Gerry McDonald: Jay Harry just keeps on hitting and his strikeout numbers have been better in AAA. What stops him from being a regular major leaguer?
12:02
Dan Szymborski: I imagine he *will* get time in 2027, should he keep hitting early next year
12:02
opifijikl: Have you ever tracked how ZiPS player comps stick or change through the years? Do you think around 50% would stay the same and 50% would change?
12:03
Dan Szymborski: Oh, they do generally change quite a bit, because no two players are truly the same, and paths vary a lot
12:03
motown oden: Hey Dan! In your milestones article, Gerrit Cole’s projections seemed a bit odd. It has him winning 30 more games, but fanning less than 400 more batters. In his career, recent and otherwise, he’s at about 1.5 K : win. Even in their 40’s, Scherzer and Verlander didn’t completely crater regarding strikeouts. Not that he’s necessarily their peers.
If you’ve been watching the Mariners the last couple of nights, you might feel like you’re back in the halcyon days of 2025. Seattle took the first two games of its series against one of baseball’s best teams, the Phillies, and Cal Raleigh launched four home runs over two nights, including a three-dinger outburst on Monday. Unfortunately, that feeling has been fleeting in 2026, as the Mariners are comfortably below .500 and have only remained in the divisional and Wild Card races because their competition has been exuberantly mediocre. As for Raleigh, after hitting 60 home runs and just missing out on the American League MVP award to Aaron Judge, he’s suffered through easily the worst season of his professional career. Raleigh’s collapse this year has been historic in its magnitude, yet both he and the M’s could still come out of this playing October baseball.
When you see a season like Raleigh’s 2025, it’s natural to expect some kind of regression toward the mean. Indeed, it’s so expected a phenomenon that a whole panoply of curses have been coined to explain it without having to get all mathy about it. Did you just have a horrible season after a great one? Well, you may have been cursed by the Sports Illustrated cover, the Madden NFL cover, a Wheaties box, a Heisman award, a sophomore slump, or an MLB the Show cover. Or maybe you just drank Jobu’s rum. Whatever you call it, Raleigh’s 2026 is well past regression to the mean and more like regression to the really frigging mean. I swear, that’s a technical term.
So just how bad a season has Raleigh had compared to expectations? You could probably just say “Austin Hedges has a wRC+ more than 20 points higher than Raleigh.” Or maybe graph Raleigh’s year-to-year home run totals, with the chart properly labeled so that people don’t think they’re viewing a win expectancy graph from a game where late-period Craig Kimbrel was the closer. But since we have a projection system handy, we can do better than that. Read the rest of this entry »
Eric Hartline, William Liang, and Vincent Carchietta-Imagn Images
Jacob deGrom tallied his 2,000th strikeout last week, providing another late-career achievement that might be engraved on a Cooperstown plaque sometime in the 2030s. That deGrom would reach 2,000 strikeouts seemed almost certain back at the end of the 2020 season, when he was 641 away. At that time, he’d struck out at least 200 batters in four of his six non-pandemic seasons. The first of those two other seasons was his 2014 NL Rookie of the Year campaign, when he didn’t debut until May 15, and in 2016, injuries limited him to just 24 starts. But since 2020, deGrom has been an example of Zeno’s paradox, seeming perpetually stuck in medias res. Achilles did eventually catch the tortoise, and deGrom and calculus won the day.
deGrom’s journey to 2,000 is a good reminder of why I like to occasionally harass ZiPS for milestone probabilities rather than simply dividing a player’s remaining distance by his most recent seasonal number and calling it a day. Whether or not there are infinite universes where every possibility happens, in this one, tendons tear and players age. A player can be on pace to finish with 500 home runs for a decade and never come particularly close, while another can sneak in the back door to a milestone that nobody saw coming, as Nick Markakis threatened to do late in his career.
