Breaking Down The Best Rotations Of All Time
The Detroit Tigers starters currently have a 2.54 FIP, which translates into a FIP- of 62, easily the best in baseball. In fact, it’s easily the best FIP- in baseball history, and as I wrote a month ago, Detroit’s starters have a chance to write themselves into the history books with their 2013 performance. But, instead of just writing a post updating their pace — they’d basically need to post a FIP- of 83 the rest of the way to break the record for best FIP relative to league average — I thought it might be interesting to look at how the best rotations in baseball history dominated.
For instance, the narrative around the Tigers current rotation mostly has to do with their strikeouts. They are on pace to shatter the all time record for strikeouts by a rotation, and the swing-and-miss stuff possessed by guys like Justin Verlander and Max Scherzer makes it easy to credit the strikeout rate as the primary driver of their success. However, once you compare the individual components to the league average, their strikeout rate becomes just a part of the story, and maybe not even the biggest part.
To help illustrate each staff’s strengths and weaknesses, I created three index stats; one each for strikeout rate, walk rate, and home run rate. For each number, I divided the rotation’s performance in that metric by the league average for that league season (so AL matched against AL, NL against NL), giving us BB-, K+, and HR- (minus for BB/HR since lower is better), with each team’s FIP component graded relative to that year’s baseline. Nine teams — including the Tigers and Cardinals this year, since it is easier to perform at an elite level for two months than six months — have gotten a FIP- below 80 from their starting pitchers. Here are how those nine teams stack up by their components relative to league average.
Season | Team | IP | FIP- | K+ | HR- | BB- |
---|---|---|---|---|---|---|
2013 | Tigers | 323 | 62 | 139 | 47 | 79 |
2013 | Cardinals | 338 | 75 | 115 | 47 | 85 |
2011 | Phillies | 1064.2 | 77 | 120 | 73 | 68 |
1971 | White Sox | 1194.1 | 77 | 114 | 73 | 85 |
1996 | Braves | 1026.2 | 78 | 121 | 69 | 87 |
1997 | Braves | 1096.2 | 78 | 110 | 66 | 73 |
2002 | Diamondbacks | 1059.1 | 79 | 137 | 103 | 69 |
1998 | Braves | 1074.2 | 79 | 120 | 67 | 83 |
1970 | Cubs | 1119 | 79 | 113 | 98 | 78 |
The Tigers strikeout rate is amazing, coming at 39 percent better than league average, and is clearly a big part of their dominance. However, the 2002 Diamondbacks posted an adjusted strikeout rate almost as high, and a bunch of deadball teams actually posted K+ rates even higher, so what the Tigers are doing with their strikeouts isn’t historically unprecedented. Instead, it’s the lack of home runs that is really driving their success.
Tigers starters have a combined 0.56 HR/9, less than half of the AL average 1.18 HR/9 for starting pitchers. Likewise, the Cardinals home run rate is even lower (0.45 HR/9), but there are far fewer balls flying out of NL parks right now, so their rate adjusted for league average is equal to Detroit’s so far. And both of these numbers are just absurd, relative to what teams have posted historically.
The last time a team’s starting rotation posted a home run rate that was less than half of the league average? 1937, when the Cincinnati Reds allowed 0.24 home runs per nine innings against a league average of 0.50. In fact, the three lowest HR- rates of all time all belong to the Reds: 1923 (36 HR-), 1925 (41 HR-), and 1937 (48 HR-), thanks to the joy of pitching in Crosley Field. If we went even further and park adjusted these numbers, those Reds performances would rise substantially, as Baseball-Statistics.com notes these fun facts about the Reds old home park:
Here’s a clue as to how difficult it was to hit a home run in old Redland Park: in 1920, the Reds pitchers gave up a grand total of one dinger to opposing hitters in their home park. Back then, the park was symmetrical – it was 360 feet down each line and 420 feet to center. The park was open almost a decade before Pat Duncan hit the first over-the-fence home run, on June 2, 1921. And in 1924, just 6 home runs were hit at Crosley, while the Reds and their opponents smacked 60 in the Reds’ road games.
So, basically, the 2013 Tigers and Cardinals are allowing home runs at a rate only seen from teams that played 75 years ago in perhaps the hardest park in baseball history in which to hit a home run. The strikeouts play a part in home run prevention — it is hard to hit the ball over the fence if you don’t first hit the ball — but other teams have struck out batters at similar rates and not been able to suppress home runs like the Tigers and Cardinals have. It’s the combination of lots of strikeouts and no home runs that explains the Tigers rotation dominance, though their walk rate has been plenty good as well.
