clock menu more-arrow no yes

Filed under:

Backpass 4-24-15: Why aren't we scoring more goals?

The collective dread among fans about the Rapids punchless offense is palpable. Going up a man with 31 minutes to go helped, but not enough. What the what is going on?

Isaiah J. Downing-USA TODAY Sports

Man, the offense. The team gets off the schneid in a big way against FCD on national TV, then gets shut down at home by Seattle. Against FC Dallas, this time at home, the team looks tentative in the first half and disorganized to start the second. After Blas Perez pushes Drew Moor and is sent off, Colorado has 31+ minutes of 11v10 soccer, and still looked ineffective as hell.

Here's a stats-heavy look at what might be happening. One important caveat to remember: the season is still young. Statistics generally don't perform as well with small sample sizes.

1) Science (Science!) says we should be scoring more.

Either we're unlucky, or terrible in front of goal.

One stat that can inform us as to whether this team is lousy at finishing is possibly my favorite stat, Expected Goals (xG). xG takes the position on the field at which the shot was taken and applies a metric that tells you what chance that shot should have of going in. It does this through the gathering of masses of data that supports that number. It is adjusted to a value of 1.0: anything higher than 1.0 means you would expect more goals based on where the shots were taken, anything less means you would expect fewer goals. It isn't a perfect number: it does not take into account the position of the defenders: so a wide open shot on a break and shot with 8 defenders in the box are rated the same. However, with enough data, those differences tend to even out.

Look below at the Rapids (in red). The Rapids have 0.83 Goals For (GF; divided for games played; we had 5 goals in 6 games before Friday). Based on xG, we'd expect to have 1.10 goals; a difference of -0.27.  That could be poor finishing: they punt the ball right into the GK or wide, when science (science!) says they should score. Or, it could be that they have been verifiably unlucky. As the worst team in MLS before Friday, if our finishing was really that bad, we'd expect to have the worst xG to GF difference in MLS.

But we don't. Look at the blue numbers. Every team in blue has a worse xG to GF number: SKC is -0.33, NE is -0.49, ORL is -0.30, NYC is -0.62, etc. Most of those teams currently sit higher in the table than the Rapids. So maybe we've been a little unlucky.

Before I was a Rapids fan, I was a big baseball fan that loved the numbers, and especially the new numbers revolution that Bill James and Billy Beane brought to the game. One stat that really informs the game is 'BABIP'- Batting Average on Balls in Play. This number can tell you if a guy has just been incredibly lucky. For instance, if Nolan Arrenado is batting .400, and .600 on BABIP, it means that no matter where he hits the ball, it falls for a hit. Lucky bloop singles land in juuuust the right place, line drives never go right to the shortstop, etc. Well, that's unsustainable. He's eventually going to come back to earth: 'regress to the mean', in nerd-talk.

In short, what I mean is, the Rapids aren't putting the ball in the net as much as you'd expect. On top of that, their Expected Goals Against (xGA) is 0.97. They have a xGD (xGF minus xGA) of +0.12. It means, statistically, the Rapids should overall have more goals than they've given up, and since scoring more than your opponent is pretty much the point in soccer, the Rapids should have more wins than they do. The difference of 0.12 ranks 5th in MLS.  The other teams above them, like Seattle and Columbus, have ridden that differential to the top of the table. The Rapids… not so much. Perhaps, then, a correction is likely.  We will likely score more goals, and if our defense stays steady, our scoring will likely result in more wins.

(stats care of www.americansocceranalysis.com)

Those teams with yellow boxes are scoring *more* goals than expected. Vancouver and NYRB are two of the best in the league, so their finishing has either been remarkably good, or they've been lucky, or both. Toronto and FCD have both underperformed. For TFC, there's a simple answer, and that is the purple box. Their defense has been awful. Imagine if the Rapids D was that bad. We'd be having a fit. So, it could always be worse.

If we want a nice instant case study, look at RSL: their GA of 0.66 and xGA of 1.26 is pretty far out of whack. Last night, they got shelled for four goals. That might be a good example of 'regression to the mean'- when your GA and xGA are way different, either you have some magical force field around the goal (or Manuel Neuer in front), or you have luck creeping in. RSL proved last night that their defense was a bit smoke and mirrors. So the Rapids offense might be better than it appears.

