stand on the merits of the argument, it’s at least a curiosity that now, perhaps more than ever, almost everybody seems to have a chance. It’s easier than it’s ever been to build a roster capable of making the playoffs, and the landscape feels like it’s mostly in balance. You probably already knew these things. If you’re a big fan of looking at stuff you already know, wait’ll you see what I have in store for you here!

Ben Lindbergh wrote an excellent article at Grantland last March about the broad trend toward greater parity within the game. There’s not very much for me to add, because in classic Lindbergh fashion, he went into incredible detail, following every thread. You should read that article from start to finish, and all I’m really here to do now is add a little more data. As I noted earlier today, this is projection season, so it’s of interest to see what the projections are saying. They’re saying what you might’ve already been able to guess about the American and the National Leagues.

What we have here right now are projections based on Steamer. Within a few weeks we’ll also fold in ZiPS to make it a 50/50 blend, and ZiPS and Steamer don’t always see eye-to-eye, but today we’ll use what’s available. Okay, now, think about the leagues. How do you expect them to shake out? You probably think of the AL as being pretty even. Meanwhile, you probably figure the NL more or less groups teams into two tiers. Maybe three. Great — let’s visualize that, also including preseason projected information going back to 2005.

I split by league, and then for each season, I calculated the standard deviation, in terms of wins, for preseason projected records. This works as a measure of parity, and though the projection systems differ over the years, all projection systems try to do the same thing so it shouldn’t be that big of a deal. Here, look at the National League, first. A dozen years of preseason standard deviations:

This year’s spread looks as big as it’s been in the window, with the 2016 figure tying the 2005 figure. You could argue there’s parity at the top of the league, I suppose, but the league itself is clearly split. Steamer’s six worst projected teams are all in the NL. Then you get the Diamondbacks as being okay, and the Marlins as being okay, and then you move into the higher tier. Every single tanking accusation — those teams are all in the same league, and the Brewers and Reds aren’t finished selling. The Padres and Rockies have yet to truly begin. The NL, in other words, has obvious winners and losers.

The American League is the complete opposite:

Look at how stable that used to be. Look at the more recent slope. If this were a mountain that you were coming down, you’d have to be careful, because with one slip you might end up dead. At the time, 2014 had the lowest observed standard deviation. That was eclipsed in 2015, and it’s been eclipsed again in 2016, at least as far as Steamer is concerned. The league is incredibly well balanced, with no dominant teams and no awful teams, and if you consider the Orioles and/or the White Sox as a little below-average, those are teams that might still add Yovani Gallardo, Dexter Fowler, and/or Ian Desmond. Everybody in the AL could conceivably be a playoff team. Everybody in the AL could conceivably end up a cellar-dweller. There are still better and worse teams, relatively speaking, but this is extreme, as parity goes. Steamer sees just an eight-win difference between the AL’s worst team and the AL’s second-best team.

Maybe you’re wondering about the outcomes, instead of just the projections. Obviously I can’t do anything about 2016, but here’s the NL over 11 years:

There was decent observed parity between 2005 – 2007, but things have been more lopsided since then, 2014 blip aside. As for the AL:

It holds well with the earlier plot. Which is funny — on a team-by-team sense, the 2015 AL projections went terribly. There was actually a point where you would’ve been better off believing the opposite of the preseason projections. But the projections still nailed the parity, forecasting a pretty even league, which is basically how things played out. There was more real-life spread than projected spread, but that’ll almost always be the case, just because of how these things work.

Steamer isn’t a perfect system, and the numbers will change some when we get ZiPS information included, but I don’t think there’s much room here for disagreement. The NL is clearly tiered, whereas the AL is more like a narrow-ranged gradient. The NL feels kind of like how baseball used to be. The AL feels more like what we should get used to. There will always be teams with more resources, but with the playoffs within easier reach, teams should be less incentivized to bottom out, and I think the NL teams doing that are mostly correcting for previous mistakes. We shouldn’t see too many teams tear down, in theory, once the current repairs have been made. It could be a future of parity, across both of the leagues.

That’s my own speculation; I don’t know for sure where baseball is going to go. But there are two things about parity: (1) everyone feels like they have a chance, and (2) there’s a greater role of luck in determining outcomes. That’s just how it is when there isn’t much spread in talent, and that can be considered either a good thing or a bad thing. You’ve heard before about how you can’t predict baseball. With greater parity, it’s only going to get even harder.