The uncertainty is what makes baseball (and life) interesting. I can tell you that Aaron Judge needs 115 homers to make it to number 500; I have a calculator for that, and even if I didn’t, I was once a third-grader. But the road is hard to map, and the aging curve, performance models, and injury models in ZiPS give us a better idea as to the probability that Judge will hit 500 or 600 in this particular universe. It’s no fun to have such a tool and not deploy it, so it’s a good time to do an update on baseball’s search for pretty, round numbers. Read the rest of this entry »
For the New York Mets, the 2026 season has been an almost unmitigated disaster. Widely expected to compete for the NL East crown, the Mets got off to a brief hot start. And by brief, I mean brief, as the season’s high point was knocking out Paul Skenes in the first inning on Opening Day against the Pirates. From there, the season unraveled like the finances of Mike Campbell from Ernest Hemingway’s The Sun Also Rises: Gradually and then suddenly.
The Mets ambled around the .500 mark for another week or so, until a 12-game losing streak knocked them firmly into the league basement, where they have mostly stayed. By the time the calendar flipped to August, they had long come to terms with the idea that they would be sellers at the deadline. Recently, though, the Mets have gone on a tear, winning 11 of their 15 games since their deadline fire sale, giving some fans an excuse to dream on a miracle wild card run. That almost recently won’t happen, but their abysmal season doesn’t have to signify the start of a long rebuild. In fact, New York teams with massive payrolls aren’t generally extended much rope, with the hope being that the club will find its way back to contention quickly. Can the Mets do that in 2027? Read the rest of this entry »
ThisIsBob: Not going to lie I expected Baez 2.0 with PCA. Are there any examples of players that dramatically improved their plate discipline and increased their walk rates like PCA has?
12:00
Dan Szymborski: Morandini
12:01
Dan Szymborski: Votto wasn’t born the plate discipline god either
12:03
Dan Szymborski: Jose Bautista too, but it could also be the rushed nature of being rule 5
12:03
Dan Szymborski: I would suspect Sosa too, but we don’t really have out-of-zone data for him
Want to build a projection system? There are a lot of things you want to prepare for ahead of time, like fans thinking that you have a secret vendetta against their favorite team. The most practically important thing to know is that you are going to be wrong. Not just occasionally, but all the time. And assuring people afterwards that yes, 10% of players are supposed to fail to even meet their 10th-percentile projection isn’t going to get you off the hook, no matter how technically correct you are. Being wrong, however, is your opportunity to improve your model. Indeed, models learn a lot more from being wrong than being right.
When it comes to projecting the number of wins for a team, it mostly comes down to two big factors: how good the players are and who actually plays. Predicting the future is an incredibly difficult activity, and that is reflected in the variance in the probability “buckets.” ZiPS entered the season highly confident that Paul Skenes was a darn good pitcher, but even with that belief, ZiPS was only 50% sure that his 2026 ERA would fall between 2.37 and 3.20, and only 80% sure that it would land between 1.80 and 3.65. But as hard as it is to project how effective a player will be, that’s still easy compared to predicting who will play. Models can get a general idea of the likelihood of injuries from some specifics, like whether a player has missed a lot of time recently or just had Tommy John surgery. But what they don’t know, and certainly never will, is who will tweak their hamstring next Thursday and be out for the rest of the season, who will suddenly lose their velocity in their Labor Day start and need a UCL repaired, or who will be abducted by giant mutant ants and forced to toil in underground sugar caves.
So, which teams have suffered the most due to injury in 2026? If you ask fans, especially those of underperforming clubs, they’ll tell you that their favorite team has been uniquely put upon by injuries this season. But we’re numbers people here. We can do a bit better than that and see which team’s fans can justify their sad laments about the trials and tribulations of the hometown 26. Read the rest of this entry »
The worst-case scenario for Hunter Greene’s elbow has come to pass. The ace right-hander had Tommy John surgery on Thursday, ending his 2026 season and almost certainly knocking him out for the entire 2027 season, too. This was Greene’s second elbow procedure this season; a previous arthroscopic surgery to remove loose bodies delayed his seasonal debut until Independence Day. Greene’s 2026 season wraps up with a 6.83 ERA, but a better 3.55 FIP over 27 2/3 innings in five starts. This is a full Tommy John revision — Greene previously had Tommy John surgery in 2019 — and not the internal brace procedure that tends to have a quicker recovery.
