The Numbers Behind the Seattle Mariners' Low World Series Odds
The four teams with the best records in baseball are all in the American League, and they are clustered in the east and west divisions.
The Red Sox lead the way with a 66-30 mark, followed by the Yankees (61-32), Astros (63-34) and Mariners (58-37). The first three teams are also at the top of our World Series odds, with Houston leading the way followed by New York and Boston (which is tied for third with the Cubs).
But Seattle is an exception. Despite their sparkling record, you need to scroll a bit to find the Mariners on the list. Seattle’s title odds stand at just 1.7%, the 12th-best odds in baseball. Contrast that with Houston (29.2%), the Yankees (14.1%) and Boston (11.0%), the AL's other best records.
Las Vegas isn't much more optimistic about the Mariners' chances, putting them at 16-to-1 (which translates to roughly a 5.9% chance). This is tied for eighth with the likes of Washington, Milwaukee, Philadelphia and Atlanta -- this group has an average winning percentage of 55.4%, which is roughly 6 percentage points lower than Seattle’s (or 10 games worse over 162 games).
The fact that the Mariners will almost certainly have to play in a wild-card game and then get through a tougher American League field partially explains why their odds are not higher, but that does not tell the whole story. The Yankees or Red Sox would face the same scenario, and Vegas sees both teams as more than twice as likely to win it all.
The bigger reason for this gap is an apparent difference in quality between the game’s elite teams and the Mariners. This disparity has not manifested itself in the win-loss column yet, but it very well might going forward.
They Are Who We Thought They Were...
Before the season, expectations were modest for the Mariners. Vegas set their win total at 81.5, while we projected them to win 82 games. They were expected to be better than average, but hardly by much.
The “over” seems like a good bet now, but signs point to the fact their “true talent” is still that of a slightly above-average team.
Seattle has outscored its opponents by only two runs over the course of the season, which is the 16th-highest margin in MLB and puts the Mariners between Tampa Bay and Colorado. This run differential translates to a .502 winning percentage, which is pretty much right where our models thought Seattle would be before the year.
Our models have not changed their “mind” much, either, as we project Seattle to go 35-33 to finish the season. Cue Dennis Green, because the Mariners look to be who we thought they were.
...Or Are They?
All that said, it would not be fair to stop here and dismiss the Mariners as “lucky” without at least looking at the factors that explain how they have outpaced their run differential so far.
As you probably guessed, the disparity is the product of a strong record in one-run games. Seattle is 26-11 in such contests (70.3%), and only Philadelphia (20-8, 71.4%) has been better in one-run contests.
Generally speaking, this is not a sustainable means of success. It has been well-documented that a team’s record in one-run games is not predictive of future success, and that a team’s run differential is a better predictor of future record than past record is. Another strong predictor is a team’s BaseRuns record, which is derived from underlying stats and independent of sequencing. Seattle’s BaseRuns record is just a game better than its Pythagorean (or run differential-based) record.
Teams tend to win about half of their one-run games in the long run, though there are some factors that can help. Rob Arthur found in 2016 that FanGraphs’ FIP-based WAR has a statistically “significant but fairly weak” relationship with a team’s record in one-run games (the correlation coefficient was 0.28, with 1.0 meaning direct correlation and 0 implying no correlation.)
Seattle’s bullpen is eighth in fWAR, which gets us some of the way there in terms of explaining the Mariners’ apparent overachievement. Their relievers rank second in MLB in “clutch” score, which is the difference between a player or team’s overall win probability added and its context-neutral WPA. This alone has accounted for about 2.7 wins.
In 387 high-leverage plate appearances (as defined by Baseball-Reference), opponents are hitting just .234 against the Mariners' bullpen, and this is actually in spite of a .320 BABIP. The Seattle bullpen has a 28.4% strikeout rate in this spot and an 8.4% walk rate, so it has genuinely been good here.
Still, this accounts for only roughly a third of Seattle’s overachievement as defined by “clutch score.” Its pitchers overall have a +4.3 clutch score, which is second in MLB, while its hitters lead baseball by a wide margin. The gap between the Mariners’ 4.6 clutch wins by Seattle's offense and second-place San Francisco’s 2.9 clip is equivalent to the margin between the Giants and eighth-place Cardinals .
In the case of the hitters, it really does seem like a BABIP thing as Seattle's BABIP gets better as the leverage of the situation gets higher.
The Mariners’ OPS in high-leverage spots is 126 percentage points higher than its overall OPS, which is more than double the next best team, per Baseball Reference. This increase owes much to a 39 point BABIP spike, which is second-highest.
Seattle has had a 32.7% increase in ISO in high-leverage spots, but we have nearly 3,000 plate appearances that suggest the Mariners' offense is more or less mediocre and only 709 plate appearances that say otherwise.
While it is true those 709 plate appearances have had the biggest impact, that is no guarantee the hits will continue to cluster in this fashion. In 2015, Jeff Sullivan of FanGraphs found there was no correlation between offensive clutch score in the first half of the season and the second half (he reached the same conclusion about starter and reliever clutch score also).
Plus, in terms of the stats that are the most reliable in a small sample -- strikeout and walk rate -- there differences are significantly less glaring. Seattle’s high-leverage strikeout percentage is virtually identical to its combined rate in medium- and low-leverage spots (20.5%) and while there is a bit more real estate between the walk rates, neither is particularly good. An overall walk rate of 7.9% would rank 23rd in MLB, while one of 6.9% would rank 29th.
Overall there do seem to be a lot of warning signs here, and it seems likely that the Mariners will likely play much closer to .500 baseball going forward than .600.
Still, while the M's great first half is largely the result of an unsustainable process, Seattle does not have to give back its 58 wins. Those victories are already banked, and they're a big reason why their playoff odds are 75.4%, per our models. If they get into the playoffs, they'll have a chance, because in the postseason, wild things happen.