Six Takeaways From Our 2025 Playoff Odds Release
Today, we released the first run of our playoff odds for the 2025 season. With both the ZiPS and Steamer projections loaded in and playing time projections added to the mix, the FanGraphs supercomputer (okay, fine, our cloud services account) can get cranking and spit out some predictions. As is customary, I’ll walk through my first thoughts on them, while later today, Michael Baumann will contribute his own takeaways on the teams most likely to surprise our model. Let’s quickly walk through the process, and then get to the takeaways.
The model itself remains simple. We use those aggregated production and playing time numbers I mentioned earlier to create team-level projections, then use BaseRuns to turn individual outcome projections into scoring and run prevention. That gives us team strength against a neutral opponent. We use those values to simulate the season 20,000 times. The odds are a summary of those simulations as of earlier this morning. That might sound intuitive, but intuition doesn’t always match reality, so let’s go division-by-division to look at how our model got there and what I think of it.
The AL East Is as Outrageous as Always
All five AL East teams project for records above .500. Not by a ton – we’re not particularly excited about the Rays or Blue Jays this year – but there are no true weak links in this division. Our model considers the Yankees the third-best team in baseball in terms of their playoff odds, but thinks they’ll have their work cut out for them in a division where every team has a reasonable shot at playing in October.
The Yankees were a question mark heading into the offseason – with Juan Soto gone, their team is going to look much different than last year. But between adding another ace in Max Fried, trading for Devin Williams, Fernando Cruz, and Cody Bellinger, and shoring up last year’s weak point of first base by signing Paul Goldschmidt, they’ve done a good job of keeping the team around Aaron Judge dangerous and playoff-ready. This isn’t a case of an obvious and uncontested favorite, but this is a good team, one we project to score the most runs and allow the fewest in a difficult division.
Second place is up for grabs, and I want focus on the Red Sox and Orioles to make note of how our model handles converting WAR into projections. Notably, it doesn’t. It doesn’t use WAR as an input at all, in fact. Instead, WAR is downstream of the outcome-level projections, just like BaseRuns is. Here’s how I think about it. A full projection for a major league team has a ton of variables. Singles, doubles, strikeout rate, baserunning, receiving, fly ball rate, the list goes on. You could try to distill those into one number — that’s WAR. You could also try to figure out how a collection of players will score and prevent runs — that’s BaseRuns. They’re taking the same raw data, putting it through their rigorous transformational calculations, and coming up with slightly different answers. This year, we’re actually running a parallel, WAR-based playoff projection, which David Appelman described here.
To wit, as measured by WAR, we project the O’s as five wins better than the Red Sox. But when you look to BaseRuns, we think the Sox are roughly a win better than the O’s. That’s strange! It comes down to a few things. First, there’s a synergy effect. We think that the Orioles’ offense will be slightly less effective than you’d expect if you merely summed up the individual options. (That’s minor in the grand scheme of things, to be clear; we still think they’re great on that side of the ball.)
Second, WAR treats all contributions as equal, but BaseRuns operates based on expected record, which doesn’t. Outscoring your opponent by 50 runs, 650-600, is likely to lead to more wins than outscoring them 850-800. The O’s have one of the best offenses in baseball. They also have the worst pitching staff of any playoff contender. They’re going to end up in shootouts, and it just makes sense that playing a lot of 6-5 games gives you less margin than playing a lot of 4-3 games. BaseRuns accounts for that imbalance in its formulation of opponent-neutral winning percentage. WAR does not. How could it? It’s just counting up the runs and turning them into wins at a linear rate.
Oh yeah, and our current calculation of BaseRuns doesn’t include framing. It’s pretty tough to squeeze into the model, and we haven’t figured out how (if you have any ideas, you know where I write). That’s a win and a half that BaseRuns isn’t accounting for; Connor Wong is meaningfully worse than Adley Rutschman when it comes to getting extra strikes. I look at these odds as a tossup, and I’ll also note that the O’s will have an easier time improving at the deadline, because the pitching gives them a clear weakness to upgrade.
As I mentioned, the Rays and Blue Jays also play in this division, which is unfortunate for them, because I think either could win the AL Central. Heck, Toronto is much closer to Detroit than it is to any team in the East — maybe they should try to move! Perhaps the Rays and the Jays can take some comfort in the fact that there are Wild Card berths very much up for grabs in the AL. Sum our odds up, and no division projects for more playoff teams than the AL East. The road to get there, though? It’s going to be rough.
The Guardians?!
I don’t know what to make of our AL Central projection. Most of it makes sense – no one’s amazingly good, the White Sox are the worst team in baseball, and the Twins are better than last year’s collapse would suggest. But I don’t see Cleveland as the fourth-best team in the division, despite what the model says.
Why is it so down on their chances? In a word, the offense. The Guardians’ offense was middling last year, with component statistics worse than that. They traded one of their best hitters, Josh Naylor, away this winter. They also traded Andrés Giménez, one of their better players, and got scant major league help in the deal; Luis L. Ortiz is the only player acquired in the de facto three-team trade who we think will play for Cleveland in the majors this year, and he’s a fourth starter. There are few teams across all of baseball, never mind playoff teams, that have a worse offensive unit, even with a few stars at the top of the lineup.
That’s a lot of work for the pitching staff to pick up. The Guardians relied on their bullpen to a huge extent last year, and the bullpen delivered. But it’ll need to be excellent again this year, while picking up a huge innings workload, to keep this team afloat. The starters are once again a dicey proposition. After Tanner Bibee, who I love, there are a lot of question marks here.
Still, I’m skeptical of the model’s view that Cleveland’s median outcome is below .500. Between the relatively soft division, the addition of Carlos Santana, and two Top 100 prospects who could debut this year and might provide an offensive jolt, I think there’s more upside here than the model suggests. The broad point, though, is a good one. If you didn’t know this team won 91 games in 2024, you’d look at it and wonder what the fuss was.
As I said, the rest of the division largely tracks with my expectations. The Twins might have had a boring winter, but they have a solid offense, some interesting young contributors, and a reasonable starting rotation. The Tigers have an even worse offense than the Guardians, but they have one of the best few pitchers in baseball so they might be able to make it work. The Royals have Bobby Witt Jr. and three great starters; the rest is up in the air. This should be a fun division in 2025, and certainly a contested one.
