In honor of the 2022 Major League Baseball season, which starts April 7 — and is actually a thing! — FiveThirtyEight will be focusing our attention on the most intriguing team in each division. Today we take a look at the American League Central, which had a new champ last season but whose old champ isn’t quite ready to give up the crown for good.
After a pair of division titles for the Minnesota Twins in 2019 and 2020, last year’s AL Central belonged to the rising Chicago White Sox. Chicago had been one of MLB’s fun breakout stories from the shortened 2020 campaign, and it proved to be just as entertaining — and good — in 2021. For this year’s follow-up, the White Sox will return a similar group, with the biggest departure being starting pitcher Carlos Rodón.
All of the different projections we use to set preseason Elo ratings
think Chicago remains the AL Central team to beat in 2022.
The projections like the ChiSox again in 2022
AL Central teams by predicted 2022 MLB win totals, according to a composite of FiveThirtyEight’s Elo ratings and three statistical projection systems
But the Twins don’t appear to agree. In something of a surprising move, Minnesota won the derby to sign former Houston Astros shortstop Carlos Correa earlier this month, outmaneuvering the New York Yankees and several other big-market suitors in the process. (Amazingly, the Yankees enabled the signing to a degree, by taking on Josh Donaldson’s contract in an earlier trade with Minnesota — freeing up cash the Twins then spent on Correa.) This was in addition to deals that brought in starter Sonny Gray (2.7 wins above replacement in 2021) from Cincinnati and catcher Gary Sánchez (1.1) and infielder Gio Urshela (0.8) from New York. Both former Yanks have the potential to be better than they were last season, as does newly signed starter Chris Archer (who was once one of the better pitchers in the AL). Minnesota is clearly positioning itself to challenge the White Sox again after Chicago’s year atop the Central standings.
How realistic is that bid? The Twins are in a really interesting spot heading into 2022 because they were really good in both 2019 and 2020 — with an average winning percentage of .612 (which equates to 99.1 wins per 162 games) across those seasons — yet they were also really bad in 2021. Suffering a stunning collapse in both pitching and defense, Minnesota won just 73 games and finished dead last in the AL Central, behind even the Detroit Tigers and Kansas City Royals. So looking ahead, we now get a fun experiment in projecting which version of the Twins was more “real,” in the sense of having greater predictive validity for 2022.
Traditionally, if we want to predict a team’s record next season from its recent records, we would assign 59 percent weight to last season, 28 percent to the season before that and 13 percent to the season before that. This implies that Minnesota’s very bad season in 2021 should “count” more than its two very good seasons in 2019 and 2020 when thinking about 2022, but that those two seasons should still carry a reasonable amount of weight.
Of course, there are a bunch of complicating factors to this: First, much of Minnesota’s roster has changed since 2019. Second, the Twins were older than average in 2021, and older teams that fit Minnesota’s pattern of two-good-years-and-a-bad tend to underperform the next season. The Twins also exceeded their Pythagorean expectation just to post 73 wins last year — another reason we would expect them to underperform in 2022 — and one of their good seasons came in the chaotic 60-game sample of 2020. And finally, a lot of the weight placed on earlier seasons is probably just attributable to regression toward the mean. If we add a weighting component for a .500 record as well, the importance of three seasons prior disappears entirely, and the weight for two seasons prior drops from 28 percent to 15 percent.
Still, Minnesota has a big bounce-back season in its sights, and the history of teams trying to accomplish a similar arc across multiple years is fascinating. Going back to 1969, there were five other teams that won more than 90 games (per 162) in two consecutive seasons, then won fewer than 75 (per 162) in the following season. What happened next for those teams?
- The 2001 Houston Astros were coming off a 72-win 2000 season that represented a 25-win decline from 1999. Injuries to Craig Biggio, Ken Caminiti, Shane Reynolds and Billy Wagner had torpedoed the team’s chances that year, but they mostly returned to better health in 2001 (and Caminiti was replaced at third base by Vinny Castilla), while an influx of fresh stars assuming larger roles — like outfielder Lance Berkman and pitchers Roy Oswalt, Wade Miller and Octavio Dotel — helped Houston bounce back to 93 wins and another division title.
- The 1972 Minnesota Twins were trying to reclaim the form of 1969 (97 wins) and 1970 (98 wins) after collapsing to 74 wins in 1971. Despite counting numerous future Hall of Famers among their ranks, Minnesota’s best players almost uniformly slumped that season — only pitchers Bert Blyleven and Jim Kaat stood out — and the team got practically nothing from its newcomers. Undeterred, the Twins ran it back in ’72 with a similar core, and the results were improved — Minnesota finished .500 (77-77) — but nothing like the first-place finishes of 1969 and ’70.
- The 2005 Seattle Mariners had hoped to hark back to an even longer run of success than the 2022 Twins. The M’s won at least 90 games each season from 2000 through 2003, including a record-tying 116-win season in 2001, and they had done it while weathering the losses of Ken Griffey Jr. and Alex Rodriguez. (Admittedly, adding Ichiro Suzuki helped take the sting out of those losses.) But the 2004 version was missing several key pieces from that run — most notably outfielder Mike Cameron — and the league’s fourth-oldest roster collapsed to 63 wins despite Ichiro’s brilliance. The following season saw more of the same, with Seattle winning just 69 games despite a new manager and the big-ticket additions of Adrian Beltre and Richie Sexson.
