Zerillo: Comparing My 2021 MLB Win Total Projections to PECOTA and the Rest of the Market
Thearon W. Henderson/Getty Images. Pictured: Jose Abreu, Tim Anderson, and Yasmani Grandal.
Now that public-facing baseball data leaders like Baseball Prospectus and FanGraphs have released initial 2021 standings projections, conversations are emerging about the upcoming MLB season, and books are starting to release win total over/unders.
Since the majority of the public uses these freely available projections, and the books certainly take a peek at them, if you want to get the most value out of the total wins market, you have to create your projections and pounce early.
However, you should also compare your own projections to those other trusted sources and create a consensus or composite projection to see how your projections stack up.
It's essential to look at more than one data point when creating projections for betting purposes; some forecasts will come in high, and others low on specific teams. But I have found that when the projection market lines up with or against an MLB win total, on aggregate, they tend to be right.
For example, during the 2018 and 2019 seasons, Davenport,FanGraphs, and PECOTA (by Baseball Prospectus) projections went a combined 92-78-10 (54.1%) against listed win totals, but their composite projection finished 34-22-4 (60.7%) over the same period.
Furthermore, I have found additional value by combining my own projection with the industry projections, whether by taking an average of the four projections, or weighing mine (55%, and 15% for each of the other three) accordingly.
How do these projection sets stack up for 2021?
2020 Win Total Composite Projections
I provided both projections (25% average, and the 55%/15%/15%/15% weighted) in the sections below, in order to compare the relative differences between my 2021 MLB win total projections and the rest of the market.
Using a cut-off of three wins (plus or minus), the average projection suggests value on four win-total wagers in the American League: Chicago Under, Houston Over, Oakland Under, and Texas Over. The weighted projection concurs on Chicago, Oakland, and Texas, with Houston (+2.9) narrowly missing the cut.
The Texas recommendation is largely the result of one outlier (Davenport at 80 wins), and although all four projections (including my own) are above their listed total, I'm still having a hard time justifying an Over wager on one of the worst teams in baseball.
I am equally hesitant about betting Oakland's Under. Publicly available projections have had a difficult time quantifying the A's for three consecutive seasons, but an under wager starts to come into range at 88. If you wait until closer to opening day, this line should continue to rise as the Athletics solidify their roster in the bottom of the free-agent pool.
I have the most optimistic projection on the White Sox (88.9), but that still falls short of their listed initial listed total.
The one obvious composite play at the Westgate is the Orioles Under 66.5 wins. While I'm lowest in the projection market at 61.8, even the high end (65.6 from PECOTA) falls short of the listed total.
I'm heavily invested in the Astros divisional and world series futures, otherwise, I would consider an Over wager at 86.5 wins.
Visually, here’s how the AL divisions break down:
Using a cut-off of three wins (plus or minus), the average projection suggests value on two win total wagers in the National League: St. Louis Under and Pittsburgh Over. The weighted projection swaps the New York Mets Over in exchange for Pittsburgh as a potential play.
All four projections have the Mets and Padres exceeding their listed totals.
I'm interested in both plays at 90.5, and 94 respectively.
The Pirates are amongst the most undervalued teams on the board — and also expected to exceed their win total by each projection. The cumulative talent level in the NL Central is so low (more on that in the next section) that it's difficult to project the worst team in the division to lose 104 games.
I have the most optimistic projection on the Cardinals by at least 2.7 wins, and each of the other three projections would suggest hitting their Under. Notably, the Cardinals have beaten PECOTA projections more consistently than any other organization, and I think their typically strong team defense is partly the cause.
I feel similar about the Cardinals' under as I do the A's Under or Rangers' Over: hesitant, but giving them adequate consideration.
PECOTA's projection of 82 wins for the Braves is one of the more surprising figures, but they have been consistently low on this Atlanta team for three consecutive years.
As for the Marlins, the opening total at the Wynn (67.5) represented a strong Over wager based on my projection alone. But the composite projection now suggests hitting the Under at the Westgate. I would be inclined to play both sides and hunt for a fairly significant middle, but I think the market settles closer to the DraftKings total (70.5).
Visually, here's how the NL divisions break down for 2021:
Balance of Power
One other thing I like to look at is to see how the balance of power stacks up between the leagues and divisions on a year-to-year basis.
The National League was stronger from 2018-19, winning 46 additional games against the American League in interleague play, but the AL knotted up the overall record in 2020 (149-149) as teams siloed off into three regional mini-leagues.
Only the FanGraphs projection set expects the AL to be the better league in 2020, and all unanimously agree that the NL Central is the weakest division.
As a reminder, the central divisions occupied seven of the 16 wild card spots last season, and they went a combined 2-14 in the playoffs, failing to win a single postseason series.
I could see both the White Sox and Twins making the postseason together, but it's difficult to imagine two NL Central teams winning enough games to qualify.