Wild, Wild AL West
Three teams within a win of first place? It almost happened in 2023, when the Astros and Rangers won 90 games and the Mariners 88. We think the division will be close like that again, despite uninspiring offseasons from Seattle and Houston. The A’s and Angels are clearly a step behind this top trio, though Sacramento’s respectable projection, in its first season as a big league city, is a breath of fresh air.
I’m surprised to see the Astros so high. They traded Kyle Tucker away and lost Alex Bregman to free agency. The model likes their offense quite a bit, though. Yordan Alvarez is great, Jose Altuve is ageless, and Christian Walker stepping in at first base should provide a huge boost relative to last year’s numbers. Heck, even the Tucker trade was less of a strict downgrade than I’m making it out to be; Isaac Paredes seems tailor-made for Houston’s snug left field dimensions. For me, this model is a useful sanity check. I’m down on the Astros’ chances relative to the projections, but there’s still a lot to like about this team. That might be the most useful part of this whole exercise. Winter vibes are strong, but the model cares quite a bit about past production too, something that’s easy to overlook when you’re thinking mainly about personnel changes.
The Mariners are seemingly perpetually competitive without ever assembling a reasonable offense. We think their pitching staff is one of the best in baseball, and they still play in an offense-suppressing stadium. We have them second only to the Dodgers in run prevention. In terms of offense, though, we have them tied with the Guardians for sixth-worst in the majors. It’s the same story in Seattle, and even with many of the same characters — no real offseason additions, few meaningful subtractions. If Julio Rodríguez returns to superstar form, I think this offense will be acceptable, but otherwise they’re looking at a lot of frustrating 3-2 losses.
Meanwhile, the Rangers are the division favorites, at least nominally; their odds are roughly two percentage points higher than Seattle’s. They’re mostly running last year’s squad back, with two major exceptions. First, we’re expecting mostly full seasons from Jacob deGrom, Tyler Mahle, and Jon Gray. We’re not talking 200 innings for any of them – the highest innings projection is Mahle’s 141 – but it’s a lot more than the 126 the three combined for in 2024. Add in contributions from the Vanderbilt duo of Kumar Rocker and Jack Leiter, and the Rangers will have an excellent pitching staff when healthy.
That just leaves the offense, and while new additions Joc Pederson and Jake Burger are going to get a lot of the attention, the real question is whether the existing team can bounce back. The Rangers had a surprisingly flat year at the plate, finishing 18th in the majors in runs scored; they were third in the league in 2023 when they won the World Series. The model thinks this squad is primed for a rebound, and I do too. Even if Adolis García is cooked, there’s just so much talent here. Forget 2024 — the computer and I agree that this lineup is formidable.
The Braves Still Have It
Last season was a nightmare for Atlanta. When I wrote about intelligently considering depth, I used the Braves as an example of a team that would still be competitive even if it lost its three best players for most of the year. That basically came to pass; Ronald Acuña Jr. and Spencer Strider barely played, and Austin Riley, Ozzie Albies, Sean Murphy, and Michael Harris II all missed more than a month of play. Even Chris Sale’s Cy Young season ended in injury; he didn’t pitch during the Braves’ brief postseason run thanks to back spasms. All that, and they won 89 games anyway. This is a very good team.
The Phillies and Mets are no slouches, of course. We have all three NL East powerhouses among the top seven teams in the majors in terms of playoff odds, and top six in terms of winning percentage against a neutral opponent. Of the two, I like the Mets’ chances of winning the division slightly more despite the model giving them worse odds, but it has less to do with the roster as currently constructed and more to do with potential upgrades. The Phillies look maxed out to me – their starters are among baseball’s best, their bullpen is strong, and they have seven-ish established offensive players who aren’t going anywhere. Unless they can do a ton of upgrading in the outfield, probably including finding someone to take Nick Castellanos off their hands, this is who the Phillies are.
The Mets, on the other hand, have all kinds of holes. The rotation could use at least one more arm, preferably an elite option to round out their playoff rotation. Brett Baty probably isn’t a big league-caliber third baseman, I’m sorry to say. Jose Siri was available essentially for free this offseason. Jeff McNeil is on the downslope of his career. The bottom third of their lineup should be trivially easy to improve on. Combine that with a GM known for his ability to make in-season improvements, and I think the Mets have a better chance of turning in a spectacular season, even if I agree that the Phillies look a hair better at present. Realistically, though, this is Atlanta’s division to lose, just like it seems to be every year.
Is ‘Ew’ a Takeaway?
I’m not going to sugarcoat it: The NL Central looks bad. There are no abjectly terrible teams here, so I don’t think it’s going to challenge any records for worst division in history, but there simply isn’t much top-end talent. The Brewers and Cubs are the only two clubs that look like legitimate championship contenders, and then only barely.
The Brewers won the division by a mile in 2024, but they did nearly nothing to improve their team. The closest thing to an upgrade here is trading Devin Williams for Nestor Cortes and Caleb Durbin, and that’s more of a sideways move. Meanwhile, Willy Adames departed for San Francisco, and a slew of depth players departed in free agency or were non-tendered. Individually speaking, losing Joe Ross, Bryse Wilson, Hoby Milner, Frankie Montas, Colin Rea, Gary Sánchez, and Wade Miley might not seem all that important. But all of those guys? That’s so many people! Pitching depth will be particularly important for Milwaukee this year, and a lot of that depth just departed.
The Brewers’ inactivity left a bit of a power vacuum at the top of the division, and the Cubs rose to the occasion, enough so that we now have them as division favorites. They made by far the biggest splash of any NL Central team this winter by trading for Kyle Tucker. Heck, they might have made the second-biggest splash by trading for Ryan Pressly. The Cubs’ vaunted farm system has mostly been churning out depth pieces and part-timers in recent years, but now they have a superstar anchor to make everyone else look a little bit better. The bullpen? It’s not great, which is why Pressly was such a nice addition. The rotation? I feel similarly; when you’re signing Colin Rea and expecting him to make meaningful contributions, it’s probably not one of the best units in the game. But in this division, Tucker, Dansby Swanson, and a reasonable supporting cast might be enough.
The rest of the division is downer after downer. The Cardinals? They’re third by default despite expressly announcing their intention to take a step back at the major league level. This feels like a team going nowhere; they failed in their quest to trade Nolan Arenado, but they’re also not trying to compete around him, which means they’re basically marking time until Chaim Bloom takes over. The Pirates have a former Boston executive of their own in Ben Cherington, and he seemed to like St. Louis’ plan of sitting out free agency, because his biggest signing was arguably Adam Frazier. Feels like quite a waste of Paul Skenes to me, and the model agrees, giving them a scant 20% chance of getting to the playoffs despite an abysmal division and a transcendent talent.