- The 2015 Texas Rangers found themselves at an interesting crossroads as a franchise. The team had gone to back-to-back World Series in 2010 and 2011, coming about as close as possible to a championship without actually winning, and it followed that with more than 90 wins in each of the next two seasons. But members of that core (such as Ian Kinsler and Nelson Cruz) eventually went to other clubs, newcomers (like Shin-Soo Choo, Prince Fielder and Colby Lewis) failed to deliver, manager Ron Washington lost his job, and the injury-riddled Rangers fell to 67 wins in 2014. Yet Texas wasn’t finished. It reloaded its suspect rotation, broke a number of younger players into its lineup and won 88 games, returning to the playoffs — and participating in one of baseball’s most memorable innings ever along the way.
- The 1996 Montreal Expos were two years removed from one of the biggest what-ifs in MLB history — a potential World Series run thwarted by the 1994 strike — but many of the best players from that team (Larry Walker, Marquis Grissom, Ken Hill and John Wetteland, among others) were already long gone. The 1995 club was still long on young talent, headlined by 23-year-old ace Pedro Martínez, but it also had the cheapest roster in baseball, and its offense collapsed (from eighth in OPS+ to 27th) en route to a 66-78 record. More departures followed for ’96, but Martínez continued to shine — and a top-10 farm system yielded pieces of the next competitive Expos squad (including a 21-year-old right fielder named Vladimir Guerrero). Though it did not make the playoffs, Montreal rebounded to 88 wins, an improvement of 13.8 wins per 162 over the previous year.
These case studies truly run the gamut of the potential paths Minnesota might follow in 2022. The Twins can’t quite hope to fully replicate the injury bounce-back potential of the 2001 Astros or 2015 Rangers, just because they were not notably hurt (relative to other teams) in 2021. But they can relate to the way a number of these teams banked on a combination of new additions and positive regression to get a do-over on a down year. For some of their historical comparisons, that plan worked; for others, not so much. On average, the teams above dipped from 96.7 wins per 162 to 70.2 in their bad year and improved to 83.8 wins per 162 the following season.
If Minnesota follows that arc — which basically matches up with our blend of projection systems, for what it’s worth — it likely won’t be enough to wrest back control of the AL Central from Chicago. It would, however, leave the Twins near the range of what it will probably take to make the playoffs in this year’s expanded 12-team postseason format. And remember, that’s just using the average outcome. For a team coming off such an up-and-down sequence of seasons — and having shaken up its roster with such splashy offseason moves — the true range of outcomes for Minnesota in 2022 could be limitless.
The author of a no-hitter last April signed with the San Francisco Giants right after the MLB lockout was lifted.
The author of a no-hitter last April signed with the San Francisco Giants right after the MLB lockout was lifted.
Which gives two-thirds weight to an average of the projections found at Baseball Prospectus, FanGraphs and Clay Davenport’s site, and one-third weight to a regressed form of last season’s final end-of-year Elo ratings.
The author of a no-hitter last April signed with the San Francisco Giants right after the MLB lockout was lifted.
Which gives two-thirds weight to an average of the projections found at Baseball Prospectus, FanGraphs and Clay Davenport’s site, and one-third weight to a regressed form of last season’s final end-of-year Elo ratings.
Using our JEFFBAGWELL metric to blend WAR from Baseball-Reference.com and FanGraphs, for which you can download data on GitHub.
The author of a no-hitter last April signed with the San Francisco Giants right after the MLB lockout was lifted.
Which gives two-thirds weight to an average of the projections found at Baseball Prospectus, FanGraphs and Clay Davenport’s site, and one-third weight to a regressed form of last season’s final end-of-year Elo ratings.
Using our JEFFBAGWELL metric to blend WAR from Baseball-Reference.com and FanGraphs, for which you can download data on GitHub.
Sánchez had 2.7 WAR as recently as 2019, while Urshela played at a 4.9-WAR pace per 162 team games in 2020 and had 3.5 WAR in the full-length schedule of 2019.
The author of a no-hitter last April signed with the San Francisco Giants right after the MLB lockout was lifted.
Which gives two-thirds weight to an average of the projections found at Baseball Prospectus, FanGraphs and Clay Davenport’s site, and one-third weight to a regressed form of last season’s final end-of-year Elo ratings.
Using our JEFFBAGWELL metric to blend WAR from Baseball-Reference.com and FanGraphs, for which you can download data on GitHub.
Sánchez had 2.7 WAR as recently as 2019, while Urshela played at a 4.9-WAR pace per 162 team games in 2020 and had 3.5 WAR in the full-length schedule of 2019.
This is based on full, 162-game seasons in MLB’s divisional era (i.e., since 1969).
The author of a no-hitter last April signed with the San Francisco Giants right after the MLB lockout was lifted.
Which gives two-thirds weight to an average of the projections found at Baseball Prospectus, FanGraphs and Clay Davenport’s site, and one-third weight to a regressed form of last season’s final end-of-year Elo ratings.
Using our JEFFBAGWELL metric to blend WAR from Baseball-Reference.com and FanGraphs, for which you can download data on GitHub.
Sánchez had 2.7 WAR as recently as 2019, while Urshela played at a 4.9-WAR pace per 162 team games in 2020 and had 3.5 WAR in the full-length schedule of 2019.
This is based on full, 162-game seasons in MLB’s divisional era (i.e., since 1969).
The weight for last season also drops from 59 percent to 43 percent, with the .500 record grabbing 42 percent of the weight. (Yes, baseball is a sport that very heavily regresses to the mean.)