New York City FC 2019 Season Preview / by Dummy Run

Point-above-replacement values are explained here. Non-penalty expected goals + expected assists are explained here, and you can see all players’ xG+xA in our interactive expected goals tables. Touch percent is the percentage of total team touches by that player while he is on the field, which can be found in our interactive expected passing tables.

By Dummy Run (@thedummyrun)

Here's the weird thing about New York City Football Club in the Year of Our Lord 2019: after all the autumnal sturm und drang, the late-season slide, the second-round playoff bounce, and Domènec Torrent's desastre of a postseason presser, the club had a whole offseason to clean house and bring in Torrent's team and they just kind of ... didn't. What you'll see on opening day is more or less the same lineup you know and love (or used to love last summer but now hate with a burning passion because they lost some soccer games). Even my boy Ben Sweat's still out there, which can be a beautiful thing as long as a foul pole’s obstructing your view of the defensive third.

So, uh, what gives? Allow me to resubmit to you the possibility that Dome’s team was maybe actually not that bad at soccer last season. That in fact their +0.55 expected goal differential per game was the second best in MLS during Torrent’s tenure, a marked improvement from under Patrick Vieira. And that their best run of games came without David Villa or Yangel Herrera, this offseason’s two major departures. Maybe Claudio Reyna wasn’t crazy to think all this team needed was a few new pieces for things to fall into place.

2018 in Review

Ugh, do we have to do this again? Okay fine.

Here’s the season the way you probably remember it: NYCFC bolted out of the gate with an away win against a strong Sporting Kansas City team and for two months they never looked back. The first loss didn't come till late April, but when the second one hit on May 5, wheewww boy: a 4-0 shellacking at Red Bull Arena sent Vieira into panic mode, and he scrapped his high-pressing 4-3-3 for a few weeks of janky back-three ball until Herrera’s long term ankle injury forced yet another switch to a 4-2-3-1. But the double pivot worked! In Vieira’s last game NYCFC spanked Atlanta up and down the baseball diamond and was unlucky to only earn a draw.

Then Vieira bailed midseason to coach a club he definitely hadn’t been talking to and in came Torrent, who wasn’t shy about reminding everybody that he’s besties with Pep Guardiola. Was this an upgrade? For a while it sure looked that way, as NYCFC took off on another 5-0-1 tear while playing a 4-2-3-1 so aggressive that it shaded into a 4-2-4. With Villa injured, the team hurled bodies into the box from all angles and came away with points. But when their star striker returned, the chances mysteriously dried up. See that big ol’ dip in the blue expected-goals-for line in the chart at the top of this article? The plunge was so sudden you could schedule your #DomeOut tweets by it.

Now that I’ve drawn your attention to the rolling xG chart, I want you to take a closer look at the purple line in the middle, the one showing expected goal differential per game. This is where the popular narrative of NYCFC’s season and the #analytics take start to part ways. See, even during the team’s worst patch under Dome, the xGD never really dipped below zero. Taking and allowing equally good shots isn’t going to win you any plaudits as the second coming of Pep, but it’s also pretty much where NYCFC was hanging around for the whole first half of the season under Vieira. Crazy huh.

This is the part where we should probably talk about luck. Imagine for a second that NYCFC had spent the entire season playing perfectly average soccer: average number of shots for and against in every game, average xGF and xGA, cruising that dotted zero xGD line from March to October. (ASA’s Eliot McKinley has dubbed this hypo the Harrison Bergeron team.) At the end of 34 games their simulated point production would look something like this:

That’s a pretty wide range of outcomes! See all that blue fuzz at both ends by matchday 34? After a full season of perfectly arithmetically average games, our 0 xGD NYCFC would manage to match last season’s Atlanta United juggernaut on points about 3% of the time but would also finish level with Orlando in around 10% of the sims. Point is, there’s a lot of uncertainty in soccer. Sometimes a bad shot goes in, sometimes a good shot doesn’t, and since goals are so rare in this game a little luck can go a long way.