That just leaves the Reds, who I’m surprised to see projected in last place. I don’t know if Gavin Lux, Jose Trevino, Brady Singer, and Taylor Rogers will all work out, but I’d bet on at least one of them exceeding expectations. A full year of Matt McLain and an encore performance from Elly De La Cruz should make Jonathan India’s departure bearable. You can squint and see a very good team here. The model is skeptical of their pitching, particularly their bullpen, and it’s weighing down a perfectly reasonable offense. This feels harsh to me; I doubt their bullpen will be quite as bad as expected, if only by virtue of in-season turnover. That doesn’t make them amazing, but it definitely puts them right in the mix in a desultory division.
Snakes Alive
We get it, the Dodgers are great. We have them down for the best offense in baseball, as well as the best pitching staff. That’s hardly surprising, though. I’m more interested in the Diamondbacks, who boast the fifth-highest projected win total in the majors and fifth-highest playoff odds too. Signing Corbin Burnes was obviously a coup, and adding Josh Naylor should keep the offense humming even after losing Joc Pederson in free agency. It feels weird, because the last time we saw them in the playoffs they were plucky underdogs, but these Diamondbacks are for real. They have multiple elite offensive players, two capital-A aces, and a deep rotation that will likely see Jordan Montgomery shipped out for surplus pieces before too long.
It feels weird to see the Padres meaningfully behind the Diamondbacks, but in looking through these projections, I’m starting to feel the downside of A.J. Preller’s quantity-for-quality trading style. The bottom of this lineup is rough. We’re talking worst left field situation in baseball rough. A catching tandem composed of guys better suited to being backups rough. Batting your DH ninth rough. Sure, the stars are bright, but the whole idea of stars and scrubs is that you can cobble together average players around the stars more easily, and San Diego absolutely hasn’t done that.
The rotation is a similar deal. The top three names? Glitzy, marquee-worthy. The next two, and the group after that? Let’s just say that if the top trio misses much time, this projection has meaningful downside risk. That’s even before considering what happens if they trade Dylan Cease, as has been widely rumored. The Padres ended last list cycle with the thinnest farm system in baseball, with just three prospects garnering a 45 FV or better on our updated 2024 report, the fewest in baseball. That’ll happen when you trade marginal prospects at every turn to avoid throwing in top names in your dealings. It has worked for the Padres for a while, but you can see the downside in this year’s projections.
Finally, our projections see the Giants as competent bordering on good. I was surprised to see them so close to San Diego in playoff odds, and if you’ll remember from up above, these BaseRuns-based odds don’t take framing into account. The Giants have the best receiver in baseball in Patrick Bailey, while the Padres alternate between two stone-handed options. I’m not surprised by the top or bottom of the NL West – but the middle three are all intriguing in their own right.
Ben is a writer at FanGraphs. He can be found on Bluesky @benclemens.
This was pretty similar to my takeaways too.
The AL West is the one I am most curious about. The Mariners, Rangers, and Astros are three very flawed stars-and-scrubs teams. The Rangers probably have the scariest health track records, as they are relying on Seager, Jung, deGrom, Mahle, Gray, and Eovaldi to stay healthy is super rough. But the Astros are flailing for outfielders and once you get beyond Framber Valdez, Hunter Brown, Josh Hader, and Bryan Abreu the pitching staff is not good at all. And we all know about the Mariners, who have apparently decided against any improvements at all. A few injuries to any one of these teams and they could send them tumbling down the standings really fast, much more so than the more balanced teams in the AL and NL East.
The Yankees going into the season with the 18th best 3b situation in baseball is such a stupid own-goal. They lost to the Dodgers who signed everyone in the offseason and they’re content to roll into opening day with the same disastrous tandem that they had to replace midseason.
Bregman is still out there!
The chances they sign Bregman are exactly 0%. Odds are that this is it, with a remote chance that they trade for someone if they can move Stroman’s contract — which they shouldn’t be using as an excuse to stop them from doing anything but very much are.
The Yankees probably have the biggest volatility of any team in MLB. This team is loaded with excellent players with health questions, like Judge, Chisholm, Bellinger, Cole, Rodon, Fried…many of whom I think the projections might be low on, even if the playing time is too optimistic. I wouldn’t be surprised if they had the best record in baseball and I wouldn’t be surprised if injuries make them miss the playoffs entirely in that division. Anywhere between 103 and 83 wins seem plausible.
I would absolutely be surprised if they had the best record in baseball. It’s possible, but they’re projected 9 wins behind the Dodgers and 5 behind the Braves, and there’s a huge bolus of good teams projected only a win or three behind them.
Perhaps one of our esteemed overlords can weigh in on how many times in those 20,000 simulations the Yankees finished with the best record.
Looking at the projected run differential, the Yankees (+63) are exactly as far behind the Dodgers (+152) as the Pirates (-26) are behind the Yankees.
(and both the Rockies and White Sox are even further into the negative than the Dodgers are into the positive, which is pretty gross)
The runs scored/runs allowed per game on the 2025 projected full season playoff standings page at the division level are all borked up. Every team is in the 2s for both runs scored and allowed. Dodgers leading the world with 2.47 runs/game! (It seems like everything is erroneously divided by two)
Dodgers look to me to have the strongest offense in the league by a lot. I think it’s going to take an upset for any team to have a better offense than them, even the Braves. They’re rolling out a lineup where three players have a projected wRC+ of 140 or higher*. If you look at the Depth Charts projections, there are 22 players projected for a 130 wRC+ or higher and 500 PAs or higher**. The Dodgers have three of them. Not a single other team even has two.
But I think the projection for the Yankees’ rotation could be low, and the Dodgers’ starters are all pretty fragile, and there’s still a lot of questions about the middle infield on defense.
I’m not saying it’s likely they finish with more wins than the Dodgers; the Dodgers have a higher ceiling and a higher floor. The the Dodgers’ “downside” is probably something like 90 wins and the Yankees is probably like 83.
But I would not be super surprised either if the Yankees won more games. Lots of volatility in the projections, no games have been played yet, and there are a lot of excellent players suiting up for the Yankees this year.
