Backpass: Mommy, where do goals come from?
Our friends the Rapids have eight goals on the year. We take a look at how they came about in order to discern some kind of pattern. How do they get goals - and how might they get more?
The other night on the podcast, my guest host Brendan Ploen and I were breaking down the RSL game (thanks Brendan!) and I reiterated a thing I’d been saying since pre-season - that the Colorado Rapids empty-bucket 3-4-2-1 formation, with Jack Price and Mark-Anthony Kaye in midfield, lends itself to spraying the ball wide and relying on creating goals in open play from the wide area. In our last match against RSL, the Rapids produced 16 shots - eight inside the box, eight outside the box; eight from set pieces, eight from open play. Six on target, seven off target, three blocked. All in all, a very even distribution of chances representing a good team with loads of possession, at home, against a struggling, bunkered opponent. And again - a lot of chances came from getting the ball wide to Jonathan Lewis or Michael Barrios or Lucas Esteves or Keegan Rosenberry in order to whip in a low pass or a cross to the middle. Colorado was a disappointing 2 for 18 on crosses for the night. So it felt like what I said was right.
But a thing I do when I watch a Rapids game is to then harden that thought into an opinion, which I subsequently, in writing or audio form, spew out to the masses. Because I am a responsible pundit, I sometimes later spend additional time wondering: “Is that thing I said/wrote really true?”
In this case, the question is: do the Rapids really prefer to go outside-to-inside in order to produce goals? Are they more apt to crosses and wide-zone dribbles than they are to attacks through the middle and into that top-of-the-box area that is sometimes called Zone 14?
A team gets roughly 30 to 60 offensive possessions per game, and the Rapids have played seven games. So I went back and watched all 329 possessions, and this is what I learned…
No, no, no. That’s nuts. I have kids and a job and cataloging every single Rapids play is maniacally insane. Instead, I just took a look at our eight goals on the year to break down if those chances can indicate, on a small scale, some tendencies we can notice in how the Rapids score. Maybe the sample size is still a little small, and maybe there will be some innate bias in just looking at the success rather than all of the near-misses or great opportunities that came up short. Nonetheless, I think there’s something, or a few somethings, to be learned here.
We start with a play that really fits the thesis I laid out: Barrios and Rosenberry are overlapping on the right wing; Rosenberry sends in a low cross to Max, who leg-chops it down and hammers it home. A fantastic goal that saved the ‘Pids, at least for a few additional minutes, from getting bounced from Concacaf Champions League. Of course, then they biffed it on Penalty Kicks.
One added side note: I believe due to the Comuniciones player’s attempt to play the ball, this one is not credited to Rosenberry for an assist.
I dunno what was up with Atlanta on this day - they got hammered 3-0 - but this one, the first one, is the worst of the three goals they conceded. Esteves beats his man so that he can unload a cross. He overcooks it completely, but there’s Barrios on the other side to scoop it up. HE gets an open cross into Rubio, who gets an uncontested header for a goal. Rubio is generously listed at 5 foot 10 on the roster - there’s no way he should have this little pressure on him for a header. Three poorly marked guys means: easy goal.
This play also fits the thesis - it is a Rapids open-play goal generated from a wide space.
If I were the head coach of the Rapids, this would be the paradigmatic way to score goals: forced turnovers from the middle third of the pitch, quickly rammed down the opponents throat for a goal. This is the way Manchester United in 2015 and 2016 might like to score a goal - get Marouane Fellani to knock down a big aerial, squirt it over to Juan Mata or Wayne Rooney, hold your 1-0 or 2-1 lead, and smother everyone in midfield. The Rapids did this too that year, with Sam Cronin and Micheal Azira (the patron saint be blessed) snuffing out all attacks right well before they even could approach the 18 yard box.
Rubio is not well known for timely precision passing - he’s never had more than 4 assists in a season - but perhaps if he gets enough chances, he will be known for that. He has produced 15 assists over the past four seasons, with just 5500 minutes. Maybe with a full 2600-2800 minutes, he turns in a proper 10 and 10 season? No, no, let’s not reopen the Diego Rubio thing for the six-millionth time.
(My bad. That's J. Lewis on the wing, not Esteves.)
This is, so far, the most gorgeously-worked goal of the year. It’s really a classic one-two, but also with the element of Rubio pulling his man out of position, then eliminating him completely with that next pass, and Andre Shinyashiki tucking into the exact right spot to capitalize. To be fair, this is also late in garbage time, against 10 men, with a 2-0 lead, so you can’t really derive much from this. But it sure is purdy.
This goal is Esteves slaying all day, mixed with Rubio breaking through the line at the precise right moment. The entire back line is sure he’s offside, but if we saw this from the sideline angle, we could see that Graham Zusi on the far right there has dropped a tad deeper than his defensive mates, keeping Rubio onside. Oops.
