For the ninth time (in 10 years — it’s a long story), we’ve reached the point in the offseason where I run down the ZiPS Top 100 prospects. For those wandering in who may hear “ZiPS” and think of the University of Akron or possibly the popular Cincinnati burger spot, ZiPS is a computer projection system that crunches a lot of data about players and attempts to peer through the fog that obscures the future. You can read more about the system here or in MLB.com’s executive summary.

ZiPS prospect projections aren’t an attempt to supplant scouting. Rather, they try to be a supplement to scout-generated lists. There’s a lot of uncertainty in lower-level minor league stats that isn’t present at the upper levels. As such, non-statistical information about players takes on added value. ZiPS doesn’t seek to be the one-ring-to-bind-them-all-unified-field-theory-giant-Katamari-Damacy-ball of prognostication; it aims to give the very best data-generated predictions possible, for people to use, ignore, mock, or worship according to their personal tastes and worldview. Read the rest of this entry »

In a winter that didn’t lack for shocking free agent moves — and even more shocking plot twists — the Rangers’ signing of Jacob deGrom was as eye-opening as any of them. After opting out despite two injury-shortened seasons, the 34-year-old righty left the Mets to sign a five-year, $185 million deal with the Rangers. The move was surprising not only because deGrom bolted from the team with which he’d spent his entire professional career, but also because he chose a club that won 68 games in 2022 over one that won 101.

The Rangers didn’t stop at adding deGrom when it came to addressing a rotation that was among the majors’ worst last year, ranking among the bottom seven in ERA, FIP, and WAR. In addition to retaining All-Star Martín Pérez and trading for Jake Odorizzi, they added both Nathan Eovaldi and Andrew Heaney via two-year free agent deals of $30 million and $20 million, respectively. Thanks to those additions plus holdovers Pérez and Jon Gray, Texas’ rotation ranks second in projected WAR via our Depth Charts, behind only the Yankees (who got some bad news on Thursday, as Carlos Rodón has suffered a mild forearm strain, cutting into the depth already compromised by Montas’ injury). The Rangers lead the majors when I compare the actual production of last year’s rotations to the projected production of this year’s units — though not without some caveats. Read the rest of this entry »

Remember the 107-win Giants, the only team to take an NL West flag away from the Dodgers over the past decade? Last year’s crash to an 81-81 record marked a rude awakening, as the post-Buster Posey era got off to a rocky start following the future Hall of Famer’s abrupt retirement after the 2021 season. Now the Giants enter the post-Brandon Belt era as well, as the regular first baseman from their last two championship teams joined the Blue Jays in January.

I mention the Giants because in this National League counterpart to my examination of the weakest spots on American League contending teams, their catching and first base options occupy the first two spots, as they did little to address either position over the winter — a point that’s been overshadowed by the Padres’ spending spree and the Dodgers’ free agent exodus (a phrase I’ve now used for three straight articles and expect to deploy at this year’s Passover seder). The Giants are projected to finish third in the NL West with 84 wins, and while they do have a 41.8% chance of making the playoffs, improvements in both spots could have bumped them up by a good 20 percentage points.

As with the AL installment, here I’m considering teams with Playoff Odds of at least 10% as contenders; that threshold describes 10 NL clubs (all but the Cubs, Pirates, Reds, Rockies, and Nationals). It’s worth noting that because of the general tendency to overproject playing time and keep even the weakest teams with positive WAR at each position (in reality over 10% of them will finish in the red), our Depth Chart values at the team level are inflated by about 20%. That is to say, instead of having a total of 1,000 WAR projected across the 30 teams, we have about 1,200. Thus, I am discounting the team values that you see on the Depth Chart pages by 20%, and focusing on the lowest-ranked contenders among those whose adjusted values fall below 2.0 WAR, the general equivalent of average play across a full season. The individual WARs cited will remain as they are on the Depth Chart pages, however, and it’s worth noting that many of the players here — particularly youngsters with shorter track records — don’t project particularly well but aren’t without upside; hope springs eternal. Read the rest of this entry »

As Gavin Lux’s season-ending knee injury reminded us, the Dodgers had quite the free agent exodus during the offseason. Nearly two months ago, they led the pack when Ben Clemens examined which teams were at the extremes in terms of production lost and gained via free agency. With all but a few free agents of note now signed, I thought it would be worth circling back to those rankings before moving on to highlight some notable holes on contending teams, something of an offseason version of my Replacement Level Killers series.