—-
*three more players between 115 and 120, two projected league-average bats (Tommy Edman, Hyeseong Kim) and a guy who could bounce back to being really good (Conforto). Yikes.
**Joc Pederson platoons
OK, let’s put a (completely made up) percentage on it! I propose 50% Dodgers and 30% Braves, leaving 20% for someone else. Of that 20%, maybe 5% Yankees? 8%?
(now I’m really curious to see the simulation results).
Gotta love the cognitive dissonance involved when considering that parenthetical alongside your comments on other articles advocating for a salary cap. How do you come across a stat like that and conclude that the issue is TOO MUCH spending?
I have no idea what you’re getting at here.
This is a good time to point out that while FG is an excellent resource for data and generally provides appropriate and useful calculations, the number one missing element of any FG-generated value is the notion of statistical error. There are 162 games in a season, not an infinite number of games. Consequently, simply due to randomness, teams will overperform or underperform their projection even if the projection value is perfect.
Using a 95% confidence interval, the error bar size here is 12ish games. Assuming FG is perfect, we can expect the Yankees to win between 76 and 100 games. Ending with 83 wins should not be a surprise, nor a losing record. But 103 would be.
These things are very reasonable statements. Well, maybe not using a 95% confidence interval, I don’t really see the point in using almost two standard deviations out from the mean. I know that science uses that standard, but none of how we talk about these projections fits the actual assumption of 95% confidence. When we say “surprise” we’re usually thinking about something that maybe happens 10% of the time, which would be 80% confidence intervals (which is not a thing IRL but could be here, in the context of this conversation).
(Also, I know that isn’t the technical frequentist definition of a confidence interval, but the technical definition is even further away from how we talk about probability in real life)
I also don’t think that changes anything I’ve said, but if we adopt this framework, it does change how I would frame it. What I would say is this:
1) I think the center for the Yankees is a bit higher than this, probably around 93 wins.
2) I think the range of “would not be surprised / 80% of cases” would be 103 at the top and 83 at the bottom.
3) I think this is wider than my “would not be surprised / 80% of cases” than for other teams.
Thank you for clarifying your statement. This is a much more intelligent response to my comment about statistical error. Too often, I get, “Huh?” I’m perfectly happy with the choice of an 80% confidence region. The number of SDs chosen is subjective. (In biology, it is 2. In physics, it is 6.)
And, like you, I still stand by my statement. FG would improve if they were to provide some sort of guidance for understanding inherent error in the calculations. Too often, I read comments from posters saying something like, “The projection system clearly shows the Yankees are the favorites, followed by the Red Sox.” If error bars were included, folks might be more likely to recognize this division is essentially a toss-up among 5 teams.
This would be great. I don’t even know that we need to talk about confidence intervals for this exercise – just a visualization of the win totals of each team in the 20,000 simulations would be simple and intuitive.
This is completely right (and “show us the error bars” has been a hobbyhorse of mine for so long I’ve basically given up even posting about it, so thanks for bringing it up). It’s not just an issue with the projections, either — there are a lot of situations where people (including Fangraphs columnists) get confused or draw incorrect conclusions about how strongly predictive a given stat will be for the same reason. (There’s one writer in particular who talks about this or that stat “stabilizing” at some threshold N in an obviously wrong way.)
Frankly I’m not sure that the baseball statistics community even wants to learn enough actual statistics statistics to figure this stuff out, but it does regularly bother me too.
I literally spent today showing some people how to use a z-score table. I had no idea physics uses 6 standard deviations out! But I guess when your research is about stable physical laws two standard deviations is too easy.
One thing that’s always been commendable about PECOTA as a projection system is that BP has always presented it with multiple percentile outcomes, which is something that Dan has started to do here with ZiPs projections the last few years as well. We could use such a display for team-level projection outcomes, because I think people have a lot worse feel for the level of uncertainty there (and that each SD represents around 8-9 games)
BP does the nice bulge charts with their standings projections so you can see the shape of these outcomes, and it’s worthwhile because it’s not the same for every team – you even see some with two-lobe structure – in this year’s first release, the Rays, Cubs, Giants and Tigers all have a 2-peak structure where the actual median win total isn’t the most likely outcome.
BP’s got some really good data visualization people and it’s an area where FG could improve
You’re making me want to subscribe to BP.
I used to feel more comfortable recommending it here when FG wasn’t also in a subscription model, but if you can afford both it’s worth it, they do a lot of things very well and IMO there’s a lot of value in the fact that they design and use their own metrics that don’t always agree, because we’re dealing with fairly imprecise measurements anyway
I’m plenty happy with what I get for my Fangraphs subscription. This would be in addition, not instead of!
Thank you for the information. Yes, providing a visualization of the distribution of potential results would be helpful. I suspect a number of people would be surprised by how much of an overlap there exists between the potential outcomes of the teams in these models.
People should assume the default is that there’s enormous overlap. Its the situations like some recent vintage Dodger teams where there actually is not significant overlap that are noteworthy and remarkable.
Where can you see the position ranking breakdown that shows Yankees have 18th best 3b situation?
Teams -> Depth Charts -> 3B
thank you
If someone gave me 25-1 on the Nats making the playoffs, I’d take it in a heartbeat. Do I think they will? No, of course not. But they do have manager that has gotten the most out of a lot of bad/mediocre players and there’s certainly hope that they hit big on a few prospects, notably Wood and Crews, but a few others, too. I do understand why the computers say otherwise.
Never give up! Never surrender! Hey, at least the Nats got a mention in the comments section since they didn’t even get a “the Nats also play in this division” mention in the article.
The Nats still have work to do. The bullpen post-deadline was atrocious, and they’ve lost Finnegan, Garcia and Jacob Barnes from that group. I anticipate they’ll sign 2-3 guys to MLB deals and 2-3 more to minor league deals, but they’ll need to get those signings inch-perfect.
I think the rotation will be good, but I’m skeptical on the lineup. It requires Garcia to maintain or build on his 2024; Abrams to put up something closer to his first half OPS (.831) rather than his second half OPS (.586); Ruiz to fully recover from the illness that wrecked his 2024; and Tena/Rosario to be good (or good enough until House is ready). It has the potential to be the worst defensive infield in the league, especially if you include the framing of Ruiz.
I think they can be improved from the last two seasons (72 wins), but it’s hard to see them getting to .500, nevermind the postseason.
