Did a Closed Roof Hurt the Blue Jays in Game 5?

Before the Indians clinched the American League Championship Series behind their improbable youngster, there was a mini controversy. Because it was 66 degrees with no chance of rain, there was a movement to keep the roof open at the Rogers Centre. It can get a little stuffy in that park; if the weather was good, why not?

It turns out the why not is in the hands of Major League Baseball in the postseason. The club is consulted, but the final decision goes to MLB. They decided the roof would be shut. It’s natural to wonder, though, after seeing a few long drives fall short of the wall, if those same batted balls would have cleared the outfield fence if the roof were open.

“They’ll leave it closed for the 3rd game. I think generally it’s felt even conditions for everybody” Phyllis Merhige, an MLB official, told Evan Drellich pregame. John Lott, at The Athletic, asked Blue Jays president about the conversation he had with the ruling organization, and the latter replied that it was a “short conversation.” Apparently, shadows may have disrupted the game unfairly and the league wanted “consistent conditions.”

That’s a little weird given the inconsistency of weather patterns from game to game during the rest of the season, but a dome offers baseball a chance to exert some control. They did, and then Edwin Encarnacion hit a drive 348 feet that threatened to pull the Blue Jays within two runs late in the game. Had it gone two feet further, it would have been a home run.

For a piece I wrote on Oakland’s Mt. Davis and its ability to knock down fly balls at The Coliseum, I had the help of baseball’s favorite physicist Alan Nathan. Accounting for weather, launch angle, and exit velocity, he created an expected distance for all balls that traveled over 300 feet during the 2015 season. Because of that work, we can compare the expected and actual distance in Toronto with the roof closed and with the roof open.

Given all the intricate variables, it’s impossible to account for moment-by-moment wind effects, any consideration of wind is omitted from this model. Opening the roof changes wind patterns, obviously. In either case, we see that, over the course of the season, the expected distance of the batted balls in the sample was nearly identical; with the roof open, the actual distance is nearly two feet greater. Based on this analysis, it’s possible that Edwin could have had the extra two feet he needed.

However! Though the spread in temperatures produced by an open versus closed roof isn’t that wide — you only really open the roof if it’s nice outside, and if you close the roof, it’s always 70 degrees — there is a spread. If we compare the roof closed to the roof open with similar conditions to the roof being closed, then we see a larger effect.

Now it really looks like Encarnacion would have walked the parrot, and history might be a little different.

But there’s a bit of a problem with this analysis, too: not only would it not have tied the game, but there were other long drives in the game. Here are the longest batted balls in a game that fell short of homers:

The Napoli and Lindor long flies were doubles, and the Napoli one helped the Indians score, but the rest were outs. If you add five feet (on the high end) to all of these fly balls, you might give the Indians two more runs (on the doubles) while also adding two more runs to the Blue Jays’s total (on Encarnacion’s drive).

In the end, it probably didn’t matter to the outcome. That other universe, where the roof was open, probably has the Indians in the World Series, as well. It’s still fun to wonder.