Sunday, February 26, 2017

NYCFC Season Preview

As an American fan of Manchester City who was born in Albany, New York, I've been avidly following the nascent New York City FC since its inception in 2015. A team affiliated with Manchester City playing in my home state? Sign me up! However, I really think all Manchester City fans should be paying attention to them and not just us Yanks. Not only is the club a place where youth players and other young City signings can be tested (such as Angelino two years ago and Yangel Herrera this coming year), NYCFC attempts to play a similar style of football as well, focusing on attacking, possession-based football. That's before coming to the fact that the manager is one Patrick Vieira, who is probably one of the favorites to take over from Pep when the time comes. Last season, the team took a real step forward and made the playoffs, before being crushed by a far superior Toronto outfit (who somehow were unable to beat Seattle in the Final). That said, this season's squad is going to have a very different look to it, as Frank Lampard and Andoni Iraola retire after playing key roles last season. Last season's finish in the table and subsequent playoff appearance covered over some long-term issues regarding the defense in particular. This team still has a lot of attacking talent, but they're gambling that a focus on ball retention can cover up a paper-thin defense. That is difficult to achieve anywhere, but especially so in a league where the talent level is below that of the top European leagues. It might be great, it might not, but it certainly won't be boring.

What Happened Last Year

NYCFC's season can be pretty easily divided into two halves: before that Hudson River Derby and after that Hudson River Derby. That Hudson River Derby was a 7-1 shellacking at the hands of New York Red Bull(s?) which was about as bad as it could possibly get for NYCFC fans. However, that game marked the first appearance of Frank Lampard in this season and was the driver to move Iraola into the base of midfield, a role he played the rest of the season. From that point forward, NYCFC had a very steady midfield trio of Lampard, Pirlo, and Iraola to support David Villa up top, along with a rotating but talented case of Jack Harrison, Stiven Mendoza, Khiry Shelton, and Tommy McNamara on the wings. It also marked the end of Mix Diskerud's and Federico Bravo's run in the side, as they played 80% and 72% of their total minutes on the season prior to that. The results speak for themselves: up to and including that Derby, NYCFC averaged 1.3 points per game, afterwards they averaged 1.8.

That uptick in form was enough to secure a second-place finish in the Eastern Conference, but it was more than a little fortuitous. NYCFC actually had a negative xGD, mainly because their defense was terrible with the 4th worst xGA. This video sheds light on some of the reasons why: little cohesion in backline, inability to track runners, and insistence on playing from the back with personnel ill-suited for it. As in that game though, the attack usually made up for it, with Lampard and Villa in particular turning in stellar seasons. The rookie Jack Harrison (a former United trainee) provided a real creative spark on the wings, coming up with perhaps one of the goals of the season by beating several players while coming in from the right wing to curl home a beautiful finish. Ultimately though, New York's defensive shortcomings came back to bite them against the talented frontline of Giovinco and Jozy Altidore (who is actually decent in MLS and yes I know that says a lot) in the playoffs, losing 7-0 on aggregate.

The Stats

Player Total Passes Failed Passes Key Passes Assists Goals Shots Take-Ons Failed Take-Ons Minutes Poss Usage Rate Positive Outcomes Negative Outcomes % Positive xG xA (xG+xA) /Poss
Villa 860 241 42 4 23 163 26 37 2779 487 14.74% 209 278 42.92% 20.32 4.75 5.15
Mendoza 425 105 20 0 5 45 27 23 1160 193 13.99% 65 128 33.68% 3.27 2.21 2.84
Shelton 418 118 13 6 4 17 21 29 1163 183 13.23% 36 147 19.67% 2.22 2.63 2.65
Diskerud 333 84 10 2 1 20 6 7 813 123 12.72% 32 91 26.02% 1.84 1.67 2.85
Pirlo 1612 302 57 6 1 26 21 7 2675 398 12.51% 89 309 22.36% 1.64 5.04 1.68
Matarrita 992 234 14 2 1 25 27 10 1987 285 12.06% 41 244 14.39% 1.28 1.8 1.08
Harrison 475 126 25 6 4 22 36 31 1482 210 11.92% 53 157 25.24% 1.61 3.18 2.28
Lampard 521 108 18 2 12 33 5 6 1187 167 11.83% 53 114 31.74% 6.76 1.41 4.89
Allen 757 180 19 4 1 4 26 22 1671 229 11.52% 27 202 11.79% 0.24 3.51 1.64
Bravo 618 135 5 0 0 7 2 2 1112 149 11.27% 12 137 8.05% 0.48 0.25 0.49
White 250 68 0 0 0 0 1 3 598 71 9.98% 0 71 0.00% 0 0 0.00
Taylor 185 39 6 1 2 12 12 12 591 70 9.96% 19 51 27.14% 1.4 0.82 3.17
Mena 554 124 1 0 1 6 2 3 1214 134 9.28% 7 127 5.22% 0.67 0.08 0.56
McNamara 761 138 26 9 4 36 7 16 2121 225 8.92% 71 154 31.56% 4.05 4.06 3.60
Lopez 285 55 3 1 0 12 2 1 683 72 8.86% 16 56 22.22% 1.19 0.33 2.11
Brillant 1478 236 1 0 1 11 1 1 2700 249 7.75% 12 237 4.82% 0.84 0.1 0.38
Iraola 1210 166 7 1 0 3 17 9 2120 186 7.38% 11 175 5.91% 0.11 0.45 0.30
Hernandez 1099 178 6 0 0 0 2 0 2291 184 6.75% 6 178 3.26% 0 0.26 0.14
Saunders 708 173 1 0 0 0 0 0 2880 174 5.08% 1 173 0.57% 0 0.03 0.02

Above is a table of the Usage Rates for NYCFC last season, and thanks to American Soccer Analysis, I have xG and xA numbers for each player as well. What this allows us to do is not only show how heavily each player was involved in the attack, but also show how efficient each player was with the possessions he used as measured by expected goals. This works as a nice complement to my usual efficiency measure: percentage of possessions with a positive outcome. As one might have guessed, Villa had the biggest role in the attack by Usage Rate, but he also produced a hefty 5.15 goals for each 100 possessions used which is also the best on the team. Second by that measure is the retiring Lampard, at 4.89, which shows just how much he'll be missed next year. Though his goals total likely would have come down anyway (he scored 12 on just 6.76 xG; take a look at this video for some of his finishing luck), his ability to score and create while not being a focal point of the offense was critical to NYCFC last season.