Their division just makes that so tough. Plop them into either central division and they could have a puncher’s chance.
Could any one of the Braves, Phillies, or Mets bottom out? Sure. Maybe even two of them. But all three of them doing so? Very long odds indeed.
I’m talking about making the playoffs. In a very technical sense, they don’t need any of them to bottom out to make the playoffs.
Winning the division would be shocking.
Fair! 25:1 on almost anybody making the playoffs (probably every single team but the ChiSox, maybe the Rockies) would be a decent enough bet considering that 40% of each league makes it in
Crews and Wood would have to improve dramatically, and Abrams would have to actually meet the projections. But they are at the right age where that is still possible.
For framing, could you use it to adjust the number of projected strikeouts for a team? Like, if you know Adley can flip X balls into called strikes then that’s worth Y extra strikeouts over what the O’s pitchers are projected for.
Another idea I had would be to use Patriot’s method of converting known linear weights into a B coefficient for your custom BaseRuns formula using the run value of flipping a ball to a called strike. I’ve never thought about including something “intermediate” like pitch framing, so I have no idea if it actually works. It wouldn’t have an A, C, or D factor and would just have some very small negative B value. But I guess it kind of makes sense since framing reduces future run expectancy (RE288), it would presumably lower the score rate B/(B+C). Just spitballing
Cleveland’s chances are panned every year — we fans are used to it. I don’t know how this season will go but I’m 100% certain that when we do eventually win the world series the pre-season prediction models will show the team limping out to the same results that this year’s do.
the closest Cleveland has come to winning a WS since 2000 was 2016, and they were tied for the 4th-best WS odds per the FG preseason projections.
Panned for a good reason, which is in the article.
Well at least they signed….uhh, Carlos Santana? I guess? Oye como va!
Projecting a 100+-run drop in their run differential from last season when they lost a grand total of one offensive player and one defense-only player while adding to the rotation and signing an unaging fey is where that argument loses me. It’s like they projected absolutely everyone to regress on both sides of the ball. I agree there should be a drop, but the depth of that drop seems like overkill. Going from 11th in run differential to 23rd is a drop only equaled by the Brewers, who go from 3rd to 15th, a team that these projections are perennially too low on.
The Yankees lost the 2nd best hitter in the AL by wRC+, one of their best relievers, and one of their better starting pitchers, and lost only 3/4 as much by run differential after a historic offensive season by both their best hitters. Gil is unlikely to repeat his outcomes from last year, and they got stupid lucky on BABIP. Yes, they added Freid, Williams and Bellinger, but that’s not going to make up that much ground. Yet still, they are projected for the 3rd best r-diff in the sport.
TLDR: I don’t buy it, and why should I? The team that won 92 games last year doesn’t look different enough from this iteration to project a massive backslide. I’d say 84 wins is fair to project. 77 is an overcorrection.
Also, as a note, the projections are pointed out as being unable to grasp framing value, and Cleveland was 3rd in framing runs last season. Their catching unit is unchanged this season. They might even be better once Fry can get back behind the dish.
Well they overperformed both pythag (by a little) and BaseRuns (by a lot) last year, so I’m not sure they were ever a true-talent 92-win team.
If they are, say, a true talent 85-win team (which seems reasonable enough) then they should be essentially as likely to win 78 as they are to win 92 (unless they have discovered some heretofore unknown secret sauce)
BaseRuns tends to struggle to grasp pitching-and-defense teams like the Guardians and Brewers. Same reason for their massive gap. It’s the reason I don’t follow that quite as closely. It’s a known flaw in that particular system. Pythagorean record does a better job with those types of teams.
BaseRuns is useful, don’t get me wrong, but it’s a tool. It cannot be the sole basis for an argument like this imo.
While that’s fair to project the true-talent as lower than 92 wins and closer to something like 84-85, the median projection should not be 77 wins in that case. That’s what the computer is predicting here. They did lose some value, but they lost a grand total of like five WAR between Giminez and Naylor the Elder, and added back at least three between Santana, Seawald, and Ortiz, even if Cecconi provides little to no value. And that’s not counting getting more than two starts out of Bieber or the upside of Bazzana/DeLauter, an obvious bounceback from Williams and Naylor the Middle, or steps forward from literally any player on MLB’s youngest (maybe 2nd youngest with Santana) roster.
I don’t think the loss of wins is wrong, I just think it’s too far.
The biggest differences I’m finding in Depth Chart projections vs ‘24 RA-9 WAR (seems more comparable to Base Runs than FIP-based) are Gaddis (0.0 vs 2.8) and Clase (1.7 vs 4.1). This maybe applies more to the WAR-based playoff odds, but that’s 5 erased wins right there.
Last year, I never thought Gaddis would keep up the charade of a high-leverage reliever with his 1.57 ERA, but I expected a slightly better projection than 4.02 ERA/4.28 FIP.
Obviously Clase’s 0.61 ERA can’t be a mean projection for this year, but 2023 was his only worse season than his projected 2.62 ERA/2.78 FIP.
Yeah, Gaddis’ projections are dragged down by him being formerly a (bad) starter. You see that a lot with converted relievers, Jeff Hoffman for example gets a lot of that. We know he’s in a completely different role, but the computer just sees numbers, not context.
“they got stupid lucky on BABIP.”
That is quite a claim to make about the team that had the fifth-worst figure in the sport. I guess they were less unlucky than they were in 2023, when they were maybe the most unlucky ever.
Pitcher BABIP, not hitter BABIP. 11th in BABIP with a RF in CF, a bad RF, a DH at 2B, a DH and converted catcher at 1B at different points, and a midseason trade to address 3B defense? A popup rate just barely behind the league leaders? Their best defensive player being their Catcher, and very low defense outside of that? No way that they got anything BUT lucky.
Okay, that makes a lot more sense.
Still, removing catchers from the equation, they were something like five runs below average by Def. Judge is a plus RF, so he can play an acceptable, if bad, CF. Gleyber’s most obvious issue is errors, which wouldn’t inflate BABIP. They upgraded 3B because they weren’t getting offense – DJLM and Oswaldo did well defensively. Also, relievers who suppress contact quality have been kind of a specialty of theirs since Matt Blake arrived.
