On Monday, reader "Kreylix" left this comment:
My understanding is that winning close games is mostly luck. Winning by 3+ is mostly skill.
It would be cool to see the MLB standings for just games decided by 3+ runs.
In that case, consider this post the coolest. Here are the 10 teams with the best records in games decided by three or more runs and the 10 teams with the best records in games decided by one run:
|
Team |
3+ Run Record |
3+ Run WP |
Team |
1-Run Record |
1-Run WP |
|
Nationals |
43-17 |
.717 |
Orioles |
23-6 |
.793 |
|
Cardinals |
48-22 |
.686 |
Indians |
15-8 |
.652 |
|
Yankees |
45-27 |
.625 |
Braves |
17-11 |
.607 |
|
White Sox |
38-23 |
.623 |
Nationals |
24-17 |
.585 |
|
Reds |
37-23 |
.617 |
Giants |
24-17 |
.585 |
|
Rays |
38-24 |
.613 |
Athletics |
19-15 |
.559 |
|
Rangers |
44-28 |
.611 |
Royals |
19-15 |
.559 |
|
Braves |
41-32 |
.562 |
Pirates |
25-20 |
.556 |
|
Diamondbacks |
41-32 |
.562 |
Reds |
21-17 |
.553 |
|
Tigers |
32-26 |
.552 |
White Sox |
18-15 |
.545 |
The most obvious difference is where the Orioles place on these two top-10 lists: at the top of one, and not at all on the other. The O's have a .387 winning percentage in games decided by three or more runs. That's the 25th-worst record in baseball, worse than those of the Phillies, the Mets, the Padres, the Royals, the Blue Jays, and a few other teams that aren't anywhere close to contention. Good teams like the Yankees, Cardinals, Rangers, and Rays, on the other hand, show up on the three-plus-run leaderboard but don't make the one-run top 10.
Clearly, Kreylix had the right idea. The correlation between overall record and record in games decided by three or more runs is .86. The correlation between overall record and record in games decided by one run is .40. The difference in correlation strength is partly due to how many more three-run games there are than one-run games, but luck plays a large role. Is it mostly ​luck? I'd prefer to put it like Matt Kory did in his response to Kreylix: "I think that scoring runs and preventing runs are skills. But the closer a specific score gets the more random variation can affect the outcome."
Here's what the playoff picture looks like now, along with how it would look if the standings were determined by record in games decided by three or more runs and games decided by one run.
|
Division |
Overall |
3+Run |
1-Run |
|
AL East |
Yankees |
Yankees |
Orioles |
|
AL Central |
White Sox |
White Sox |
Indians |
|
AL West |
Rangers |
Rangers |
Athletics |
|
AL WC1 |
Rays |
Rays |
Royals |
|
AL WC2 |
Orioles |
Tigers |
White Sox |
|
NL East |
Nationals |
Nationals |
Braves |
|
NL Central |
Reds |
Cardinals |
Pirates |
|
NL West |
Giants |
Diamondbacks |
Giants |
|
NL WC1 |
Braves |
Reds |
Nationals |
|
NL WC2 |
Pirates |
Braves |
Reds |
If the standings were determined by record in one-run games, six of today's 10 playoff teams would still make the playoffs, but only the Giants would make it via the same route (winning the NL West). We'd also have three AL Central teams in the playoffs, and the Tigers wouldn't be one of them. I wouldn't want to live in a one-run world. The current standings are unpredictable enough, as the 2012 Orioles remind us.
​Thanks to Rob McQuown for research assistance.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
We're starting to see some writing in Baltimore criticising "statheads" for emphasizing the O's run differential, and saying, similar to the "Keylix" post, that the O's record is luck.
The key fact in this article is the correlation difference between one run wins and overall record, and three+ run wins and overall record.
I'm also aware, from reading BP, that Pythagorean record is a better predictor of future record than current record. This was the basis of the recent Jay Jaffe SI article stating that the O's are likely to fade.
Unlike some of the commentators in Baltimore, I don't see pointing to the O's run differential as belittling their accomplishments thus far this season. Frankly, the O's record in this run differential environment is even more remarkable than the improvement from last year.
The reason for the record in one run games (and extra inning games), I believe, is clearly the performance of the bullpen. For that I credit Duquette's (and MacPhail's) acquisitions and Showalter's management of the bullpen, as well, of course, as the credit due to the pitchers themselves.
This is the most fun I've had watching the O's in years.
I despise it when some people use advance metrics in an attempt to negate what actually happened, as if their virtual statistical world is more meaningful than the real world.
In particular to deny an award to a pitcher or player because one disputes that they didn't "deserve" such impressive statistics is inane.
It is the same as assigning the World Championship to a team which was bounced out of the playoffs, because on paper it "should" have won.
But with awards season soon upon us, we will see several faux "stat heads" do exactly that. And usually with an air of superiority and a sneering condescension towards anyone who disagrees with them.