Match analysis: New York City FC 1 – 1 Philadelphia Union

Photo by Stephen Speer

The Union’s regular season concluded Sunday afternoon with draw against one of the hottest teams in MLS, New York City FC. As the playoffs loom, a few things stood out in the final match that could also be considered part of the team’s 2021 summary.

Lots of points, lots of points dropped

The Union had a man advantage and the lead on Sunday. It wasn’t enough to take the game’s full allotment of points. In fact, the full allotment is something the Union have struggled to gather in two important situations: in road matches and up a man.

However one slices it, winning points on the road is hard in MLS.

How hard?

Harder than any professional soccer league in the world, more than 60% harder than the Premier League or the Bundesliga (and no one in either of those leagues has to play away games on recently re-sodded baseball fields).

Irrespective of those difficult odds, fans will say that the Union should have closed the door on Sunday anyway. City made one key change that – even down a man – made that task more difficult.

They moved Maxi Morales farther up the pitch. He was already playing the 10 for the Pigeons, but was dropping deep to try and link the team’s trademark through balls and switches. Down a man and down a goal, moving him back toward the attack is counterintuitive because most teams would try and bunker, keep the score close, and see if they could nip one on the counter. This is in fact something City is really good at! However, their strategy was different on Sunday and for an obvious reason: there was no downside to losing by a lot and plenty of upside to throwing everything forward for the win.


In order to move up in playoff seeding, City needed those points. In order to fall out of the top four, City needed to lose by 12 goals. So, it’s no wonder they regrouped at halftime and pushed everything forward.

Strikers, or “The Tyranny of the Or,” or The Backup Quarterback Fallacy

Union fans want a new striker.

This has been true since Sebastien Le Toux was scoring goals (“He’s more effective on the wing!”), through when CJ Sapong was setting team records (“He’s too streaky for the team to rely on him!”), to today, when Kacper Pryzybylko is earning continental Golden Boot awards and consistently hitting double digits in goals per season (“His feet are bad, he’s slow, etc…”).

Jim Curtin said in his post game press conference that they gave City star Valentin Castellanos one chance all game and he scored. He meant that as a compliment to the guile of the opposing striker. Meanwhile, Union fans lamented that their striker only got one chance on the night and scored it, feeling that the goal itself wasn’t impressive and the Muffin simply wasn’t involved enough in the game’s outcome.

Beauty, as they say, is in the eye of the beholder.

The comparison is an apt one given the weekend’s matchup, but also because the league-wide narrative is that Castellanos is far and away a better attacking player than is Pryzbylko. The numbers suggest that’s not quite the case.

Each player has played nearly exactly the same number of minutes this season with the following scoring output.

  • Castellanos, his team playing a possession-heavy, short-pass-centric 4-2-3-1, took 132 shots this season, scoring 19 goals.
  • Pryzybylko, his team playing a chaotic, possession-lacking, cross-heavy 4-4-2/4-3-2-1, took 63 shots this season, scoring 12 goals.

Knowing how frustrated Union fans have been with their striker, few give him credit for having the unmeasurable “nose for goal” they desire in their ideal marksman. And yet, he measures quite nicely against his City counterpart in that statistic too.

  • Castellanos’s Expected Goals number for 2021 was 19.05, meaning he scored exactly as many goals as prediction models thought he might.
  • Pryzybylko’s Expected Goals number for 2021 was 11.16, meaning he scored nearly a full goal more than those same models predicted.

The Union’s striker scored a goal for every 5.25 shots he took, while City’s tallied once every 7 shots. There isn’t a debate here: Pryzbylko is more efficient . The debate is about his efficiency when given a chance, which is a crucial part of the conversation that can’t be left out.

City is built to get the striker the ball on the ground far more often than are the Union, evidenced in part by the fact that Castellanos took twice as many shots as his Philadelphian counterpart (how many times did Union fans scream as another aimless cross was sent in the direction of Pryzbylko, but not to him in any meaningful way?). The Hometown team’s formational transition (more on that below) should create a lot more quality chances for the striker than did their prior iteration.

Castellanos does come out ahead in total Goals Added to the team, beating Pryzbylko by roughly 3 over the course of the season. However, most of that comes down to the amount of shots he gets, while the specific qualities Union fans gnash their teeth about in the overall total aren’t all that dissimilar between the two.

  • Castellanos is a net negative to his team in both dribbling and passing
  • Pryzybylko is also a net negative in dribbling and passing (cue teeth gnashing), though less in aggregate than Castellanos in those categories

This isn’t the first time this opinion has graced the Philly Soccer Page pages. Given Union fan’s passion about this position in particular, it seems certain not to be the last either.

Jamiro or Paxten, Leon or Jack

Sometimes the numbers tell the most compelling story, other times the eye test is where truth can be found.