Don’t be so sure that Siri was given away for nothing the Rays have fixed plenty of relievers with interesting traits like Orze has. I agree that the Mets offense is shallow and I think they’ll get the badly needed Alvarez breakout and I’m not throwing in the towel in McNeil just yet. Jeff said he changed his approach hitting for damage over just making contact in the second half, and the results showed as he had an over .900 OPS over a couple months before getting hurt. I don’t believe in Baty or Mauricio so bringing Pete back on a short term deal with opt out would be very helpful.
I like rotation more than most as I believe in Manaea’s changes and results in the second half and if they get 20+ starts from Senga they’ll have a top of the rotation guy I think Sproat will be up in late May or June. The bullpen looks pretty good especially if Nunez and Minter can stay healthy. I have the team as of now at around 90 wins and Stearns will tinker with the roster as you said as the season goes along.
I’ll differ with you on Mauricio, who I think is a good post-hype breakout candidate as a hitter if he can find a position in the big leagues. But if Alonso returns, the Mets won’t have a lot of at-bats to give him.
And now they don’t have at bats to give him but honestly after a missed season he should start the season in Syracuse.
The NL West is the only division without an obviously competitive division race, and still has three additional teams with at least a 25% chance of getting a wild card.
I don’t think it’s that surprising that the central divisions are this mediocre. This is where most of the teams with revenue problems reside, and the three that are in decent markets are retooling after their team struggled the last couple of years (Cardinals), seem to be incapable of making the most of their resources (Cubs), or a incompetently run (White Sox).
Detroit’s market is much bigger than STL.4.3M in Metro Detroit, STL has 2.7M. (Minnesota is at 3.7M. The other 5 teams are all lower than 2.5M). Detroit is 14th biggest metro area in the US.
& when they win, the fans WILL show out. They were top half of the AL from 2006-2015 & had at least 2.7M every year. Attendance sucked from 2018 until last year because the team sucked,
& for all of the talk of Detroit itself, the suburbs have $$, esp Oakland & Washtenaw county. GM, Ford, Chrysler employ a lot of white collar folks & there is a bunch of legacy money from the auto industry, too..
The Tigers probably have more resources than most of the central teams. The big difference between the Cardinals and other metro areas (even larger ones) is that they built a huge fanbase through decades of decent-to-good management. It’s a small metro area but they’ve made their own market to some extent. They have better resources than the Tigers for that reason. Despite being in a much smaller area. The estimated TV deals back in 2020 had the Cardinals at $65M and the Tigers at $64M and they had the 7th most attendance in MLB (35.8K) versus 24th for the Tigers (23.8K).
Even then, the Cardinals are probably only middle of the pack in terms of resources. They’re very likely below the Astros, Nationals, Mariners, and Rangers in terms of resources, much less the Braves and Phillies. But compared to the Pirates, Guardians, Reds, Royals, Twins, and Brewers they’re in much better shape. The Brewers and Marlins have always had the worst TV deals in MLB, and the others aren’t great either. And it’s probably worse now with all the deals getting renegotiated.
Totally understood. You’re right, STL was pretty much a model org for close to 40 years. That makes a difference. I do think a large chunk of that is Detroit tanked in 2017 & was god awful (110+ loss awful) in 2018-19 & was bad in 2020-23 as the rebuild petered out.
Guess what I am getting at it is there is no reason Detroit shouldn’t be able to spend more than Cleveland and KC if they want to & the market is even 15% bigger than Minnesota. & they have shown they WILL spend if the team is good enough.
I don’t consider them a small market like KC, Cle, MKE, Pittsburgh, Cincy, etc. More a mid-market team that can spend like STL, Seattle, Arizona, Colorado (if they’re ever good), SD, etc. when it makes sense. & yes, behind the big market teams.
The Cardinals have a good combination of longstanding success and geographic good fortune. They have a fairly wide regional fanbase, while teams like Cleveland, Cincinnati, and Pittsburgh are more boxed in by each other.
It’s also helpful for the Cardinals that the Blues are the only other Big 4 pro sports team in St. Louis. I’d say it’s particularly good for the Cardinals’ continued fan interest / revenue outlook that there isn’t an NFL team in St. Louis.
The Cubs are, going by outside estimates, a revenue powerhouse that probably ranks #3 or #4 in team revenue. The only clearly higher revenue teams are the Yankees and Dodgers. Cubs are in a revenue tier with the Red Sox (and maybe Blue Jays, with the Rogers broadcasting angle) that’s probably above Phillies, Braves, Giants, Rangers, and Astros. So the Cubs look like a team that could comfortably raise payroll above the level where they’ve been operating.
The White Sox have problems with how they’ve been run from both a baseball and business perspective, but I wouldn’t say their market situation is that good. They’re the longstanding #2 team in what’s now the smallest two-team MLB market (since the A’s left the Bay Area). I find it telling that none of the other Big 4 pro sports leagues have two teams in Chicago (and haven’t since the NFL’s Chicago Cardinals left in 1959).
The attendance history of the White Sox is that they’ve gotten relatively short-lived pops in periods of success, such as after the 2005 WS win. And the White Sox have still trailed losing Cubs teams in attendance during those stretches. https://www.baseball-reference.com/teams/CHW/attend.shtml and https://www.baseball-reference.com/teams/CHC/attend.shtml
Speaking anecdotally, my impression is that the White Sox also have a very limited regional fan base outside their (smaller than the Cubs) share of Chicagoland fans. I grew up in downstate Illinois, and the baseball fan allegiance question there is (and has long been) “Cubs or Cardinals?” The White Sox are an afterthought.
The size of Chicago was possibly a revenue boon for the White Sox when the RSN model was going strong, but the time has passed when a team could count on strong RSN carriage revenue solely because it plays in a large population market. The late 2024 launch of Chicago Sports Network (carrying Blackhawks, Bulls, and White Sox) as a successor to the now defunct NBC Sports Chicago does not seem to be going well, to put it mildly.
Add it all up, and I don’t think that the White Sox market position is meaningfully (if any) better than the AL/NL Central teams that are sole teams in smaller markets.
Yeah I’m looking it up again and the White Sox’s TV deal tanked. It was estimated to have something like the #7 TV deal in 2020 and now probably is like at #17.
If you look at TV money only, the order is probably:
Tier 1: Dodgers and Yankees
Tier 2: Cubs, Red Sox, Angels, Braves, Phillies, and Blue Jays (their TV money isn’t public, but the org is fully vertically integrated. they print money).
