The 2025 Playoff Odds Are Here!
The playoff odds and projected standings for the 2025 season are now available!
As a refresher, to generate our playoff odds, we take each team’s projected performance and the schedule, and use those inputs to simulate the remaining season 20,000 times. We aggregate these outcomes to find the probability of a team winning its division or a Wild Card spot, along with its chances of winning the World Series and various playoff rounds. If a team has a 90% chance of making the playoffs, it means that 18,000 out of the 20,000 simulated seasons end with the team playing in October.
To calculate each team’s initial projected performance, we use individual player projections from the FanGraphs Depth Charts, which are a 50/50 blend of ZiPS and Steamer, prorated to our RosterResource Depth Chart playing time. We then aggregate those individual player projections by team and apply the BaseRuns calculation to each team’s batters and pitchers to get projected runs scored and allowed. Those BaseRuns runs scored and allowed calculations are used to calculate a projected winning percentage using the Pythagorean win expectancy. This is the number you’ll see on our projected standings page, which amounts to a team’s projected winning percentage versus neutral opposition. Here I’ll remind everyone that this is calculated before being run through the season simulation 20,000 times, so the projected standings can, and often do, differ from what you’ll see on the playoff odds page.
New this season, both by popular demand and as a matter of curiosity, we’ve added a version of the playoff odds that use the FanGraphs Depth Chart projections to calculate those initial team projections using WAR. Please note the existing BaseRuns version will continue to be our “official” version for the time being. Instead of running player performance through BaseRuns and calculating winning percentage using Pythagorean win expectancy, we add a replacement level to team WAR as calculated by our Depth Charts. For instance, our projections have the Dodgers forecast for 55.9 WAR. We add in a replacement-level team’s wins (for this projection set, that’s about 40 wins) and the Dodgers end up with a projected winning percentage of .595.
You’ll notice that the .595 winning percentage using the WAR calculation is pretty similar to the .599 Dodgers winning percentage we forecast using the BaseRuns calculation. Both methods produce similar results, though there can be significant differences depending on the team. Here’s a chart showing the correlation in 2024 and 2025. I’ve included the FanGraphs Depth Charts, ATC, THE BAT, and OOPSY projection systems as data points:
Since we only have this data archived for the past year, it’s not yet clear if there’s an advantage to using WAR for the initial projections instead of BaseRuns. The difference between the two is also not easily explained. These are two completely separate calculations. WAR includes numerous additional data points that the BaseRuns framework does not, such as park adjustments, league adjustments, fielding, catcher framing, context neutral linear weights, and FIP for pitcher calculations, just to name a few. But BaseRuns continues to be an accurate run estimator based on standard counting stats.
BaseRuns has a slightly higher correlation with team wins since 2002 than WAR does (.83 to .81 r-squared). This appears to be a slight advantage for BaseRuns, but it’s not a huge one, and they clearly both work pretty well. We’ll continue to monitor whether one or the other work betters for projections.
David Appelman is the creator of FanGraphs.
Since this is occasioned by the release of ZiPS, I wanted to point out that in the ZiPS 2027 projections there are about 400 pitchers at the end of the file whose WAR is not listed.
That’s an unfortunate projection for the Guardians.
I do think that they’re not as good as they were last year and they’re very likely going to take a step back. Into second-to-last place in the AL Central? I am not buying it. But maybe it is too much to project the four non-White Sox teams for 81-81?
Yeah, that’s not a correct projection. Even if you project the Guardians to drop last year’s run differential by a full third, that would still put them above what the Diamondbacks are projected for. I think this is massive over-projection on the part of the system here, largely on the bullpen if I had to guess. I know reliever WAR is finnicky and difficult, but there’s very little chance of the entire unit turning pumpkin.
Granted, the run differential for a lot of these teams seems wonky at best. Really, only five (1/6) teams seem like their projections in that regard look plausible.
21 CLE (80 W – 82 L)
+5.08 bullpen WPA (7th)
22 CLE (92 W – 70 L)
+10.41 bullpen WPA (1st)
23 CLE (76 W – 86 L)
-1.44 bullpen WPA (23rd)
24 CLE (92 W – 69 L)
+14.88 bullpen WPA (1st)
Obviously doesn’t account for the entirety of their yearly W/L variance, but Cleveland has needed not just a good, but the very best bullpen by leverage performance to break through the doldrums of late.
