Move over, Houston: The Astros are no longer the most surprising team in baseball. That distinction now belongs to the American League Central-leading Twins, a team that opened the season by becoming the first in major league history to fail to score an earned run in its first three games. Minnesota opened 0–3, then 1–6, but since falling to 6–10 on April 24, the Twins have gone 24–9, good for a .727 winning percentage, the best mark in baseball over that stretch and six games better than any other team in their division. So how has a team that lost more than 90 games in each of the last four seasons managed to open June with the second-best record in the AL at 30–19? The answer, unsurprisingly, is by overachieving in likely unsustainable ways.
The clearest indication that the Twins are playing over their heads is the tremendous gap between their actual and third-order records. Third-order record goes a step beyond Pythagorean record to determine a team’s expected runs scored and allowed based on the component parts of run scoring (hits, walks, etc.), then calculates their expected record based on those expected run totals. The Astros have won three more games than their actual run differential would suggest, but their third-order record (31–21 compared to their actual 32–20 record) supports their performance as legitimate. In sharp contrast to that, the Twins have also outplayed their actual run differential by about three games, but their third-order record suggests that they are much further over their heads than that.
• POWER RANKINGS: Twins jump into top 10, Cardinals grab first place
Minnesota’s third-order record is 22–27, a full eight games worse than their actual record. No other team in baseball has outplayed their third-order record by a full five wins (the Braves, another team that clearly appears to be overachieving, is at 4.7 wins above their third-order mark). The size of that difference is a clear indication that correction is coming for the Twins, but how have they managed to win so much without putting together the underlying performance to support that success?
There are a few factors at play here. To begin with, the Twins have had a fairly easy schedule to this point. Of the six other AL teams entering Tuesday’s action with winning records, Minnesota has played just two of them, the Royals and Tigers, and gone just 5–10 against those two. The weighted winning percentage of the other eight teams the Twins have faced: .461. Of those teams, only the Pirates—whom the Twins, to their credit, swept in a two-game series in Pittsburgh two weeks ago—currently boast a winning record.
Still, even if the Twins had been beating up on weaker teams, their performance should support their results, and it doesn’t. Their 11–6 record in one-run games, which is identical to the Astros’ mark, can explain why they have out-performed their run differential by three games. There is likely to be some correction there both because of the nature of one-run records, which trend toward .500, and because of the team’s two end-game relievers. Veteran righty Blaine Boyer—who has emerged as the Twins' primary setup man since signing a minor league deal with Minnesota in January—and lefty closer Glen Perkins are enjoying results far exceeding the quality of their pitching. This is particularly true of Boyer, whose weak peripherals, including a 3.9 strikeouts-per-nine rate, translate to a 3.96 Fielding Independent Pitching mark, nearly two runs higher than his actual 2.10 ERA. Still, that doesn’t explain why even the Twins' run differential fails to reflect their true level of performance.
Generally speaking, the Twins’ pitching isn’t causing the problem. Even with Perkins and Boyer getting unsustainable results, the bullpen is 19th in the majors in ERA (3.72), and the pitching staff as a whole has been roughly average, allowing 4.16 runs per game, just below the AL average of 4.19 and in sync with the team FIP of 4.13. There’s a lot of individual correction to come, with Kyle Gibson (2.61 ERA) and Mike Pelfrey (2.77) currently pitching over their heads, while Trevor May (5.07 but a 3.10 FIP) and Ricky Nolasco (5.51 against a 2.79 FIP) have pitched in bad luck.
Most of that should even out, however, with only a slight drop in overall staff effectiveness due to the combination of the Twins’ pitch-to-contact approach (they have the fewest strikeouts per nine innings and third-fewest walks per nine innings in the majors) and poor defense (they are 24th in park-adjusted defensive efficiency). Another mitigating factor could be the second-half return of Ervin Santana from his performance-enhancing drug suspension.
Where the real dissonance occurs is on offense, where the Twins are fourth in the majors in runs scored per game (4.59) but a lowly 19th in team OPS+ (93). How does that happen? The answer is clutch hitting.
There’s a popular misconception that the sabermetric community doesn’t believe that clutch hitting exists. You’d be hard pressed, however, to find anyone who would deny that clutch hits exist: A two-RBI hit with two outs and your team trailing by one in the bottom of the ninth inning is undeniably clutch, and the situation need not be that extreme for a hit to count as clutch. What advanced analysis has largely debunked is not the existence of clutch hits, but the idea of clutch hitting as a repeatable skill. On occasion you can find a hitter who has had more success in clutch situations than in general. But for every David Ortiz—who has a 105 tOPS+ in what Baseball-Reference terms “high leverage” regular-season situations, meaning his adjusted OPS in those situations has been about five percent better than his overall career OPS+—there’s a Derek Jeter with a 99 tOPS+ in those situations.
The Twins this season have posted a 129 tOPS+ in high-leverage situations, batting .301/.350/.453 as a team in those circumstances, compared to their overall team batting line of .257/.309/.391. Only the Angels have outperformed their overall line by more in high-leverage situations (133 tOPS+), and no team has exceeded its overall performance with runners in scoring position or with runners in scoring position and two outs more than the Twins, who have a 137 tOPS+ in both situations. None of that is sustainable. Last year, no team had a tOPS+ higher than 119 with runners in scoring position, and that team, the Mariners, is down to a 96 tOPS+ in those situations this year, a good indication of why they’ve been disappointing and why we may have expected too much from them coming into the year.
Leading the Twins’ clutch-hitting charge: veterans Joe Mauer and Torii Hunter. Mauer has gone 14-for-28 with five extra-base hits in high-leverage situations, good for a 1.410 OPS and a 297 tOPS+. Hunter has gone 12-for-29 (.414) with another five extra base hits, a 1.061 OPS and a 166 tOPS+. Trevor Plouffe, meanwhile, has four home runs in 40 high-leverage plate appearances, helping him match Hunter’s 1.061 OPS in those situations and giving him a 147 tOPS+ in that category. The career tOPS+ figures in high-leverage situations for those three players are 100, 97, and 80, respectively, and those include their 2015 performances.
The Twins simply aren’t going to keep this up, and when that correction comes and they start scoring runs at a rate more commensurate with the quality of their lineup, they’re going to sink in the standings because their pitching and defense won’t be able to compensate. Looking at the schedule, that reckoning could start as early as this weekend, when they start a 13-game stretch against the Royals, Cardinals, Cubs and suddenly-hot Rangers.
Third-order record tells us the Twins’ actual performance thus far this season is that of a team with a .449 winning percentage. Over 162 games, that translates to 73 wins. Over the remaining 113, it translates to 51. Add those to the 30 wins Minnesota has already banked, and it has an outside chance at finishing at or even above .500. But even if the Twins' hot start keeps them hanging around the wild-card race well into the second half, there’s unlikely to be much reason to take them seriously as contenders.