If the Rangers beat the O's, don't call it an upset

The 101-win Orioles, the top seed in the American League, head into Tuesday’s ALDS Game 3 down 2-0 to the 90-win Rangers, the fifth seed. Unless Baltimore can manage to come back and win three consecutive games, which has happened just 11% of the time when teams drop the first two of a best-of-five, the top seed will go home.

You can imagine what will happen next. Questions will be raised about whether the layoff the Orioles had while waiting out the Wild Card Series hurt their chances (probably not), or about how they choked away their best chance to go all the way in decades. It will be considered an "upset" in the books.

But is it, really?

If we just look at the seeds, yes. (One is higher than five.) If we just look at the number of team wins, yes. (One hundred and one is more than ninety.) If we look at the head-to-head record this year, then _no_; Baltimore took the first two of three in Texas in April, and Texas took the first two of three in Baltimore in May, so it was an even 3-3. If we look at run differential, then also no, because the Orioles outscored their opponents by 129 runs, while the Rangers outscored theirs by a superior 165 runs.

It’s indisputable that the Orioles won 11 more games than the Rangers did, but the flaw in all of this – in all the data presented so far – is clear, in that all of it is based on a full season of play. While April wins matter just as much as September ones in playoff seeding, many of the players who contributed to those numbers aren’t here right now. Jacob deGrom, Max Scherzer, Félix Bautista, John Means, Shintaro Fujinami, and Ezequiel Durán aren’t on these rosters. Evan Carter, Jordan Montgomery, Brian Baker, Aaron Hicks, and Heston Kjerstad are. The uniforms are the same. Some of the players are not.

Even the players who are the same may be different – The Rangers’ Dane Dunning had a sparkling first half that declined into a just-OK second half, and Baltimore’s Grayson Rodriguez had a very rough early-season debut that saw him sent back to Triple-A, before he returned to become one of the best starters in baseball in the second half.

Maybe, then, we should be focusing on the players who are actually participating, and how they’re playing now, and so maybe the better question is this: Is there that big of a gap between the October Orioles and the October Rangers? And if not, how unsettling should it be that a team that sent six players to the All-Star Game might actually (gasp!) win two consecutive games?

There’s no one right way to answer this, though FanGraphs hosts a pair of playoff odds projections that, if we go back to see what they said on Friday, Oct. 6, before the series began, should help us to …

• Standard playoff odds: Rangers 55/45
• ZiPS playoff odds: Orioles 56/44

… sort nothing out. The differences there are ones of methodology (the ZiPS version uses more information about projected starting pitchers, for example), but even though they pick different teams, they end up on slightly different sides of “coin flip.” Baseball Prospectus, for their part, also gave Texas a very slight edge here.

What if we instead look at A) the players on the ALDS rosters and B) what they did in the final month of the season?

The hitters

• Rangers: .783 OPS
• Orioles: .728 OPS

The pitchers

• Rangers: .710 OPS against
• Orioles: .666 OPS against

This is an admittedly imperfect way to look at rosters – surely the Rangers did not expect Nathan Eovaldi’s 1.028 OPS September disaster to continue into the playoffs, and it did not against the Rays – but what you can glean from this, comparing players only on the ALDS roster, is that Orioles pitchers were 44 points better in September, and Rangers hitters were 55 points better. So where exactly is the large edge here?

Maybe it's better to try to figure out how the losses actually happened.

In Game 1, a mere 3-2 Texas win, the Orioles fell because veteran outfielder Aaron Hicks missed a sign on a busted hit-and-run, and future Hall of Fame manager Bruce Bochy masterfully handled a tandem start of Andrew Heaney ahead of Dunning. In Game 2, they lost because the rookie Rodriguez, who had been so strong for months, was unwilling or unable to throw breaking balls against one of baseball’s best fastball-hitting teams and simply couldn’t throw strikes – nor could Bryan Baker, who had been in the minors for nearly all of the second half, nor Jack Flaherty, whose unsuccessful midseason acquisition stands in stark contrast to Texas adding Montgomery.

The Orioles lost because Gunnar Henderson and Adley Rutschman are a combined 4-for-18 (plus two walks); they’ve lost because their pitchers have walked 16 Rangers batters in two games; they’ve lost because a player who hadn’t been with Texas for the first five months of the season – Carter – is off to one of the best starts in playoff history. They lost because this is the third time these clubs have faced off this year, and it’s the third time that the visiting team has won the first two games, and because they’re two evenly matched teams, and because it’s two games between two good playoff teams.

As MLB.com’s Thomas Harrigan researched, going back to 2012, the team with the superior regular season record went 48-39 in postseason series that were more than a mere one-game playoff. (Six were excluded, as the two teams had the same record.)

That record, 48-39, also works out to 55/45 proportionately. If not exactly a coin flip, it’s about as close as you’ll get.

It is, to some extent, similar to the discussion had last year when the lower-seeded Phillies and Padres shocked higher seeds on their way to the NLCS, because the emphasis on their relatively paltry win totals (under 90 for each) ignored just how different their rosters were _right then_, at that moment, about the improvements that brought them closer to the Braves and Dodgers, respectively.

It’s a conversation that’s playing out again in the NL, when the lower-ranked Phillies have a 1-0 lead on the higher-ranked Braves, another outcome that would play out as an ‘upset’ when maybe it really isn’t, given Atlanta’s well-discussed pitching injuries and Philadelphia’s late-season bullpen improvement. Can it really be an upset when just before the series began, six of the 11 MLB.com writers polled said that yes, the Phillies would in fact come out on top?

The Orioles have lost so far because the Rangers played better baseball over a sample of exactly two games. It shouldn’t be surprising, given the similarities in talent between these teams – and it wouldn’t be surprising if the Orioles managed to win three in a row, either, something they did on 15 different occasions in 2023. It's not really an upset. It's two similarly good teams over 18 innings of baseball. They have, after all, exactly 19 hits apiece.