Of course, NYCFC wasn’t an average team: they were a good one, or so we lonely stat nerds would have you believe. From atop my pile of spreadsheets (and my season ticket perch up the first base line), the majorleaguesoccersoccer dot com party line about Dome breaking something beautiful looked like lazy box score punditry. If anything was broken, it was NYCFC’s rolling points minus expected points—what we might for want of a better word call luck.

And yet. Something had gone wrong, hadn’t it? Even on the relatively rosy xG plot up top, the September dive toward the mean is tough to ignore. And it’s more than a little worrying that the team appeared to be moving in the wrong direction the longer NYCFC was Torrent’s team. Numbers or no numbers, the dwindling camp of Dome apologists has some serious ‘splainin’ to do.

You’re not going to like this, but one problem may have been David Villa. With no disrespect to the greatest player in NYCFC’s very short history, a 36-year-old striker ain’t always what he used to be, and there were signs last season that Reyna may have made the right call by gently refusing to slide another $5.6m Villa’s way in 2019. Over two comparable seven-game samples under Torrent, NYCFC’s attack was noticeably more effective without its marquee DP, especially at creating shots close to goal. Here’s a stat to ponder: when Villa was out hurt from July 1 to August 5, Jesús Medina received six passes within 10 yards of goal, trailing only Josef Martínez over that span. Villa himself only got on the end of one such pass the entire season.

You’ll be able to read more about NYCFC’s attack with and without Villa (including video analysis!) later this week at The Outfield and should definitely follow us on Twitter at @OutfieldNYCFC. I wish I could tell you that’s the last time I’m going to promote my side project in this article but it’s just not true.

Offseason Changes (from

F - David Villa (out of contract)
M - Eloi Amagat (option declined)
M - Kwame Awuah (option declined)
M - Tommy McNamara (option declined)
GK - Andre Rawls (option declined)
M - Rodney Wallace ( option declined)
M - Yangel Herrera (out of contract)
D - Saad Abdul-Salaam (out of contract)
F - Jo Inge Berget (mutually terminated contract)
D - Cedric Hountondji (waived)

M - Tony Rocha ( trade from Orlando)
M - Keaton Parks (on loan from Benfica)
M - Justin Haak ( Homegrown)
M - Juan Pablo Torres (transfer from Lokeren)
GK - Luis Barraza (SuperDraft)
F - Alexandru Mitrita (transfer from Universitatea Craiova)

NYCFC returns around 80% of its minutes from 2018, slightly more than the league average. The big names out are Villa (20% of NYCFC’s total xG) and Herrera (19% of the team’s defensive actions while on the field); depth guys like Rodney Wallace, Tommy McNamara, and Eloi Amagat aren’t likely to be missed on the field. While Reyna was happy to consciously uncouple from Jo Inge Berget and the $800k contract he massively underperformed, the club has struggled to land a promised replacement center forward.

What they got instead is the 24-year-old Romanian attacker Alexandru Mitriţă, whose reported $8.5 million transfer fee makes him the third most expensive signing in MLS history. That’s a lot to live up to, but highlight-reel goals in each of his first two preseason games have helped to assuage some fears. As Christopher Jee wrote for The Outfield (oh hey there it is again), “If you squint and turn the Eastern European dance music up loud enough, you can sort of see what Dome’s talking about when he compares his new DP to Philippe Coutinho.”

The other big pickup is Keaton Parks, a 21-year-old USMNT prospect on loan from Benfica. There’s no shame at that age in getting edged out by Champions League players, and Keaton’s silky ball skills should earn him good marks on ASA’s expected passing model, which also happens to be the one metric where newcomer Tony Rocha stood out for Orlando last season.

Tactical Outlook

Along with technique to sustain a ball-hungry possession game, NYCFC picked its offseason targets for their positional versatility, signaling more tactical fluidity to come from the famously tinker-happy Torrent. The Outfield’s NYCFC Tactics has a good rundown of the most likely formations here, but the two basic shapes will involve either the more defensive double pivot of Ebenezer Ofori and Alex Ring with Maxi Morález roaming free as a central attacking mid or a more attacking look with Ring as a lone defensive midfielder and Morález and Keaton pushing high up the channels, Man City style. As you might have guessed from Maxi claiming three out of the four 2018 Team Leaders spots in that graphic up top, the key to success for this team is going to be finding the system that gets the most out of his enormous talent (and no I’m not going to end that sentence with a joke about his height).