But, besides just focusing on the Tigers, it’s interesting to look at how the other great staffs achieved their success. The 2011 Phillies didn’t walk anyone, and have the best BB- of the rotations listed above. The 2002 Diamondbacks just controlled the strike zone entirely, posting the second lowest walk rate and second highest strikeout rate, but they gave up some home runs as part of the trade-off.
Then there’s the 1996 to 1998 Braves rotations. It’s saying something that a team has only gotten a FIP- below 80 from their starters seven times over a full season, and three of those seven came in back-to-back-to-back years from basically the same group of pitchers. There’s a reason those mid-90s Braves staffs are considered the best of all time. Their remarkable dominance over 3,000+ innings is really incredible, and look at the consistency of their home run numbers; 69, 66, 67. The walks and strikeouts bounced around a little bit, but they just kept giving up about 2/3 as many home runs as other starting staffs, even though home run rate has the most noise of the three true outcome stats.
I think it’s probably fair to say that those teams established something like the lower bound of HR- over any sustained period of time for starting pitchers. You didn’t need this post to know that the Tigers and Cardinals home run rates are in for some regression, but I think it’s worth noting just how far removed from something considered normal both teams staffs have been.
The good news for the Tigers is that when the ball starts going over the fence again, they’re still going to have the strikeouts and walks, and barring injury, they’re still going to make a real run at the title of best single season rotation performance in baseball history.
Dave is the Managing Editor of FanGraphs.
Aren’t the Braves underrated by FIP- mainly because Maddux does not get credit for consistently having a lower ERA than FIP? I believe over a 4 year period, his FIP was 1 run higher than his ERA. Is 4 seasons luck or skill?
I think you mean Glavine. In that three year period, Maddux had a negative E-F, but it was only ~0.2 runs or so. Glavine’s E-F for those three years was -0.5, -1, and -1 (approximately). Smoltz had a slightly positive E-F, of the same magnitude of Maddux’ negative E-F.
Never hurts to check the data on the very site that has it at your fingertips.
Yup, was looking at the wrong years, my bad.
I think a partial answer to both of those, at least for that time period, was Andruw Jones. Guy was phenomenal defensively during the late 90’s
I doubt a single player could have made such a large impact, even Andruw.
Well the defense did save 191 runs above average according to Fangraphs during those 3 years — about 60 runs/year, and the ERA- was 76 vs 79 for FIP-. Defense could have been the difference, and Andruw saved about 20 runs/year above average so he might have been about 1/3 of the difference.
I’m surprised no one has mentioned the not infrequently generous strike zone afforded Atlanta pitching…
The things I would do to have Pitch F/X backdated…
Wouldn’t Smoltz’s FIP most likely be lower than his ERA relative to Glavine and Maddux since he relied on defense less (a good Atlanta defense, supposedly).
Or the difference between their FIP and ERA I mean
For his career, Maddux had an ERA- of 76 and a FIP- of 77. He had some seasons where it was higher, some where it was lower.
Ah, I was looking at his Cy Young years of 92-95, not 1996-1998, my bad.
“Aren’t the Braves underrated by FIP-”
Yes, probably. If you look at something like ERA+, the ’97 Braves starters blow away what the Tigers are doing this year. In fact, the ’13 Tigers haven’t been historically great by ERA+ (the Rangers starters have been better this year).
I get all the reasons why Dave used FIP, but the title of this post should come with an asterisk (*By Fielding Independent Stats).
Using ERA to evaluate a pitching staff that is supported by Miguel Cabrera and Prince Fielder (among others) is not a good idea.
Yes, I get that. But assuming you can parse out the greatest rotations of all-time using only FIP is also probably not a good idea.
Well sure nothing is a perfect idea, but this doesn’t seem to be a bad idea.
Well, Dave’s results should at least be taken with a medium to medium-large sized grain of salt. Keeping with the ’97 Braves example, using B-R WAR which accounts for defense and park, the Braves rotation produced 4.50 WAR per 200 innings, the Tigers currently are at 4.06 WAR. So “a real run at the title of best single season rotation performance in baseball history”. Eh… I’m dubious.
“I get all the reasons why Dave used FIP, but the title of this post should come with an asterisk (*By Fielding Independent Stats).”
This is a good point. Though it would make for a less interesting title.
Even more underrated by this metric than the Braves were the 1871 White Stockings who had an ERA- of 57 and a FIP- of 83. George Zettlein knew how to pitch to the defense.