One More Incredible Rangers Statistic

As you’re no doubt aware, it’s been a hell of a season in Texas. The Rangers own the best record in the American League, with the eighth-best run differential. They’ve destroyed their Pythagorean record, which has caused them to destroy their BaseRuns record. Much of this has been fueled by historic success in one-run games, and much of that has been fueled by historically clutch hitting. Teams are successful every year, but the Rangers have followed an unusual course. It’s been simultaneously thrilling and bizarre, something difficult for analysts to explain. At this point, there might not be any sense in trying.

There’s one more nugget I want to throw on top of the others. For the reasons detailed above, this Rangers season has been truly exceptional. It’s hard to imagine a team drawing it up like this. Yet there’s another split you might have trouble believing. I know I did! Which is why I’m writing this in the first place. I don’t really know what it means, but I can’t not bring it to your attention.

Baseball-Reference draws a handy line. Let’s say you wanted to split teams into two groups, by success. How would you do it? The easiest thing would probably be dividing at the .500 mark. Baseball-Reference does that, and while I definitely understand this is far from perfect, it’s neat and it’s useful and it makes enough sense. Teams over .500 are pretty good. Teams under .500 are mediocre. This year, when playing teams .500 or better, opponents have won just 45% of the time. When playing teams under .500, opponents have won 56% of the time. I can’t imagine I need to explain this.

If you’ve followed the Rangers, you know they’ve dominated the Astros in head-to-head competition. It’s gone a long ways toward shaping the AL West race. It’s possible to go one step further, though. When playing teams who are at least .500, the Rangers have gone 60-31, which is very easily the best record in baseball. The Dodgers are a distant 31-23. The Rangers are just running away with that category.

And yet, almost inexplicably, when playing teams who are below .500, the Rangers have gone 34-34. That’s tied for baseball’s 23rd-best record, even with the Rays, and sandwiched between the Brewers and the Braves. Against quality opponents, the Rangers have played like a total juggernaut. Against weaker opponents, they’ve been indistinguishable from those organizations that are rebuilding.

Not only is this backward. This is extremely backward, and you can see that in the following plot. This looks at every team going all the way back to 1913, which was made easily possible by the Baseball-Reference Play Index. The Rangers are clearly highlighted.


Of course there’s noise, or randomness, and of course circumstances can be different when a team plays a certain other one. Maybe the Rangers have gotten really lucky in terms of missing the better teams’ better starters. But we haven’t seen something quite like this in a century. In a little more than a century, for that matter.

Split Winning Percentages, 1913 – 2016
Team Year Win% vs. .500+ G Win% vs. < .500 G Difference
Rangers 2016 0.659 91 0.500 68 0.159
Cardinals 2010 0.617 60 0.480 102 0.137
Reds 1932 0.427 110 0.295 45 0.132
Giants 1979 0.500 84 0.372 78 0.128
Twins 1998 0.493 75 0.379 87 0.114
Pirates 1918 0.585 54 0.472 72 0.113
Browns 1924 0.545 66 0.442 87 0.103
Angels 1974 0.454 109 0.352 54 0.102
Mets 1980 0.451 102 0.350 60 0.101
Twins 1973 0.556 72 0.456 90 0.100
SOURCE: Baseball-Reference

Almost 2,300 team-seasons in the sample. For each of them, I calculated winning percentage against teams who’re at least .500, and winning percentage against teams below .500. I then found the difference between those two winning percentages. This year’s Rangers have the biggest positive difference, and from what I can tell, it would be almost impossible for them to fall out of first place here over the season’s final games. It wouldn’t matter if they swept the Rays. They’d still be No. 1 in one of the most unusual splits you could come up with.

So add this to the list. Clutch hitting. One-run-game winning. Run-differential defeating. And, better against good teams than bad ones. It doesn’t have to make sense for it to be, and there’s no undoing what the Rangers have accomplished. All told, they’ve had one of the weirdest damn seasons in at least recent memory. It still stands to be mostly defined by what happens in the weeks ahead, but Rangers fans could tell you this year has never been dull.