Again - open play, from the wing, working outside-to-in.
A sneaky variant of Loki’s Toboggan ; the Rapids run a stack, but Captain Jack Price bombs it deep to a free swinging, wide open Steven Beitashour. First, check out the work Lalas Abubakar does in posting up Graham Zusi and eliminating him, then screening out #3 , Andreu Fontás, who is absolutely nowhere on this play even though he clearly should be marking Mark-Anthony Kaye in the box. Colorado works a good set piece here, but this is also terrible space-cadet ball-watching zonal marking from SKC. Would absolutely not want to have been anywhere near head coach Peter Vermes after this.
“Let’s leave Colorado’s Canadian international completely alone at the back post for an absolute tap-in from a laser-guided precision strike from the league’s best set piece specialist.”
This one is interesting, in that RSL’s defense is set and in defending mode against the Rapids, and yet they still find a way to manipulate the defense into opening a window for them to exploit. That’s different than all of the previous open play goals, in which Colorado took advantage of a transition rush or the move from the middle of the pitch to the final third in order to strike. These goals - the bunkered, get behind 10 guys in the final third goals - are really hard to come by. It does help, however, that Colorado dominated possession and had most of the shots. RSL made it hard to find the net, but they didn’t make it hard for the Rapids to waltz into their end.
…
OK, so all in all, eight goals;
6 from open play; 2 from set pieces. (Fotmob says there are 3 set piece goals. I look it over, and I have no idea what they’re talking about).
3 open play goals that develop from the right side; 3 that develop from the left
3 goals via crosses (#1 is a low cross, #2 a high cross, and #8 a ground cross)
2 goals via over-the-top throughballs (#3 and #5) [one might arguably call this an ‘early cross’]
1 headed goal (off a cross)
1 goal via a dribble-drive
…
Rubio is involved1 in 4 goals
Barrios is involved in 3 goals
Lucas Esteves is involved in 3 goals
Jonathan Lewis is involved in 2 goals
Mark Anthony Kaye involved in 2 goals
Jack Price involved in 2 goals
Rosenberry involved in 2 goals
Abubakar, Shinyashiki, Trusty, and Beitashour all involved in 1 goal.
…
All 8 goals were scored from shots inside the box; zero goals were produced from shots outside the box
Zero goals were via a true throughball (AKA on the ground)
Zero goals were from shots outside the box … on 32 attempts (13 alone in the CCL Leg 2)
Zero goals were from Zone 14
Zero goals were from cutback passes in the box to the front of goal
Similarly, zero goals from late-arriving midfielders in the box
What I learn from all this is that my hunch was right - Colorado has scored by getting wide, or by getting set pieces. But wait! There’s more! Colorado is not scoring a lot of headed goals - which, when Diego Rubio is your center striker, is to be expected. Colorado is smartly scoring by getting close to goal, as 50 of their 82 shots this year have occurred inside the 18 yard box, but they still take the long shots, and they haven’t succeeded with any of them. In MLS matches, the Rapids have 7 goals on 6.59 Expected Goals, slightly exceeding expectation2.
Maybe, all in all, I should be satisfied. The team scores goals - well worked goals, clever goals, goals from active states, team goals involving a lot of guys. However, I can’t help but worry that it’s still all originating in similar places. I wonder if a clever or effective team couldn’t totally shut down the Rapids by playing a little bit wider - filling the space line to line and trying to force Colorado to go up the gut. Maybe a bunkered 5-3-2 where the full backs don’t sally forth; or an old school, extra wide 4-4-2. We’re also clearly missing that Cole Bassett-type goal scorer - the late arrival in the box, or the guy with the tricky slashing-dribble that gets into the box; or any other kind of dribble-drive central attacker. We knew that. But still, we really need another dimension.
It’s good, don’t get me wrong. And of course, I feel good about this team right now because, of course, I only looked at Goals Scored this week - we look great! That tends to happen when you conveniently ignore all goals conceded, or yellow and read cards, PKs, and the overall outcome of a game. Just looking at golazos on any highlight film will skew your perception of a team to feeling good and invincible about your club - ‘gee that Max goal sure was great. I assume the Rapids musta beat those kids from Guatemala easy, huh?’
But if we want to repeat our success of last year, the manner in which we create goals might need to become a bit more varied. Maybe I shouldn’t worry. But I worry.
Involved isn’t a technical term, just an acknowledgement of an immediate impact by a player in the attacking phase. For most cases thats a goal, an assist, or an ‘MLS assist’; in one case Trusty gets credit for starting the attack with a really critical pass.
I could have also manually added in the xG from the 2 Concacaf matches, except that xG number is produced by Opta while I’m using American Soccer Analysis’ number, and the two systems actually count xG differently. Also, all xG data is really weak until we’re at least 10 to 15 games in.