’21 and ’23 were both injury-riddled seasons for the whole roster. Doesn’t account for much of anything.
None of the projection systems like the Guardians much this year.
FGDC – 77.8
FGWAR – 79.0
ATC – 77.3
THE BAT X – 77.7
OOPSY – 75.6
Also we’re just taking player projections and aggregating them, and only the new WAR version involves any WAR calculations. Otherwise we’re plugging counting stats into BaseRuns (as the post describes).
No doubt the Guardians were a bit lucky last year. And I’m not a fan of their offseason moves. Still, if I were a betting man, I’d bet the over on those projections.
Off the top of my head I’d put their floor at 85 wins, if only because of their games against the ChiSox.
And while they haven’t done any big spending, neither have the Twins.
It’s a young team and one coming off injuries and bad years. Regression to the means works both ways.
Their floor is at 85 wins?! Oh my no, floors don’t work that way. A floor means 0.0% of their outcomes would be below that number, a/k/a worst case scenario imaginable. Their floor may be more like…66, 68 wins? Over half of their outcomes are probably 85 wins or fewer which isn’t a floor at all.
The Dodgers likely have around an 85-win floor. I wouldn’t guess anyone else is close to that.
In projections maybe.
in the real world the floor is whatever there roster *today* can be expected to achieve with only minor injuries or overperformance.
The FO isn’t planning for total disaster or miracles.
They do expect a mix of good and bad news.
Which means a few games above .500.
Call it a baseline if real world expectations offend you.
Too much proven talent to expect much less than .500 and too much uncertainty from the kiddie korps to expect much more.
As I said elsewhere, regression to the mean runs both ways: the ‘pen can be expected to be somewhat worse, the rotation can be expected to be somewhat better.
A dropoff of 8 games from last year is not unreasonable.
It’s not even third WC territory. Just enough not to be embarrasing.
At that point the fans start dreaming on Brito, DeLauter, and the rest of the auditioners and you start looking at how high they might go. You have to be a true believer to expect much more than 90 but 85 is reasonable enough for this crew.
That’s not how language works, or how statistical probability works. “Regression to the mean runs both ways: the ‘pen can be expected to be somewhat worse, the rotation can be expected to be somewhat better” – in theory, sure, but one of those things is inherently more volatile than the other, over any period of time, and it’s the one that was best in the league last year.
And, that’s not how “regression to the mean” works here, either. Over time, your overall results will regress to the mean value of your true talent, not to the league average results. If it worked the way you say, then every team would operate like the Marlins, because why would you bother spending if any player or set of players could perform as well or as poorly as any other at any time?
They don’t play the White Sox that many times. It’s not like the old days where you had the 18 intra division games.
The Guardians outperformed their Base Runs by 11 games in 2024, so the system is already seeing last year as a “true-talent” .500 team. You can think of it as ZiPS predicting them to lose 4-5 more games in 2025, not 15 more.
BaseRuns tends to struggle with teams that have an emphasis on pitching and defense, like the Brewers and Guardians, which they tend to over-project. That’s a known flaw. Sure enough, the two teams projected for the largest drops in that regard this season. My doubt stands.
The framing gap that Ben mentions in his article also over-penalizes the Guardians, who were 3rd in that category last season with an unchanged staff.
I don’t believe BaseRuns cares whether or not a team emphasizes pitching and/or defense. It’s just a run estimator and I don’t think it does any better or worse with teams that have better pitching and/or better defense.
It is certainly possible that projection systems have issues with this, but I don’t understand why BaseRuns would be the culprit.
Does BaseRuns actually take defense into account somewhere?
My only thought is maybe if the pitchers’ counting stats used in the calculations already have team defense baked in on the ZiPS/Steamer end.
Defense makes no sense but it is teams with great high leverage bullpens that will be underestimated, since removing leverage from a team built around maximizing leverage is going to underestimate that team. At the same time bullpens are volatile and it’s a tough way to consistently make a living so projection systems being skeptical is warranted anyway
The Twins went from 87 wins in 2023 to 82 in 2024, but also went from first to “second-to-last.”
Great projections for the Rangers and D-Backs. I am buying the D-Backs as a top 7 team but that Rangers projection is based on the team staying healthy and that seems unlikely.
There’s a lot of missed innings by Degrom/Mahle/Eovaldi/Gray counted in the frame already.
