Diving Deep: Hartford Athletic 0-1 Birmingham Legion

You have to give the Legion credit: they are keeping those of us in the analysis business busy.

Three games in and we have seen three entirely different game plans. Well, not entirely, but certainly different enough that we are kept guessing as to what the team will do in any given game. Against Hartford we saw another counter-attacking strategy. In general, that has to be coupled with a very disciplined defensive setup. There is an obvious need to have the ability to absorb pressure and not allow the opponent too many scoring chances. That was certainly evident in Connecticut, with the Athletic having just 4 shots on goal and only 2 big scoring chances. Their xG was 2.10, which was the highest of any team in the USL Championship last week. They came up against a brick wall. The Legion’s xG was 1.61, by the way.

But that wasn’t the only secret ingredient here. There was a second, and far less expected.

Possession.

It is almost axiomatic that a counter-attacking team will not dominate possession. That makes complete sense. After all, if you are willing to sit back, drag the opponent’s defense out of position and look for a big opening, then that opponent is almost certain to have more of the ball. The tactic is also referred to as the low press and is essentially the soccer equivalent of the rope-a-dope. Let the opponents tire themselves out and then hit back hard.

Fair enough. But in this case the Three Sparks, who were on the short end of the possession game in both previous outings this season, had a 9-point advantage in ownership of the ball. That’s rather a double whammy. If you drag the opponent into your own half, but don’t let them have the ball, then the tiredness (and frustration as well, come to that) will come from chasing the ball for the better part of 90 minutes. Let’s examine a couple of graphics to see just exactly what the Legion did to achieve this. First, we’ll check the heatmap:

This shows the Legion only; Hartford have been taken out. On the left is the first half; the second half is on the right. In both cases the Legion is playing right to left. Now we will add a graphic we’ve not used here before. This one is called the gameflow:

This comes to us from the Twitter account @GameFlowxPG, which is operated by American Soccer Analysis. What it shows is the relative usefulness of possession in a game on a minute-by-minute basis. The longer the bar, the more likely a goal is to result. A really good scoring chance doesn’t materialize until the bar gets to about 0.5. The chart also very helpfully shows who dominated possession each minute.

So, from the heatmap we can see that the Legion was very much on the back foot in the first half, stuck pretty deep in its own end, and with almost no penetration at all into the final third. That correlates with the fact that they had the better of the possession in just 16 minutes in the half. But they kept the Athletic to just one decent scoring chance in the entire half, which was a long-range attempt by Antoine Hoppenot. For their own part, they had two good chances, one being Gabriel Alves’ goal, the other being a tight-angled shot by Tyler Pasher.

After half time though, they changed it up. No parking the bus here. Instead, the Legion pressed the attack, pushing upfield and holding the possession advantage in 27 minutes, including 4 in stoppage time (when they buried the ball in the corner). This did lead to some chances for Hartford, but they were all snuffed out. It wasn’t until after the 80th minute that Trevor Spangenberg even had to make any saves (he made a double and a single after that point). In fact, over the course of the game Hartford had considerably less penetration than the Legion. Of Hartford’s 14 shots, 8 were in the second half, but most of those were ineffective.

The average positions in the game are also telling:

As you can see, this is seriously clogged up in midfield, with almost every field player in the middle third. Moreover, this is kinda messy, formationally speaking. But, if anything, the messiness is on the Athletic side. Both teams came out in a 4-2-3-1. It’s nearly impossible to discern that from the graphic. Hartford looks more like a 3-3-4, which obviously is a majorly aggressive setup that pretty much no one uses. And also didn’t work.

The Legion were a different matter. If we remove Hartford from the picture it gets a bit easier to see:

I’ve highlighted the back four. Note that the two fullbacks, Gabriel (#16) and Collin Smith (#4) are both equally high. We haven’t seen that before this season, with the attacking fullback being mostly on the left. With both flanks offering potential threats, that’s harder for a defense to handle. I’ve also highlighted the two defensive mids, the double pivot, Anderson Asiedu (#6) and Matthew Corcoran (#17). More on them in a moment.

The front four, in contrast, look pretty flat. And that’s where the genius of formational flexibility comes in. All told I think you can extract as many as four different setups here. The initial 4-2-3-1 is there, although Tyler Pasher (#15) is rather higher than the striker, Neco Brett (#11). And because of that, you can also see a 4-4-2 setup, with Tyler being the second striker. In that alignment, Juan Agudelo (#9) and Enzo Martinez (#19) are the attacking mids (the 4-4-2 usually plays as a 4-2-2-2). But if you also see Enzo as working in tandem with the defensive mids, it’s a 4-3-3. Finally, if you consider that Matthew looks to be playing as the defensive half of the double pivot, you can also argue that we have a 4-1-4-1.

Well, I think we did in fact see all of these formations at various points in the game, although the 4-4-2 probably dominated. What I am trying to convey here is that switching formations can be relatively easy, especially if you stick with a 4-man back line. Change it up to a 3-man back line and it gets a tad more complex. But this really only needed Tyler and Enzo to adjust their roles.

Here’s the secret: it all rests on the performance of the defensive mids. In that light, it was a huge risk to start Matthew and play him the full 90. As it turned out, he played a brilliant game. Ando, of course, thrives in that role and did so on Saturday. But in a typical double pivot, one player takes the more defensive position and the other the more attacking position. That didn’t happen here. If you check the duo’s attacking and defensive statistics, they are very similar. Another way of looking at it:

Here I’ve merged the touchmaps for the two (as before, playing right to left). Ando is the black dots, Matthew the blue. Other than Matthew taking a corner, there’s not a great deal of difference. Moreover, they both played the entire width of the field. The conclusion being that they traded off the roles in the pivot. So, although Matthew looks like he was the single pivot in the 4-1-4-1, he did not limit himself in any way. It also means I don’t think they used it much.

Another requirement of course is that every player has to understand his roles in the ever-changing formation. I think this was very evident. I started out this post saying that the defense was disciplined. Well, it surely was. But so was the rest of the team, and it was glorious. Last week we had an incredible display of never-say-die play. This week we got to observe 90 minutes of intelligent organized soccer. And it’s tough to say which was the more entertaining to watch.