The Morrison Report – Running at the Defense Edition

Earlier today, Kelly McLain, my esteemed counterpart at NASN Portland, wrote an articulate and impassioned defense of the Timbers defense.  Although I strongly encourage you to give Kelly’s column a full read, his working thesis is essentially that the Timbers’ concessions are not yet cause for great alarm due to their flukey, once-in-a-blue-moon-mistake, nature.

Kelly is certainly correct that it would be imprudent to write the book on a defense one match into a 34-game season.  Nonetheless, I respectfully dissent from his conclusion that it is too soon to harbor concerns about Portland’s defense.

In the three serious first team games in 2013 – San Jose, AIK, and New York – the Timbers have conceded seven goals on 31 shots.  Hidden within this statistic is both good news and bad news.

Before we get to the good and bad news, keep this one caveat in mind: we’re dealing with a very small sample size with the 2013 Timbers right now.  Thus, it is likely these numbers will change – and perhaps normalize – as the Timbers add to the sample.  The point of this column, however is not undercut by this, as the only point I make here is that the Timbers’ current performance must improve dramatically if they want to compete for a spot in the playoffs.  My thesis, then, is this: If the Timbers want to compete, their defense must improve substantially.  That, by itself, is cause for concern.

Back to the news, starting with the good news.  Through these first three games, the Timbers are conceding 10.33 shots per game.  This is remarkable defensive efficiency.[1] In 2012, the team with the fewest shots in the league – Vancouver – averaged 10.44 shots per game.[2]  The team with the most shots, Sporting K.C., averaged 15.94.  The median Seattle Sounders averaged 12.91.[3]

From this, we can infer that the 10.33 shots per game conceded by Portland would be at or near the top of MLS if extended over the course of a season.

The bad news, however, is very troubling.  The Timbers have conceded goals on 22.58% of the shots they have allowed.  That is astronomically high.  In 2012, the team with the best shooting percentage (the percentage of shots that resulted in goals – which I will refer to as “offensive efficiency”) was New York with 14.47%; substantially lower than the Timbers’ 22.58% defensive inefficiency rating right now.   The median offensive efficiency was Chicago at a round 10%, with the mean just slightly higher at 10.27%.[4]  The Timbers’ defensive inefficiency rating, if you will, is more than 8% higher than the highest offensive efficiency rate in 2012, and more than double the mean and median offensive efficiency rates.[5]

Here’s why this is really cause for concern.  While the statistical correlation between the number of shots taken and the number of goals scored in 2012 was somewhat weaker than I thought it would be, the relationship between offensive efficiency and goals scored was very strong.  The top seven teams in goals scored were also the top seven teams – albeit in a slightly different order – in offensive efficiency.  If the defensive corollary holds true, and there is a strong relationship between defensive inefficiency and goals conceded – something that is a statistical likelihood, though not a certainty – it becomes clear the Timbers defense must improve significantly or the season’s success is in serious peril.

This by itself, however, does not altogether disprove Kelly’s thesis – that the Timbers’ concessions may be more fluke than trend.  What’s troubling, though, is that the Timbers’ manner of concession is consistent with what would be expected given their tactics.

Portland’s 4-2-3-1 plays a very narrow, very high midfield.  All five players in the middle levels of the Timbers’ formation – Diego Chara, Will Johnson, Darlington Nagbe, Kalif Alhassan, and Diego Valeri – play very centrally.  The width in the Timbers’ offense comes from fullbacks Michael Harrington and Ryan Miller, who both are pulled very high.

The theory behind this is essentially to force the Timbers’ opponents to devote massive numbers to the middle of the field.  In doing so, the opponents won’t be able to exploit the space on the Timbers’ flanks because all of their wide attacking players will have been sucked into a losing battle in the middle.

We saw this for extended periods against the Red Bulls.  From about the 30-minute mark on, the Red Bulls “offense” was essentially reduced to thumping long balls over the top to a stranded Thierry Henry.  Even though there were acres in which to run on the outside of Portland’s defense, New York couldn’t get there because the Timbers had pulled all of the Red Bulls’ wings inside.

This largely explains why the Timbers have conceded so few shots.  Because Portland’s opponents get mired in a central midfield battle, and because the Timbers are good at holding the ball for extended periods, the opponents’ opportunities to build offense and put together an organized attack are limited.

The problem, however, is that this is bound to break down on the Timbers a handful of times per game.  A pass is given away in the midfield.  An opposing fullback sends forward a good long ball.  If an opposing attacking player can get into that space on the flank, the defense is unlocked, essentially leaving two center backs to cover the width of the field.