Tier 3: Mets, Giants, Mariners, Rangers
And then a huge number of teams between about $60M and $80M in TV money. The Astros, Nationals, Orioles, Cardinals, and even the Athletics fall into this bucket. The Tigers and White Sox are at the lower end of this range, with the Reds and Pirates below it, the Royals quite a bit below that, and the Brewers even lower than that.
The Mets tv deal is pretty terrible— one of the worst tv deals in the game due to the Wilpons at the time owning the team and SNY before regional sports networks fell apart.
The White Sox have apparently self-immolated their TV deal. The network that broadcast White Sox, Bulls and Black Hawks games disappeared after the last game of the White Sox season. It was replaced by one owned by the teams or Reinsdorf or something. It’s apparently available for free over-the-air in Chicagoland, but it’s been unable to make a deal with YouTubeTV or other streamers. Here in Iowa, where it was the only MLB team I got, it’s not possible to watch. Great marketing job, guys!
The Cardinals were absolutely terrible, but they’d still be in Chicago if the NFL had figured out its TV contract problems earlier and wasn’t terrified of the AFL putting a team in St. Louis.
White Sox fans attend games when the team is good, which you can see from the four (would be five but for the 1994 strike), straight years of 2MM+ attendance from 1990 – 1993, and seven straight years from 2005 – 2011, but the stadium isn’t a tourist attraction located in the middle of the city’s primary party district, and they’ve been terrible (minus COVID years) for decade and a half now with no end in sight.
The Sox also do have a regional fanbase, but it’s in Northwest Indiana, not downstate Illinois.
The Nationals and Orioles technically share a smaller two-team market according to the MSA categorizations.
I’m in complete agreement with the model on Cleveland.
Their starting pitching is:
-Bibee
-Half a year of Bieber coming off injury
-And a lot of bubble gum and string
Their offense is:
-Jose Ramirez
-Steven Kwan
-And a whole lot of “league average would be a good year for this guy”
The bullpen should be excellent, but this team feels like it’s on the knife’s edge to me.
and like others have said, it’s reasonable to project the bullpen to be excellent, but the combination of the workload these pitchers had in 2024 and the inherent volatility of bullpen performance year-over-year makes it even more precarious on the knife’s edge.
Bullpens are inconsistent. But the Cleveland bullpen wasn’t overworked last year (at least during the regular season). Sure they were 8th in innings pitched but because they were so good they were only 17th in batters faced and 14th in pitches thrown.
I’m more concerned about the heavy hitters (Clase, Smith, Herrin and Gaddis) throwing 300 innings between them, many of them important and medium-to-high leverage. Their effectiveness waned in the ALCS, and hopefully it doesn’t lead to decreased effectiveness, fatigue or injury in 2025.
To me, KC is the team that will backslide.
Bobby Witt likely just had his career year. Even 8 WAR is a 2 WAR drop.
& Lugo had a career year, too. Ragans & Wacha, 2 guys with injury concerns, had none in 2024.
Perez had a bounceback last year, too.
Feels like they were an “everything went right”/85th percentile of expectations last year & if 1-2 of the 5 guys I mentioned above gets hurt/drops in performance, they’ll disappoint.
100% agree. The Royals do have a bit of room to get worse, seeing as their Pythagorean record was a quite good 91-71, but I can’t see any way they get better.
Biggest thing for KC is that their pitching stayed healthy all last year in a way that it’s highly improbable to do twice in a row. Their front 5 starters combined to start 151 games last season, and they traded Singer from that group. It’s just too iffy of a depth chart behind Ragans, Lugo, and Wacha, and if any of the three goes down it gets uncomfortable pretty quickly.
It is really hard to see how they would improve from last year, at the minimum. The bullpen should be better to start the year but there’s going to be some regression from the rest of the team, if only due to injuries.
Interestingly the Royals are projected to score more runs than the Twins. I would be willing to bet a significant sum against that happening.
Cleveland and Detroit have some similarities, with strong bullpens and relying heavily on about 7 players on the rest of the roster. The Tigers are basically Skubal, Flaherty, Olson, Greene, Carpenter, Parker Meadows, and Colt Keith. I’m not buying Gleyber Torres or Jake Rogers. For the Guardians, it is Ramirez, Kwan, Bibee, Gavin Williams (watch for him), and Bo Naylor (also a believer here). I’m not in on Lane Thomas or Brayan Rocchio or Carlos Santana, although maybe Manzardo will be good sooner rather than later.
But they’re helped out a bit by the Royals basically having the same problems without the excellent bullpen, and by the Twins’ very deep and extremely injured team.
Eventually the Twins are bound to roll a 20 on the injury die. Maybe this is the year.
Yeah, if they ever got a year where Buxton, Correa, & Lewis all played 140 games & the pitching was relatively healthy, they would likely win 95 games.
Whoa, let’s not go crazy! I’d like to see them make it through one game without any of those guys getting hurt. That’d be an improvement from last year. Two of the three giving them 140 healthy games would be enough. Correa says his feet feel great, so that’s a good start I guess?
Yeah, but for Cleveland it sure seems like you had trouble filling out the list of 7. Ramirez and Kwan are stars, Bibee is solid, and literally everyone else is “hoping for upside”.
Seems more like “obvious bounceback” or “health bounceback” to me. Williams and Naylor both showed they can ball out in their rookie campaigns and Williams spent half the year injured. Upside is more like hoping Rocchio can tap into the power he showed in the postseason, hoping Lane Thomas becomes more like his 2023 self, or betting on Ben Lively to take a second late-career jump. The issue I have with your interpretation is that you seem to be all-in on the negatives from last season and few to none of the positives.
Yeah I guess I only listed 5 there, didn’t I? Same sort of construction, although the bullpen is better.
I think you can have confidence in Fry as a 6th easily. Dude hit for a 129 wRC+ through a shoulder injury (yes he will miss some time this year, but he’s still valuable). Clase is 7th. 1.7 WAR is the lowest number he put up in a full season. Manzardo should also be someone you can have confidence in. He hit a 152 wRC+ at home but only 54 away. Taking a steep over on his away numbers is logical. Even if you’d drop his home numbers by a reasonable amount, that’s still a good hitter.
Add in that you’ve got DeLauter and Bazzana likely making their debuts and they don’t look nearly as bad as some are saying. It’s kind of comical that some people are taking one look at base-runs and deciding that everything must follow that projection.