The defense is pretty much the same as last year, although Rocha’s preseason emergence as a candidate at inside left back is a wrinkle worth watching. And anyway defense shmefense, right? This team is all about counterpressing high, which they’re very good at (just go ahead and ignore all that blue at the back, thank you kindly).

While the front line returns most of its young wingers, Mitri’s arrival has shaken up the tactical landscape. So far he’s preferred to line up on the left and dribble in at goal, leaving Valentín Castellanos, Jesús Medina, or Jonathan Lewis to lead the line and run the channels as an unconventional center forward. Look for plenty of rotation and fluidity from NYCFC’s attackers in both the 4-3-3 and the 4-2-3-1, at least until that mystery striker shows up.

Roster Outlook


There’s no question that recent USMNT call-up Sean Johnson is the starter here, although his average shotstopping numbers and good-but-not-great distribution make his recent contract extension at a reportedly doubled salary feel a little generous. Down the depth chart things get more interesting, as the Edersonesque draftee Luis Barraza may push Brad Stuver for backup minutes.


The starting back line of Anton Tinnerholm, Maxime Chanot, Alexander Callens, and Ben Sweat looks intact headed into the season. Tinnerholm has been particularly effective getting forward under Dome, racking up 83% of his 5.8 xG+xA in the 19 games after the coaching change. Callens drives the buildup with an 11.3% touch percentage and the second-highest total passes and pass score among center backs in 2018 (Sporting Kansas City’s Matt Besler led both), while Chanot and backup CB Sebastien Ibeagha are more notable for their defensive presence in the box. With Saad Abdul-Salaam gone to Seattle, look for 16-year-old homegrown Joe Scally to get spot minutes as a backup right back. 


Along with Ring, Ofori, and Morález in their usual roles, expect Parks to earn significant minutes as a better-passing but less defensive replacement for Herrera. Rocha, Juan Pablo Torres, and new homegrown Justin Haak will help fill out the depth chart, but the real guy to watch here is James Sands, whose comfort manhandling Wayne Rooney last season and doing stuff like this in the preseason suggest that the heralded homegrown is more than ready to graduate from NYCFC’s U-19 squad and start rotating with Ring at DM.


Ismael Tajouri-Shradi probably can’t repeat his insane 11 goals on 3.8 xG from last season, but he and fellow returning wingers Medina, Castellanos, and Lewis averaged a combined 0.52 xG+xA per 96 last season, dead level with LAFC’s heralded Young DP Diego Rossi. If Romanian former MLS defender Alex Zotinca is anywhere close to the mark with his prediction that Mitriţă will be a “Romanian Giovinco” and score 20 goals this season, this could be a pretty potent attack. On the other hand, it’d be a whole lot easier to feel good about playing tactically creative small ball if there were a conventional center forward anywhere on the roster as a fallback plan, and there are whispers back home about Mitri’s attitude—he’s got more career yellow cards than assists.

2019 Expectations

I mean, I guess at this point I’m pretty much committed to predicting a good season for NYCFC. Are they going to miss the playoffs, as some of our good friends at expect? Or are they a dark horse candidate to win the East when Frank de Boer inevitably reduces Atlanta United to rubble? Nobody knows anything! All I can tell you is NYCFC had pretty decent underlying numbers last year and spent some money (though not as much as they could have) improving the squad this offseason, and that those are about the only weak empirical predictors of success we have in this topsy turvy world. FiveThirtyEight, which is pretty good at this stuff, has NYCFC running fifth for this year’s MLS Cup. So for that matter do the bookies, which would be more comforting if third place weren’t the L.A. Galaxy, who at this point are basically just Zlatan standing with outstretched arms atop a giant tire fire, laughing maniacally as the flames consume him. I don’t know, man. I just don’t know.