What drove the Red Bulls late play-off push? / by Harrison Crow

By Harrison Crow (@harrison_crow)

I'm going to throw a quote at you from an American Soccer Now article. Read it below then we'll talk about it and break stuff down from a data perspective. Okay,  BRB.

I look at the tactical shift to a two defensive midfielder formation, putting Eric [Alexander] and Dax [McCarty] together, as the key to turning around our season,” Robles said. “Coming off the Supporters Shield we got off to a slow start and were trying a lot of things and there was a lot of trial and error. But by moving Eric next to Dax it resulted in fewer chances conceded while allowing our offensive juggernauts to do their thing.
— Red Bulls Goalkeeper, Luis Robles

Luis Robles doesn't quote any stats or data here. It's more of an intellectual response in attempting to encapsulate two different points in the season. Obviously, Robles has a good frame of reference from which to put together this thought. Unfortunately it comes across more as a feeling that manifested as result of the situational outcomes.

The last 9 games of the season the Red Bulls average 2.1 points per game and were very good as a whole. This was also a much needed improvement upon heir 1.24 points per game they average the first 25 matches this season. It begs the question was this accurate line of thinking by Robles and did the Red Bulls really get that much better from a shot perspective or was it a swing in luck?

I'm not trying to specifically get into the nitty-gritty of the tactics to determine whether McCarty or Alexander directly influenced shot creation or allowance. The point of this exercise isn't to get that granular, but to look at the shots from a general perspective. Looking for indications of how the Red Bulls turn around came about.

Robles mentions that he felt they conceded fewer shots a game and author Brooke Tunstall back that point referencing Robles made less saves on average during those 9 matches. That doesn't really mean that they took less shots. What's the difference? Well, luck is the difference. Shots on target are a subset of the greater whole in the sense that it only shows you a small picture of the attempts to score.

Logic would tell you that just because the opposing team didn't put a shot on target doesn't mean they shouldn't have scored a goal or that they didn't create a dangerous opportunity. It just means they lacked in execution of the final component of scoring a goal, getting it on frame. Albeit a very important detail, just not the only detail to note.

Below is a quick synopsis of shots before September 6th and then the few following along with some aggregate data.

You can see that there are a few differences of note that occur before September 6th and then again after. Let's take a look at this snap shot and discuss three specific things from it.


Slow start: Red Bulls attack was good, defense not so much

There are a lot of things that you can say about the Red Bulls start at the beginning of the season. Chalk it up to personnel or to bad luck or even just bad coaching decisions. Whatever, dealers choice. You just might as well sum it up as not being good. That is magnified all the more by the fact they over performed in the expected goals metric. Unfortunately for them, they shot themselves in the foot with how many opportunities they in turn gave their opponent.

You can't expect good results when you give your opponent that type of advantage. It doesn't work in your favor and this is exactly how we would have expected it playing out.

It's easy to pander to the narrative that New York was simply unlucky in the first half. It's actually even a bit convenient in this situation with trying to explain the last minute "jolt" they received an effort to make the play-offs. The thing about that is that luck wasn't really involved, it was simply about the amount of volume they were allowing to be shot at their own goal. That's just not playing good.


Red Bulls got a bit crazy in their final nine games.

They scored 16 goals in their final 9 games and out produced their expected outcome (13.97) by two whole goals. They did this upon an average of 1.552 expected goals per game a slight increase upon their first 25 game total (1.521).

Looking at the expected goals with shot totals can give us a rough idea of the average shot placement on the pitch. Which indicates they were still firing from pretty much the same places as they were earlier in the season just scoring more frequently than before.

The result of the shots being converted so frequently could mean either one of two things in my mind. It could be positive regression to the mean. This would be assuming that the players who were taking the largest portion of the shots the first 25 matches had finishing that was performing below their skill level. Or, and I feel this is more likely the case, the Red Bulls just got lucky in a situation where they needed it.


Luck isn't a bad word

Yes, so there was likely some luck involved in how they found their goals. That isn't suppose to undercut the performance of either Bradley Wright-Phillips or Thierry Henry. Both had great seasons and it's arguable that if it wasn't for one or even both that the season could have been much, much worse. Great strikers have a habit of out producing their expected goal. Now, that's not always the case but in this situation it was. Wright-Phillips scored 4 more goals than what we projected for him and Henry was good for an extra two goals himself. That's an additional six goals and some of that is luck but certainly there is another portion that comes from their skill.


Red Bulls reduced the attacks at their goal. 

It wasn't just that Luis Robles saved less shots on average in those nine games. Teams were giving him less chances to make those saves. The defense stepped up in a big way by reducing the volume of shots being sprayed at Robles' net. By proxy this lowered the average expected goals by their opponents. Is this the foot print of Alexander and McCarty? It's very possible, and taking it a step further I think that Robles portrayed the Red Bulls turn around very accurately.


The luck in attack and the corralling of opponents shots is exactly what helped propel the Red Bulls to big wins. The question going forward is whether or not they can continue to do it against New England. With Jermaine Jones roving both ways, the creativity of Lee Nguyen and the playoff resurgence of Charlie Davies.

The Red Bulls still have a real challenge ahead of them and while it's clear that they have the pieces that can slow down defenses and throw up goals on the score board. The transformation is complete, the question is can it get the Red Bulls to an MLS Cup? And if so, could they win it? Many questions and only two matches for answers.