Dan’s write up on Cleveland indicated that ZiPS had this team in the 85-88 win range if I remember correctly. His Bregman article yesterday suggested ZiPS had Cleveland’s playoff odds at a coin flip, best in the division at the moment. If the projected standings and playoff odds are based on a 50/50 mix of ZiPS and Steamer, how is a 77 projected win total possible?
Just checked. Yes, he said 85-88. I thought maybe he ran the projections before the various trades. But he did it after. So year, something is definitely off here.
The NL is projected to be slightly better overall. I keep coming back to that thing from 20 years ago somewhere that concluded the AL’s advantage at the time was primarycausally because of the DH.
Yeah, really down on Frazier. Not in the least because the Pirates have like 5 utility infielders. I was really expecting them to sign Kepler, just like everybody knew the Rays would sign Barry Bonds in 2008.
I was about to list some things that the NL Central teams have going for them but that’s part of why all five have some amorphous on the cusp but never there hype around them. It’s more of a frustrating than a bad division, but then again with spending that’s its own problem too.
I do think the Dodgers are not all-time supermen and more like the 2000s Yankees. There’s fewer top tier teams than in previous decades, and the annual contenders that are fading do so much slower.
NL Central gets somewhat of a bad rap as they had a 84-76 record vs the NL East and a 124-106 record versus the AL. Yes, the NL west had a 86-74 advantage, but its not like the rest of baseball is beating up on the NL Central. There is something to be said for not have a doormat to play 13 games against like the other divisions sans the AL East. MIL won 93 games last year in a division that projects almost the same as it did entering 2024, 6 wins separate 1st and 5th. From a competitive viewpoint having 5 teams with a shot instead of 2 can be attractive.
I think the bad rap is more related to any one of the Brewers or Cubs could spend another $20 million and significantly increase their odds of winning that division but they don’t because they don’t have to. They may not have a doormat in the division but they have two bottom 10 teams, which is only rivaled by the NL East where it can be easily argued that’s due to 3 playoff teams (borderline 3 contenders) beating up on the Marlins and Nats.
A big difference I see for the current Dodgers vs. the mid-2000’s Yankees is the amount of depth that the Dodgers have, both on the MLB roster and with a highly ranked farm system.
The poor Dodgers don’t have Carl Pavano though smh
No I think they just signed him for 3/40 to be their 13th starter
I realize this is probably a very stupid question, but I don’t understand this: “Outscoring your opponent by 50 runs, 650-600, is likely to lead to more wins than outscoring them 850-800. … [P]laying a lot of 6-5 games gives you less margin than playing a lot of 4-3 games.”
I feel like I am missing something/just fundamentally not understanding something obvious, but why isn’t the only pertinent thing the net run scoring? If a team is +50 runs on a season, why would it matter if they outscore 300-250, 700-650, or 1,600-1,550?
Think of it this way: if you outscored your opponents 50-0, you’d never lose. If you outscored them 10,050 to 10,000, you’d probably be quite close to .500. The differential doesn’t matter — it’s the percentage of the runs scored in the game that matters. At 650-600, you’re scoring 52% of the runs. At 850-800, you’re scoring 51.5% of the runs. The farther from 50% that gets, the better your winning percentage will be.
The unstated premise here is of course that all these runs are being distributed randomly across the 162-game season. And this is definitely not 100% true (for all the reasons that we know teams can systematically outperform or underperform their run differential) but it’s good enough to a first approximation.
Thanks Ben! I follow now.
Ah yes, the fruitless game of trying to predict the season has come again so I think I will join in. The team that I think has the best chance to surprise is the Red Sox. The value of an ace cannot be overestimated and Crochet brings some of the best stuff in the game to Fenway. Houck can get major league hitters out, Bello is still improving while Buehler, Giolito and the rest should be able to cobble together a decent rotation. If, and of course all rookies carry a big “IF”, Campbell and Anthony approach their hype then the team looks much better than a team that won 81 games in 2024. Bullpens are bullpens but at least I won’t have to watch Luis Garcia and Lucas Sims any more. There should be close races in several divisions and I will be watching to see if the White Sox, whose roster looks even worse than last year, can lose 120 again or if the Dodgers might win 120. I will enjoy the exploits of Duran, Witt, Freeman, and my favorite non-Red Sox player, Jose Ramirez along with all the others who make baseball the great game that it is. Opening Day can’t arrive soon enough!
I “get” that the Dodgers are projected to have the best record in the league (by a meaningful margin), but it still doesn’t “feel right’ to project them to win fewer games than they won last season…
A 97 win projection feels both super crazy and somehow not strong enough?
Let me just say that I hope i can find a place to bet the over if that number shows up which, of course, it never will.
“Signing Corbin Burnes was obviously a coup, and adding Josh Naylor should keep the offense humming even after losing Joc Pederson in free agency.”
They also lost Christian Walker.
The only place that Josh Naylor outperforms Christian Walker is on the scale. This is a large down move for the Snakes.
I’d love to better understand this comment: “In terms of offense, though, we have them tied with the Guardians for sixth-worst in the majors. It’s the same story in Seattle, and even with many of the same characters — no real offseason additions, few meaningful subtractions.”
The Mariners have finished 8th, 6th, and 12th in MLB in wRC+ the last three years, and as noted they have basically the same roster back. If this is just a comment about raw runs scored, then sure, because T-Mobile kills offense, but that seems like a far too simple bit of analysis. I know the narrative about the Mariners having a terrible offense is all-powerful, but it’s also non-factual.
Fair comment. Their offense is not as bad as the surface stats, and their pitching is not as good as the surface stats. Tmobile (Statcast 3-year park factor of 91) is just about as extreme an environment as Coors (112). Every other park is between 96 and 107.
Yeh, that line made no sense.
People perceive the Mariners to have a bad offense because of their ballpark which makes them look awful at times. But even with T-Mobile, they were 21st in runs scored and wOBA. Not good, but not awful. And then 12th in wRC+ and Off. The had a league average offense last year and get a full year of Arozarena and Robles and presumably some bounce back from J-Rod.
I can imagine projections wind up with some regression but not sure how this isn’t a league average offense again. Depth chart show the 10th in Bat, which I assume includes defense as well.
Looking at the projected standings, something seems off with the run scoring environment. The median teams scored 700 runs last year, 4.32 per game. Projections have a median close to 4.5. That’s a huge jump across 2400 games.