When Ben checked in just after the New Year, the Dodgers were runaway leaders in terms of net WAR lost to free agency, having already parted ways with Tyler Anderson, Cody Bellinger, Justin Turner, Trea Turner and a dozen other players in an effort to trim payroll and lessen (if not eliminate) their Competitive Balance Tax burden. Towards that end, only two of their incoming free agents (J.D. Martinez and Noah Syndergaard) will make more than $10 million annually. To put it another way, the team won 111 games last year with a payroll just south of $270 million for CBT purposes, but after falling short of the National League Championship Series, they’re hoping to get as far or further by winning 80-something games and spending maybe $25 million less. Such are the vagaries of postseason baseball that it just might work. Read the rest of this entry »

On Thursday, MLB announced the 2023 schedule, implementing the alterations originally announced when the current collective bargaining agreement was signed back in March. The existing format, under which the 2022 season is being played, has been largely stable since 2013, the season the Houston Astros moved to the American League. That change evened out all six divisions to five teams each, making for a tidy format in which every team played their divisional opponents 19 times and the rest of the teams in their league six or seven times, with 20 interleague games against a rotating division and officially designated MLB rivals.

Before 2001, MLB’s schedule tended to be a good deal more balanced. During the divisional era before interleague play, six-team divisions typically played 18 games against their divisional opponents and 12 against non-divisional opponents; seven-team divisions had a nearly even 13/12 split (the American League did 15 vs. 10 or 11 for a couple years after the 1977 expansion). In 2001, MLB went all-in on an unbalanced schedule, with the idea being that by having teams play their divisional rivals more often, you’d create greater tension in the divisional races and more intense regional rivalries. Whether this approach actually accomplished its goals is difficult to tell. I can’t think of any new rivalries that were created simply by playing more games and tend to believe that rivalries are born from teams playing more meaningful games against each other, not simply from seeing each other more often. Red Sox and Yankees fans don’t appear to have hated each other any more in 2010 than they did in 2000, and the endless Orioles-Rays series in the days before Tampa Bay was competitive made this O’s fan click over to other games, not foster a hatred for the Rays.

Be that as it may, from a philosophical standpoint, heavily unbalanced schedules make the most sense when winning divisional races is the sole or at least primary way of making the playoffs and much less so when more Wild Card spots exist. When you have a lot of Wild Card spots, you create a fundamental bit of unfairness when the divisions are of meaningfully differing strengths; teams in weak divisions are competing directly against teams in stronger divisions for those Wild Card spots, with the former generally having easier schedules. Read the rest of this entry »

It arrived stressfully, chaotically, and slightly late, but the 2022 season is here. And that means it’s time for one last important sabermetric ritual: the final ZiPS projected standings that will surely come back and haunt me multiple times as the season progresses.

The methodology I’m using here isn’t identical to the one we use in our Projected Standings, so there will naturally be some important differences in the results. So how does ZiPS calculate the season? Stored within ZiPS are the first through 99th percentile projections for each player. I start by making a generalized depth chart, using our Depth Charts as an initial starting point. Since these are my curated projections, I make changes based on my personal feelings about who will receive playing time, as filtered by arbitrary whimsy my logic and reasoning. ZiPS then generates a million versions of each team in Monte Carlo fashion — the computational algorithms, that is (no one is dressing up in a tuxedo and playing baccarat like James Bond).

After that is done, ZiPS applies another set of algorithms with a generalized distribution of injury risk, which change the baseline PAs/IPs selected for each player. Of note is that higher-percentile projections already have more playing time than lower-percentile projections before this step. ZiPS then automatically “fills in” playing time from the next players on the list (proportionally) to get to a full slate of plate appearances and innings.