Suffice it to say, this is a major blow for both Greene and the Cincinnati Reds. For Greene, it’s the latest (and scariest) injury setback in a professional career that has never seen him throw enough innings to qualify for an ERA title. There comes a point at which one starts to wonder if he has crossed the proverbial Rubicon, the point where teams now can’t realistically tell themselves that they can keep him healthy as a starting pitcher. His fastball-splitter-slider repertoire would almost certainly play nicely in relief, though he’ll be hard-pressed to make the same impact as a closer as he could as a legitimate frontline starter. Read the rest of this entry »
With the trade deadline having come and gone, we’ve officially entered the homestretch of the 2026 season. And while teams surely want to avoid tempting the fury of the fates by assuming they’re postseason-bound, they’d be lying if they said they weren’t thinking about October baseball at least a little bit. And why shouldn’t they? Winning the World Series is the ultimate goal, and many of the league’s playoff-relevant teams made deadline moves for the express purpose of improving their postseason chances.
Before you set out on a vacation, it’s natural, even prudent, to contemplate your trip, though it might mean making plans for a destination you never reach. If you’re heading to the beach, you’re likely going to think about the sun and the sand, even though there’s always a chance the alternator in your 1988 Ford Tempo dies on U.S. Route 50 near Easton, Maryland, forcing you to call your friend Alan on your janky 1990s Motorola “car phone,” only for him to gormlessly drive right past you despite your hazard lights and frantic waving. Baseball has its own broken alternators, injuries and losing streaks that render October plans suddenly irrelevant. Still, teams have good reason to think about what will happen if they get there; after all, it’s important that someone remember to pack sunscreen. And since we’re nerdy baseball fans and have access to a projection system, we might as well look ahead too. So which teams are actually built to be the most dangerous in October, and which ones might be better or worse postseason teams than their regular-season records suggest?
When it comes to projecting October, we have to think about teams somewhat differently than we do during the regular season. Projecting the regular season and a playoff series are two very different tasks. Not because of any playoff-related mythology, or because a team possesses specific characteristics that work better or worse in the postseason, but simply because teams use players very differently in October. It’s not quite as extreme as, say, the difference between how 100m gold medalists and 500m gold medalists are built, but it at least rhymes. In the playoffs, you get to ride your top talent hard since you no longer need to keep them fresh and healthy for a six-month marathon. And because of the postseason’s frequent off days, you also get to have a higher proportion of your innings pitched by the pitchers you actually want in the game. Read the rest of this entry »
Dan Szymborski: To every thing there is a season, and a time to every purpose under the heaven: Except maybe SzymChat.
12:02
war2d2: Dan! Good news! I just realized you can sing your name to the tune of “Black Betty” by Ram Jam. Do with this information what you will.
12:04
Dan Szymborski: I’m not sure if the lyric “Szymborski had a child bam ba lam” really works
12:04
Chandler White: How does Zips react to a player repeating a minor league level and performing much better? The general notion is that it’s expected from potential major leaguers but was unsure if Zips is only factoring in age and other league factors or if there was something built into the algo to reference a repeat in level.
12:04
Dan Szymborski: There’s a bit of a penalty for repeating after a previous stint of at least 1/2 a season
12:05
war2d2: I felt the Cubs did an okay job at the deadline, addressing the needs they had with decent options (ZiPS seems to agree, judging by your piece on the deadline). Obviously I’m disappointed they didn’t land Skubal, as I think anyone would be. You had said previously that, from a playoff odds perspective, Skubal made the most impact on the Cubs’ chances out of anyone. Would he have made more of an impact alone than the package of Gausman et al did in aggregate?
It looked like it was going to be a quiet deadline for a while, but the floodgate opened over the weekend, and we got a lot of fun moves, including a deal involving two-time reigning Cy Young winner Tarik Skubal. In the end, it feels better than last year’s deadline, when every team seemed preoccupied mainly with shoring up its bullpen. Now, there were certainly a lot of relievers shifting around, but plenty of starting pitchers and position players were also traded, and variety is the spice of deadline life.
Deciding the victor and the goat of the trade deadline is a vexing philosophical exercise, as teams come into swapping season with different goals. That makes it hard to distill deadline performance into a single, all-purpose number. But since distillation is a fun process — we would not be blessed with bourbon without it — let’s narrow down the question to something a projection system can get its virtual mind around. I focused on the 2026 part of the equation to cover the ZiPS portion of the deadline: Which teams helped their 2026 chances the most? Read the rest of this entry »
Dan Szymborski: The appointed time has arrived.
