Astros' rotation making case as best ever

It doesn't take much effort to make a case that the Astros have the best rotation in the Major Leagues in 2018. It's actually harder to find reasons to suggest that they don't. The Astros have a 2.25 ERA, by far the best in the Majors. They have the best strikeout rate (30.1 percent), the lowest average against (.189), the lowest slugging against (.316), and they've done it all while throwing the most innings (308). They've been so good that their .188/.253/.316 line against basically means that every hitter against them turns into Michael A. Taylor, who's hitting .185/.255/.305.
Even better, think about it this way: The entire Astros rotation is striking hitters out like Luis Severino (30.8 percent), allowing the same average as Chris Sale (.188) and limiting slugging like Max Scherzer (.319). That is, of course, ridiculous.
Houston has three legitimate American League Cy Young Award candidates in Justin Verlander, Gerrit Cole and Charlie Morton, who all live at the top of the AL ERA leaderboards. The Astros also have Dallas Keuchel, who won the 2015 AL Cy Young Award. Houstonn also has Lance McCullers Jr., who just shut out Cleveland (baseball's hottest offense over the past 30 days) for seven innings in Sunday's 3-1 win. The Astros are so good and deep that Brad Peacock, who had a 3.22 ERA in his 21 starts in 2017, has been relegated to the bullpen. He can't crack this rotation.

This isn't a surprise. The Astros were expected to have the best rotation, and they do. Will that be all they are? We're nearly 30 percent of the way through the season. It's not too soon to see if they can stack up with the best rotations of all time.
The first question to ask is, how in the world do you even measure that? How do you compare a rotation in the baseball world of 2018 to ones we saw in the 1990s or the '60s or the '40s?
It's not easy, but it's not impossible. Let's try this out a few different ways. Let's start with the absolute simplest one possible.
Through 48 games, Houston's starters have allowed 83 runs. The fewest runs allowed by a rotation in the modern era (since 1920, non-strike years) are 342, by the 1967 White Sox, who pitched right in the middle of the historically low-offense late-'60s. If the Astros were to maintain this pace -- easier said than done, of course, and perhaps impossible -- they would allow about 300 runs. They'd shatter the record.
Now, there's some caveats here, of course. Raw runs scored totals can be affected by defense, and starters simply don't pitch as many innings as they used to. Last year, no rotation threw 1,000 innings. In that 1967 season, 15 teams (out of only 20) did. 
So give this one the grain of salt it deserves, but even then, it's impressive. The job of any pitcher is to prevent runs. This Astros crew could allow fewer than any modern rotation.

ERA helps us put earned runs on a "per nine innings" basis, which helps with the issue above, but isn't perfect for a few reasons, namely that the sport's offensive environment does not stay consistent. In 1968's "Year of the Pitcher," the Major League ERA was 2.98. At the height of the high-offense era in 2000, it was 4.40. That means that depending on the year, a 3.50 ERA could either be very good or very poor.
We'll get to that. Let's start simple, though. Which rotations since the modern era began in 1920 have had the best ERAs from their starting pitchers?
Best rotation ERAs from 1920-present (non-strike years)
2.25 -- 2018 Astros
2.49 -- 1968 Cardinals
2.50 -- 1968 Indians
2.52 -- 1943 Cardinals
2.58 -- 1972 Orioles
With another caveat that we'll get to in a second, that's the top of a very impressive list. The 1968 Cardinals featured Bob Gibson's magical 1.12 ERA year and a 23-year-old Steve Carlton; the '68 Indians had stellar years from Luis Tiant and Sam McDowell. You might not remember the '43 Cardinals, who had the benefit of war-weakened competition; you probably do remember the early '70s Orioles of Jim Palmer.
The Astros top them all. One problem: Comparing 48 games to a full season. That said, no team this century has had a lower ERA through 48 team games. No team since the 1972 Dodgers has gotten off to a better start (for this look, we're talking full staffs, not just starters, though in the past those were largely the same thing). With the understand that there's more season ahead than behind, this is meaningful.

Remember when we said that raw ERA is fine, but it doesn't tell you enough if you don't know what run-scoring was like that year? A better way is to look at a pitching version of OPS+ (called ERA-), where 100 is "league average" for that year and each point above or below 100 represents a percent above or below that league average.
Let's do exactly that. Compared to the average for that year, which teams have had the best ERA marks above average?
Best adjusted rotation ERAs from 1920-present (non-strike years)
45 percent above average -- 2018 Astros (2.25 vs. 4.12 MLB)
29 percent above average -- 2016 Cubs
27 percent above average -- six teams including 1997 and '98 Braves
That's an enormous gap, and it's some impressive competition. You might remember that the 2016 Cubs won the World Series, and they did so on the strength of an elite starting rotation (and a great defense) that performed like one of the best ever. The 1997 and '98 Braves had Greg Maddux, Tom Glavine and John Smoltz. This has the same partial-season issue as above, but the point remains: What these Astros are doing is rarely ever seen.
All this talk about ERA ignores the issue about present-day starters throwing fewer innings, and a low ERA is more impressive over more innings. (Although, again, no rotation has thrown more innings in 2018 than Houston's.) 
That being the case, we can turn to Wins Above Replacement, which does -- unlike ERA -- account for how many innings you pitch. If we look at FanGraphs and put every rotation since 1920 on a 162-game scale, you'll see where the 2018 Astros would hypothetically end up.
Best rotation WAR, FanGraphs from 1920-present (non-strike years)
30.2 (projected) -- 2018 Astros
26.0 -- 2011 Phillies
25.9 -- 1971 White Sox
25.4 -- 1997 Braves
24.7 -- 1967 Twins
There's some new teams on this list, but it's a fun one too, with the Astros topping the 2011 Phillies of Roy Halladay, Cliff Lee, Cole Hamels and Roy Oswalt. There's those 1997 Braves again. The Astros are on track to beat this record by 16 percent. That's a ton.
Pretty much all of this is subject to the fact that we can't simply take what has happened and expect it will happen for the rest of the year; "on pace" doesn't work this way. Mookie Betts is on a 162-game pace to hit 58 homers. The Yankees are on pace to win 112 games. Those things probably are not going to happen. For the Astros, someone will get hurt or slump. Peacock, Francis Martes David Paulino,or someone else will have to enter and make some starts. Who knows how they'll treat September if they have a large divisional lead.
Tons of things could go wrong, is the point. But we've also seen more than enough to know that this isn't some kind of fluke. The 2018 Astros rotation is the best in baseball, by a lot. (Their bullpen has the AL's lowest ERA, too. It doesn't get easier.) At the end of the year, we may be talking about them as one of the best ever.