The result is a million different rosters for each team and an associated winning percentage for each of those million teams. After applying the new strength of schedule calculations based on the other 29 teams, I end up with the standings for each of the million seasons. This is actually much less complex than it sounds. Read the rest of this entry »

Earlier this offseason, we released our team expected win totals and playoff odds for the 2022 season. These are based upon the FanGraphs Depth Charts, which use a 50/50 blend of ZiPS and Steamer and our manually maintained playing time estimates. To arrive at the playoff odds, we then simulate the upcoming season 20,000 times, taking strength of schedule into account. (You can learn more about the FanGraphs playoff odds here.)

For the second straight year, we’ve also run the same process using the Average Total Cost (ATC) Projections as our base.

The ATC Projections have been available on the pages of FanGraphs since 2017. ATC is smart aggregation of other projections; its methodology is based on the process that Nate Silver uses with his political forecasting model over at FiveThirtyEight. Read the rest of this entry »

It arrived stressfully, chaotically, and slightly late, but the 2022 season is here. And that means it’s time for one last important sabermetric ritual: the final ZiPS projected standings that will surely come back and haunt me multiple times as the season progresses.

The methodology I’m using here isn’t identical to the one we use in our Projected Standings, so there will naturally be some important differences in the results. So how does ZiPS calculate the season? Stored within ZiPS are the first through 99th percentile projections for each player. I start by making a generalized depth chart, using our Depth Charts as an initial starting point. Since these are my curated projections, I make changes based on my personal feelings about who will receive playing time, as filtered by arbitrary whimsy my logic and reasoning. ZiPS then generates a million versions of each team in Monte Carlo fashion — the computational algorithms, that is (no one is dressing up in a tuxedo and playing baccarat like James Bond).

After that is done, ZiPS applies another set of algorithms with a generalized distribution of injury risk, which change the baseline PAs/IPs selected for each player. Of note is that higher-percentile projections already have more playing time than lower-percentile projections before this step. ZiPS then automatically “fills in” playing time from the next players on the list (proportionally) to get to a full slate of plate appearances and innings.

The result is a million different rosters for each team and an associated winning percentage for each of those million teams. After applying the new strength of schedule calculations based on the other 29 teams, I end up with the standings for each of the million seasons. This is actually much less complex than it sounds. Read the rest of this entry »

We passed the halfway mark of the 2021 season over the long holiday weekend, providing a convenient spot to take a break, look back over the preseason projections, and hopefully not cringe too much about how the predictions are shaking out. Since this is the big midseason update, I used the full-fat ZiPS model for individual players in addition to the normal depth chart reconfiguring, with all the high-fructose algorithms rather than the leaner one used for daily updates.

I went through the American League on Wednesday, so now it’s the Senior Circuit’s turn.

The Mets only averaging 89 wins in the update might feel a bit disappointing, but that negative inclination is misplaced. ZiPS actually likes the team’s talent slightly more than it did in March, with the difference being that the injury situation has been worse than expected. Use the preseason playing time predictions with the up-to-date player projections, and ZiPS believes that New York would have a 93-win roster, good enough to be the third-best team in the National League.

Read the rest of this entry »

MLB passed the halfway mark of the 2021 season over the long holiday weekend, providing a convenient spot to take a break, look back over the preseason projections, and hopefully not cringe too much about how the predictions are shaking out. Since this is the big midseason update, I used the full-fat ZiPS model for individual players in addition to the normal depth chart reconfiguring, with all the high-fructose algorithms rather than the leaner one used for daily updates.

Let’s start with the American League standings.

I was making a “do not panic” argument on behalf of the Yankees back when they were 5–10 and some people were digging for their doomsday preparedness kits, and while it might not be time to find where you left those water purification tablets, the situation is bleaker now than it was three months ago. Not that the team is actually worse; New York has been on an 88-win pace in the games since that reference point. But an 88-win pace isn’t nearly enough to get out of an early-season hole in a division where there are three other teams with more than detectable pulses. Even projected to play solid baseball the rest of the season, the Yankees have gone from the favorite to the projected fourth-place team.

Read the rest of this entry »