Two other players whose numbers stick out are Ronald Matarrita and Iraola. I mentioned Iraola's impact before, but the stats give a clearer picture as to why his move to midfield helped NYCFC so much. His passing range was far superior to Bravo's and enabled NYCFC to build up much more effectively when installed in the deep-lying role, while also bringing better defensive positional sense. The low number of turnovers (and correspondingly low Usage Rate) and high passing volume (3rd most passes on the team) distributed the ball to Pirlo and Lampard excellently allowing them to dominate play further up the pitch. As with Lampard, he will not be easily replaced. Meanwhile, the statistics for Matarrita (and to a lesser extent RJ Allen) show just how high up the fullbacks were in the attack. His Usage Rate is comparable to Pirlo, but because he didn't use the ball very efficiently (only 14% resulted in a shot or chance created) numerous times the opposition came right back down his side. Being outnumbered on the counter was one of many defensive problems City faced last season as a result.

Squad and Salary Makeup

MLS squad construction is always tricky, see this excellent flowchart for its intricacies. The basic gist is that there is a salary cap for each team, but in order to attract top names from other leagues, a team can have a certain number of Designated Players whose salary can be much higher but counts for relatively little towards the cap. That is how David Villa, Lampard, and Pirlo were brought to the club, so it obviously works as intended, but it also makes vast intra-club salary disparities and relatedly tends to make for a stars and scrubs approach to squad building. As such, it's important that your DP players live up to expectation and also critical to find players outperforming their salary.

One of the great things about MLS is that salary information is public, so we can see how players performed relative to their cost:

Player Salary (xG+xA)/$100,000
McNamara 85000 9.54
Allen 63000 5.95
Shelton 94500 5.13
Harrison 160500 2.98
Lopez 51500 2.95
Mendoza 207276 2.64
Taylor 91875 2.42
Matarrita 175000 1.76
Bravo 110000 0.66
Diskerud 761250 0.46
Villa 5610000 0.45
Mena 231400 0.32
Brillant 299666.7 0.31
Iraola 200004 0.28
Lampard 6000000 0.14
Hernandez 210000 0.12
Pirlo 5915690 0.11
Saunders 150000 0.02
White 62500 0.00

As you can see, because the salaries for the Designated Players are an order of magnitude higher, they don't look so good once you take into account their cost (though they would look much better when it comes to their salary cap hold). It also highlights just how valuable Tommy McNamara is, since his salary is so much lower than most of the other players and he's contributing these goals from midfield. The salaries spent on defenders seem high given their abysmal defensive performance last season.

What to Expect

As mentioned, this will be a very different NYCFC team compared to last season. Over half of the players with MLS minutes last season will be gone this season (assuming Diskerud leaves), and 9 of the 19 with more than 500 minutes. In addition to Iraola and Lampard, starting keeper Josh Saunders and defender Jason Hernandez will be leaving, as well as forward Stiven Mendoza. Not all of these departures are necessarily a bad thing: Saunders was overrated as a keeper and disastrous with the ball at his feet (he makes Joe Hart seem a natural at playing from the back). Hernandez too, despite his New York roots, seemed a bit out of place at both center-back and right-back. Other than Saunders' replacement Sean Johnson, the incoming players don't really seem to fix either the preexisting or the looming holes in the squad. NYCFC's only signed two defenders despite their poor results, one of whom was a trialist in training camp. There are two possible replacements for Iraola, but Yangel Herrera is 19 and Alexander Ring was struggling for minutes in the German second division. Hopefully one of them will be able to step up and make the position his own.

The primary focus of the window has been attack once again. NYCFC's first signing was striker Sean Okoli from the USL, they drafted attacker Jonathan Lewis, and their international signing was Maxi Moralez from Leon, formerly of Atalanta in Serie A.With Villa entrenched at striker, Okoli is presumably relegated to backup duty, Lewis will have to contend with NYCFC's last two rookies Shelton and Harrison to get minutes. Moralez is primarily a Number 10, so it will be interesting to see if NYCFC shoehorn him into the 4-3-3 they had played with last season in Lampard's role or give him more freedom further forward. His Usage Rate is also higher than Lampard's, which make it tougher to integrate in a midfield with Pirlo. That said, he played with a Pirlo clone in Luca Cigarini at Atalanta, so he can probably make it work here.

NYCFC seem to be hoping that Moralez's creativity can create enough chances for Villa so that Lampard's magical finishing is not missed, that better passing from the keeper and backline reduce the number of easy opportunities for opponents, and that one or both of Herrera and Ring can solidify the midfield. The promising youth talent will likely improve once more, as Harrison and Shelton in particular stand to gain from Stiven Mendoza leaving, providing an extra boost. However, there are still clear issues across the defense and the age of key players Villa and Pirlo are also a concern. All in all, I'd expect something similar to last season: a dangerous attacking outfit with defensive issues that can make the playoffs, but not make too much noise once there. That's not really a bad thing, but whether that will meet the demands of the fans and ownership is another thing entirely.