I’d like a 2023 WS rematch
The true sickos are here for the White Sox, and only the White Sox, projection. Dodgers? Yankees? Even the Pirates? Who cares… gimme the good stuff!
On that note, the AL Central teams look to all be losing about 3-4 wins because of the Sox 22 game improvement!
Not all.
Cleveland had a hard time with the ChiSox early in the year.
If there’s ever a time one can say that a team has been given a “rosy” projection at 62-100, it’s the one given to the White Sox this year. I think there’s a decent chance this year’s team could actually do WORSE than last year’s, but odds are they will do slightly better. I don’t see a world where they avoid triple digit losses.
I’m surprised the Mets at 85 wins I like the Mets bullpen and their rotation has high upside(though lots of risk too), especially if Senga stays healthy. Soto is a a huge upgrade and Lindor is Lindor and while I see a rebound for Nimmo I can see why projections would be down on him given his age and foot injury. I think Alvarez breaks out and Vientos imo is gonna be a 125 wRC+ guy. I don’t believe in Baty or Mauricio so I’d like Pete back at the money the Mets offered. I’m more with Baseball prospectus seeing this team at 89-90 wins.
The new WAR version has the Mets at 87.9, which is the highest of any. Generally we’re pretty conservative on win totals, and this year, for whatever reason, they seem even more compressed than usual (I’d have to take a look to see if that’s completely true, but observationally, it seems to be the case).
I don’t know why we’re more compressed than usual. Nothing about how we calculate these has changed, except the underlying projections.
Thank you!
They’re in a tough division to start with and the NL west is looking to be a terror.
Who are they playing in the AL? If it’s the east even 87 might be a reach. That rotation doesn’t inspire Dodger-like confidence.
They play every team in the al as been the case since mlb made the decision to have less division games and more interleague years ago.
So they play the AL east. Not fun.
But the central and west too.
Might be a wash then.
I’ve got a slight bit of confusion here. This article says that a replacement team’s level of wins is about 40. But the glossary notes that replacement level is set to have 1000 WAR per year, a replacement level of .294. .294 times 162 games is 47.628 wins per team at a replacement level (closer to 50). So, not a huge deal, just trying to figure out what I am missing.
For projections we’re not calibrating things to ~48 wins, so we recalculate what the replacement level team would be, based on the actual number of projected WAR.
Projected 98 wins for the Dodgers, finishing 13th in runs allowed per game?
At the risk of being a biased and entitled fan, I would place bets that both projections are low.
That comment is pretty stupid-can’t you even check the correct columns?
The guy writes a great article, then you come along.
You gave yourself quite the tongue-lashing there!
So you’re telling us the Mariners have a chance?
The projections say the Mariners have a chance. The projections always say that. But somehow the Mariners seem to fall short. Their offense is constrained by almost 1,500 projected plate appearances for Raley, Haniger and Garver, which translates to 450 Ks with only 55 or so home runs (mostly from Raley) to show for it. Add Polanco and Dylan Moore, projected for a combined 1,000 PAs with a close to 30% K ratio and there are too many empty ABs. (Signing Alonso or Bergman could have made a big difference.) Their tremendous starting pitching has kept them in the hunt. But they’ve been very lucky to avoid injuries to starters the last two seasons and that’s not something one can expect to keep up. I’d take the under on that win total.
I would not expect this projection system to include the changes in two stadiums for 2025. However, I suspect they will affect the number of runs scored in the league this year. I am going to take the over for both the Rays’ RS/G and RA/G values. The three-year PF for Steinbrenner Stadium for LHB HRs is 150.
Yeah the Trop was benefiting the Rays pitchers
So you’re saying there’s a chance.
–Jerry Reinsdorf
Wonder if Montfort will make wild claims about the Rockies potential win totals again? I sure hope so.
Way too high a percentage on the Braves to win the NL East. It’s like downturns are forecast for everyone but the oft injured Braves. This is one projection that will not age well.
Yeah, crazy to think the Braves, who’ve won the East 60% of the time throughout history (since joining), would have a 65% chance this year.
And “oft-injured”? In 2024? Yes. Otherwise? No, not at all.
The Braves have been tremendously successful historically, but their past division wins have zero bearing on their odds for 2025
Which teams will playing in the 2025 postseason? Probably the same ones that have been for the past 10 years.