This, then, is consistent with the Timbers tremendous shots allowed numbers, but very, very poor defensive inefficiency rate.  Simply put, Portland’s tactics limit the number of opportunities opponents have, but when an opportunity inevitably comes along, they are significantly more lethal than the average chance.

It is too simple, then, to explain the Timbers’ defensive struggles as coincidence.  New York’s second goal is the perfect anecdote.  Yes, it was proximately caused by a mistake when Mikael Silvestre whiffed in his attempt to cut out Kosuke Kimura’s long ball forward.  But that sort of mistake is made more likely by Portland’s tactics, which force central defenders to defend wide spaces they aren’t necessarily accustomed or well suited to defending.  Add in a central defense that appears to be mistake-prone under the best of circumstances,[6] and you have a perfect storm of defensive exploitation.

In sum, it is no coincidence that the Timbers first team has played three home games, crushed all three teams in the eyeball test, and yet failed to come away with a single maximal result.  The defense just hasn’t been good enough at putting out the inevitable fires to let the offense win games for them.

There is still plenty that can change.  The return of David Horst or Hanyer Mosquera could alleviate some of the struggles.  The improving health of Jack Jewsbury could permit the Timbers to go to a narrow a diamond 4-4-2, which would give the flanks a little bit more cover because Chara and Johnson could play a little bit wider.

So, I agree with Kelly that it is too soon to say that the poor defense will end the Timbers’ season.  But serious concern about the defense is perfectly justified at this point.  If the Timbers’ defense can’t start putting out more of these inevitable fires, the current short trend of dominant play and disappointing results will likely continue.

That is why the state of the Timbers’ defense is the single most important issue facing the team right now.

Onward, Rose City!


[1] I use offensive stats as comparison because, well, they’re easier to find and compile than defensive ones.  The defensive statistics, however, are often corollary to the offensive stats cited here.  This is no more complex than saying if one team takes a shot, their opponent concedes one.  Thus, with an appropriate sample size, the defensive statistics will generally be close to a mirror image of the offensive statistics.

[2] Vancouver was last by quite a ways in 2012, as Philadelphia was second worst at 11.08.

[3] Interestingly, the correlation between shots taken and goals scored was weaker than I expected.  The third best goal-scoring team in 2012, New York, had the third fewest shots with 394.  The most prolific shooting team, SKC, had a relatively middling 42 goals.

[4] The mean was pulled up by New York and San Jose (13.92%), both of whom had absurd offensive efficiency statistics.  The Galaxy were third at a comparatively meager 12.29%.  Just for fun, the least efficient team was Chivas USA who scored on a disastrous 6.05% of their shots.

[5] Keep the small sample size in mind, however.

[6] See, e.g., Silvestre’s awkward back pass on the first concession or Andrew Jean-Baptiste’s lost mark on the third concession against New York – a set piece.

This entry was posted in Gossip and Rumors, TA Blogs. Bookmark the permalink.

3 Responses to The Morrison Report – Running at the Defense Edition

  1. Nick says:

    I’m really worried if our hopes for improvement on defense are the return of Horst, Mosquera, and Jack Jewsbury. I don’t think Andrew Jean-Baptiste’s judgement is more questionable than Horst’s and Mosquera didn’t seem to be picking up Porter’s approach very quickly. Zemanski may return to form and adjust to the system faster than Jack recovers. It seems like Porter would prefer the 4-4-2 diamond but is just working with what he’s got right now. Hopefully when we’ve got a more fully healthy roster we’ll see more tactical flexibility on top of improved performances as the team gels.

  2. Jon Green says:

    I think the second half of the New York game is much more indicative of the potential of the team than the goals surrendered so far. That is very easily attributed to having a brand new back five who are learning to communicate with each other than it is to the sum of their abilities. But they definitely must improve this area soon.

  3. cg campbell says:

    Great blog! Unfortunately, we had another bad game defensively last night. I really don’t understand the lack of urgency back there. We hold possession for a few minutes and then it seems as if Silvestre is using the “Toro” bull-fighting defense on every counterattack. He’s certainly lost a step and is not playing with heart. To me, this is another personnel mistake where we brought over an older high profile player from Europe who doesn’t have it anymore. Is it possible to ship him to wherever Kris Boyd went? Jean-Baptiste played admirably last night, but we’ve got to shore up that left side and give him some help.

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>