For the Union, the eye test suggests the team’s 4-3-2-1 is better at playing to the team’s (healthy) player’s strengths. Daniel Gazdag seems more comfortable sharing the spotlight at the top of the formation, and the Union’s spine seems significantly more solid with Alejandro Bedoya and Leon Flach protecting the team’s outside backs via their inside channel.

And yet, the eye test also suggests the team only hits its highest gear when their highest paid player and their best offseason acquisition are on the bench.

Jamiro Monteiro looks exquisitely talented on the ball and has regained his form as a defensive menace. Paxten Aaronson gets forward more quickly and gets dangerous shots off more effectively.

Leon Flach is a tireless defensive wall and is adding an offensive dimension to his tool set every match. Jack McGlynn is a sublime passer of the ball with a knack for finding space, either to ping into or run into himself – oh, and he’s been compared to Mesut Ozil.

Maybe it’s that the homegrowns look more energetic precisely because they’re entering matches after most of the players around them have sprinted for an hour. Maybe it’s because they’re just really good and should see the field more often. The eye test suggests it’s a conversation worth having, and one that is certainly being had down at The Wharf.

Regardless, don’t hold a breath for either of the youngsters to start this postseason, barring an injury. For as much hand-wringing as Union fans have done about playing them more often, it’s impossible to argue with the team’s results using them exactly as they have in 2021. When they’re ready for the lead roles they’re destined to play however, look out.


  1. Jeremy Lane says:

    I’m very interested in what everybody has to say about this. For my part, I both totally agree with what Chris just said vis a vis the strikers, and yet also believe there are gains to be made (marginal or more) by improving the striking situation.

  2. Unfortunately for Kacper the stats provided don’t include pressures.

  3. Enjoyed the whole article but respectfully disagree with this conclusion – it’s impossible to argue with the team’s results using them exactly as they have in 2021.
    I think that you could argue that a major reason Pryzybylko has had half as many chance as Castellanos is that for most of the season the Union have been dismal at moving the ball on the ground in the final 3rd – something BA was very good at last year. Instead its all wing play and nonthreatening crosses ad nauseam.
    You could also argue that Flach hampered the offense as much as he helped the D.
    Paxten might not have been ready to start at the beginning of the season, but I think he is now, and at dual #10 he gives this team something the other players mostly don’t – the ability to quickly and skillfully move the ball up the middle. To be fair, Gazdag has also gotten better at this recently. Montiero though continues to dribble & pass much more horizontally than vertically.
    Curtin will start who he starts, but for my money in the current Union best 11, Paxten pushes Montiero back into Flach’s spot at left wing and Flach/Martinez is a toss up at D Mid.

  4. I’ve been leaning most of the season toward believing the system just doesn’t suit Przybylko — or scoring in general. Przybylko certainly has his faults, but he’s the best finisher on this team by quite a distance. Often his runs aren’t picked up by teammates who will loft in hapless crosses to no one or simply take a poor shot on their own. Someone on here in the last week or two (sorry I can’t recall!) made a good point that Przybylko is hurt by not having the Santos to play off of. If Santos is anything, he’s a nightmare for defenders to handle. So I’m on team Przybylko.

    I hate to sound sour, but it’s mostly because I think this team is underperforming the talent it has. I’ll stick by that. Yeah, it came in second, but were 19 points off the pace of first, only 7 points away from falling below MLS’s very generous playoff line.

    That said, the glass-half-full perspective is that the team battled well given that two key players were sold and there was a short mutiny from Montiero at a crucial point in the season. They remained resilient and pushed on. There’s no reason they can’t beat a couple of these other equally flawed teams in the playoffs. Rooting for them to do well.