2) The offense isn't getting in front of goal, is heavy on the left side, and takes too many low percentage shots.

So bad luck explains a little. But still: the Rapids xGF is only 1.10; good enough for 15th in MLS. That's not gonna get us to the playoffs unless the defense is unbelievable or Don Garber decides to add three more spots to the playoffs. Could happen. Probably not.

So what's wrong with the attack? One thing especially noted by @losthooligan and @sec106COrapids: the team attacks from wide, crosses too much, and doesn't challenge enough in zone 14: the spot in front of goal which generates some of your best chances. So, lets take a look.

Wow. The Rapids are dead last in shots from the middle of the field, and lead MLS in shots from the left side. This makes a lot of sense: Juan Ramirez likes to work from wide, and has been taking more shots than whoever's been on the right. Dom Badji overwhelming works on the left. Dillon Powers favors the leftish side too. But put more simply, when your team is clumped to one side, you aren't taking advantage of the whole field: you are denying yourself space. Part of this is players moving together and getting close for short passes, predominantly on the left. Great. But they've got to get from left to central for shots. Shots from left and right, very simply, are lower percentage, because the shooter is looking at a much smaller space and the GK can cut down on a very small angle.

More importantly, again, look at the top of this list. NY Red Bulls: best in MLS. Orlando City : Kaka understands where to generate goals on the pitch, but also has the skills to weave into those spots. Sounders FC: best offense in MLS. And the Philadelphia Union: ok, numbers can't explain everything.

Still, the Rapids offense needs to work on taking shots in better places. In the second half against FCD, we saw some of that. Here are shots and Key Chances from the last 45'.

Good! More central play! Shots from Zone 14! Of course, most of that time (31 minutes of 45) was spent 11v10. Call it training wheels. Now lets hope the boys can do it versus a full strength side.

That all said, I'm not opposed to an offense that tries often to get wide and fire crosses into the box. English soccer teams have used that tactic to devastating effect for years. Alan Gordon and Conor Conner Casey and Peter Crouch and Olivier Giroud can do that all day. Marcelo Sarvas can deliver that kind of service. Only, Gaby Torres is not a tall, strong, headed goal guy. I'm really hoping Kevin Doyle's addition in July can give this team an added dimension that makes a strategy like this effective. Right now, it's not such a good idea though.

3) We waste perfectly good possession without getting forward and look totally toothless at times.

Sometimes numbers tell you something. Sometimes, you watch a game and your eyes tell you whats wrong: you don't need any numbers. Look at these tweets:

There was just a lot of pointless possession in the second half. Wasted time, lateral passing, getting wide and crossing to nobody in the box, etc. Before the red card, the offense looked disjointed. After, they took a while to get going. THe team starting to look a bit more threatening after the 70th minute, culminating in LaBrocca's 79th minute equalizer off a gorgeous little header from Dillon Powers. Vincente Sanchez subbed on in the 62nd minute, but didn't deliver much impact either.

This is a game with a clock, so futzing around with the ball for 25 minutes when behind isn't ok. I think Marcelo Sarvas' absence was a big factor; players operate off him well because he gets into spaces where others can operate off him. LaBrocca and Pittinari don't do that as much.

Quick Takes

Last week I was tough on Juan Ramirez, and said he wasn't being as aggressive on the dribble. This week he had SIX take-ons,and three were successful. Great step forward...

I put my faith in Dom Badji early. He had a lousy game: 50% passing, 38% on Aeriel Duels, zero shots, zero take-ons. His hold-up play back-to-goal resulted in lots of final-third turnovers. In case you were wondering how good a defender Matt Hedges is he can basically disappear a player like he was a character on 'Sopranos'. Is that too dated a reference? I have little kids, I don't really watch tv anymore...

Axel Sjoberg and Michael Harrington totally got clowned by Fabian Castillo...

It happens. Don't panic. Go watch RSL or Philadelphia's defense get destroyed from this past weekend, and you'll see that Axel and Mikey giving up one goal in 90 ain't no thang. But if it repeats, then we can start panicking. However, based on the things I've seen on twitter, maybe most of us already are.