I’m sure you’re wondering if this means anything. Like, in terms of looking ahead. It seems like it would reflect well on the team if it “plays up” against stronger opponents. The counter-argument would be that it reflects poorly on the team if it plays down to weaker opponents. But as it happens, conveniently, I took a little look at this line of thought a year ago, when the Mets did the opposite of what the Rangers have done. The Mets beat up on softies and were relatively weak against quality clubs. I couldn’t find any signal in my quick investigation, and of course the Mets made it to the World Series. I don’t think this is much of an indicator in any direction. There are too many variables, and too many points in conflict.

But, hell, what’s true for a group doesn’t have to be true for an individual. Maybe the Rangers somehow do play better against better teams. There’s no real harm in believing it, from the perspective of a fan or a player. I don’t know the ins and outs of that dynamic. And no matter what, this whole year has just been extraordinary. For a variety of reasons, the 2016 Texas Rangers haven’t made great sense. I’m going to guess that’s made them all the more enjoyable to cheer for.

Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

Newest Most Voted
Inline Feedbacks
View all comments
David Palardymember
6 years ago

Interesting. What team season is the dot all the way on the right?

Jackie T.
6 years ago
Reply to  David Palardy

I’m also curious about the extreme outliers on each side of the cluster. Left, down and right seem like pretty unique extremes in their own right.

6 years ago
Reply to  David Palardy

Probably some Yankee team. Or the 116 win M’s

6 years ago
Reply to  Doorknob11

My guess is the 1954 Cleveland Indians. They went something like 89-11 against sub-.500 teams.

EDIT: Checked just now and remembered just how weird that 1954 AL was. The standings were:

CLE 111-43
NYY 103-51
CHW 94-60
BOS 69-85
DET 68-86
WSH 66-88
BAL 54-100
PHA 51-103

Against the five teams with losing records, Cleveland went 89-21. They were exactly .500 (22-22) against the Yankees and White Sox. If you count the World Series, where they were swept by the Giants, Cleveland was played .458 ball against winning teams.

6 years ago
Reply to  Doorknob11

I think I found the team. Strangely, it’s not any of the great regular season teams that we think of, like the 1900s Cubs, several Yankees teams, or the 2001 Mariners.

It’s the 1928 St. Louis Cardinals, who won the National League pennant with a 95-59 record, finishing two games ahead of the Giants and four ahead of the Cubs. The Cardinals managed to go 38-6 (.864) against losing teams and 57-53 (.518) against winning teams—or .500 if you include their World Series sweep at the hands of Ruth, Gehrig, and the Yankees.

The Cards weren’t the only NL squad feasting on losing teams. The Pirates had a spread nearly as large at St. Louis, going .794 against losers and .468 against winners.

What the hell happened that year? Well, a whole bunch of weirdness did:

STL 95-59
NYG 93-61
CHC 91-63
PIT 85-67
CIN 78-74
BRO 77-76
BSN 50-103
PHI 43-109

The Braves and Phillies were so terrible that they made the other six teams look much better by comparison. How terrible were these two teams? Consider this:

Boston finished 53 games under .500 with the best player in the National League. Player-manager Rogers Hornsby had 9.0 WAR in 140 games. He did so on the strength of a 196 wRC+, which was a close second for the MLB lead behind Babe Ruth at 200.

The Braves also got a useful season from an aging George Sisler (2.4 WAR), as well as a career year from a journeyman RF named Lance Richbourg (4.1 WAR).

The 15.5 WAR put up Hornsby, Sisler, and Richbourg made them about as valuable a trio as the three best players on the NL champions. Jim Bottomley, Chick Hafey, and Frankie Frisch posted a combined 15.8 WAR for St. Louis.

Boston’s fortunes diverged from those of the Cardinals due to the trivial matter of the rest of the Braves team being worth -4 WAR.

Somehow, the Phillies managed to finish seven games worse.

While Boston’s unique losing season featured an upstairs-downstairs dynamic that would have put “Downton Abbey” to shame, Philadelphia followed through on an impressive commitment to uniformly horrible performance. Chuck Klein and Don Hurst tied for the team “lead” in WAR despite their best efforts. Their tie at 2.1 WAR has to be some kind of modern record for the least valuable “best” player on a team.

If Boston and Philadelphia were even a little bit better, then the Reds and Brooklyn Robins would have had losing records as well.

Returning to the weirdness in St. Louis, moving Cincinnati and Brooklyn to the losing team bin would change the Cardinals’ losing/winning team split to 61-27 (.693) and 34-32 (.515). They go from being the outlier in the table above to placing very close to the imaginary regression line going through the data.