  5. I mean, I totally expected them to lose

  6. I think there are reasons that MLS has the biggest home field advantage, some of which apply as normal while others are either essentially not applicable since COVID or at least weren’t applicable to the Union in their run over the last 2 months.
    First, the difference between the top and bottom of the league is much less in MLS than it is in other leagues. Teams like the top teams in the Premier League are so much better than the bottom teams that we are not surprised if one or more of the top teams were to pick up 24 points on the road against the bottom 8 teams in the league and it would be a major shock if they picked up less than 16. There is no team like that in MLS. New England “only” picked up 20 points in 9 games on the road against non playoff teams which is toward the bottom of what we regularly see for the top of the Premier League. So that still applies this year.
    The second issue is travel. The longest road trip in the Premier League is approximately equal to the distance between Boston and Washington, DC. There are no time changes in most leagues. In MLS, especially in the eastern conference, travel was significantly reduced this season. The Union only played 4 games outside the Eastern time zone. No team had more than one cross country trip (excluding CCL). In a normal season there are multiple games where the road team is at a significant disadvantage due to travel that doesn’t happen in other leagues.
    The other big one in some seasons (like this one) is the inequality in rest due to having an odd number of teams. When a team is on the road (especially for a longer trip) on two days rest playing against a team at home who hasn’t played in a week, there is a significant advantage for the home team. However, for the Union’s closing run (since they were eliminated from CCL) they were on essentially the same rest as their opponent every game, home and away, except the one after the international break where they had played 1 week earlier and Montreal had played 2.
    All that being said, I don’t think the frustration is from the lack of wins so much as how they happened. It would have been one thing if it were a bunch of games like the one in Nashville where they gave up an early goal and just couldn’t get it back. But that wasn’t the case. They had the lead in 7 straight road games and only beat Cincinnati (which is the only one they had a two goal lead).
    Against DC United, they led on a fluke own goal and were completely outplayed. Fine, it happens. Against NYRB, they got a somewhat lucky goal by Sergio, gave up a well taken tying goal later in the first half, and then never really threatened. A result against a team that made the playoffs. Cincinnati was the win. Then came Montreal. Although they didn’t play great, they were able to get a 2-1 lead and keep it well into stoppage time. Then they gave up the tying goal on a late throw in and cross. This was probably the most frustrating non win. Minnesota saw them take a 2-1 lead early in the second half only to give up 2 quick ones before going up a man for the last 15 minutes to no avail. By itself, not so bad, but we are seeing a pattern here. Against Toronto they scored less than a minute in against a team who had nothing to play for but they couldn’t score a second until after Toronto had taken the lead. (They probably should have had a PK late here for a chance to win but MLS officiating is a different conversation.) Finally we have the NYCFC game which we know about.
    Individually, none of these games is awful, but collectively they point to something that needs addressing in the offseason. Winning 6 of these games (still losing in DC) still would have left them 8 points short of New England in the conference, but a month from now, if the Union are still playing these dropped points from winning positions (along with the blown lead in the home opener against Miami) will likely be the difference between a game a Subaru Park on December 11 and one somewhere on the other side of the Missouri River.

    • The Union Jack says:

      A boss from decades ago use to say “take care of the minutes and hours and the days and weeks will take care of themselves”. This year Bruce Arena had New England focused on the minutes and hours.

    • Chris Gibbons says:

      Would your perception be different if they had eked out results in each one instead of let them slip? Wouldn’t the narrative then be that they couldn’t compete for most of the match and we’re lucky to steal a point? I’m using idiomatic phrases here on purpose because they serve both of our means. The game goes 90 minutes, and goals count the same in every one. It’s harder to defend with the lead because teams make adjustments to address those deficits. I’m not pleased any points were dropped, but the game state changes after goals and as time goes on. Holding the lead isn’t linear, it becomes more difficult every second closer to the final one. Also, whether you believe that each game is its own data set or that momentum is real, the Union were really, really good down the stretch and made up a ton of playoff ground in a dual competition season that almost always sees MLS teams fall by the wayside. That’s the important narrative.

      • First, I definitely agree with you that the Union have been really, really good over the last 2 months and it’s the first time they’ve gone into the playoffs playing the best they’ve done all season and I’m hoping that with some help from NYCFC or Atlanta, I’m spending money on games at Subaru Park into next month (although I don’t think it would be a stretch for them to get a result in New England considering they haven’t put their first team out against the Revs since mid May).
        As far as the difference between getting the tying goal and giving up the tying goal, the bad in the latter outweighs the good in the former by a considerable margin for one simple reason. In one case you are gaining 1 point. In the other you are giving up 2. So the effect on the standings is twice as big. A win and a loss is 50% better than two draws. Heck, a win and two losses are slightly better than three draws due to the tiebreakers.
        One of the things I least like hearing late in a tied game is when JP says something along the lines of the Union have to be careful because they don’t want to give one up now and come out of here with nothing. To me, that’s the wrong attitude. Just to exaggerate a little because nothing in soccer has these kind of swings, but if your play is going to win the game 30% of the time, lose 50% of time, and draw 20% of the time, it’s a better strategy than one that will get a draw 100% of the time since the former has an expected result of 1.1 points per game rather than 1 point per game for the latter.

  7. I don’t want Castellanos – who needs another diver? I think it would be interesting to look at the same stats for, say Ruidaz or Arango.

    To me, it’s not that Kacper is bad. It’s that there are better players who can be acquired without breaking Sugarman’s bank.

    • Chris Gibbons says:

      I think Tanner’s counter to that would be that have a guy that scored 17 goals this season, was acquired for free (I think), and costs less than $1 million a year. Guys like that don’t grow on trees.

      • And there’s also the possibility that any replacement doesn’t fit in with MLS physicality or spends a lot of time on IR like Santos and Burke.

  8. Union all the way

Leave a Reply

Your email address will not be published. Required fields are marked *


%d bloggers like this: