I’m probably making way too much of Louisville City’s dire performance in Harrisburg over the weekend, but the result still leaves me scratching my head to the point that I felt the need to write about it AGAIN. The guys on Barrel Proof last night talked about whether we’ll see the 3-4-3 again, or if James O’Connor will revert to his trusted 4-2-3-1 based on last week’s result. That caused me to want to look at the last two games and figure out why we definitely shouldn’t do that.

As I mentioned on Monday in the STATCAP, passing in this game was bad. Our two midfield hubs, Paolo DelPiccolo and Guy Abend, had subpar games, to be sure. To illustrate that point, here’s their full-game passing chart from the totally awesome Pittsburgh game:

Notice how the two central mids mostly stayed in the middle of the field, and most of their passes in Pittsburgh’s half were pointing toward the Hounds’ goal line or upfield toward the sideline, where we can assume Kyle Smith and Oscar Jimenez got the ball. That’s good!

Now let’s look at the first half passing chart for the Harrisburg game.

You’ll notice a couple things: First, they’re playing way wider than they usually do. Second, there are a lot fewer passes even being attempted, and the ones that are are horizontal or backwards. Third, very few balls are getting to the left wing, where Oscar generally is. That’s a problem, since he’s the team’s second best chance creator!

I mentioned earlier this week that while I hadn’t watched the game, I was pretty sure Richard Ballard’s halftime sub appearance for Sean Totsch meant that O’Connor was scrapping the three man back line for the second half against the City Islanders. That led to a transition for the central defenders from a touchmap that looks like this:

where the defenders are mostly higher than their own 18 yard box, even at midfield in Totsch’s case on the right, to a touchmap that looks like this in the second half:

where guys are barely in the opposition half at all, and much more active in their own 18 yard box. That’s not what you want. Combine that with they LouCity defensive midfield’s passing chart in the second half, which looks like this:

and what you’ve got is a team that basically played the entire second half on its heels. Paolo and Guy completed, by my eyeball count, just four passes that could be considered attacking, and only two of those were made in Harrisburg’s half of the field.

This, in my opinion, is the direct result of O’Connor’s switch to the four man back line in the second half. I don’t think it was a good move, and these charts bear that out. No, the game was not going great in the first half, but at least it was scoreless. Changing tactics midstream just made it worse.

The three-man back line works because we have tall-ish central defenders playing it that are also adept at playing defensive midfield, meaning they’re good at tracking runners and playing on the ball. This allows O’Connor to send his fullbacks higher up the field and stretch out the opposition defense. Pulling the fullbacks back into a more defensive position where they can’t link up from defense to offense as easily led to a touchmap for City’s attackers, Lancaster, Ownby and Davis, that looked like this in the first half against Harrisburg:

To one that looked like this in the second half:

Fewer touches, far more defending from the attackers, and barely anything in the opposition penalty area. Compare that with this from the Pittsburgh game:

and you’ll get kinda sad.

The 3-4-3 setup is the best use for the players we have at our disposal. It opens up space for our wingers to play the ball into, which is the preferred method of receipt, a term I just made up, for Cam Lancaster and company. It maximizes the talent our central defenders have at their disposal. I don’t know if the weak link in this chain is the central midfield; maybe they just had a bad game. It happens. But the “chuck it, let’s flatten out in the back” plan didn’t work well early in the season, and it certainly flunked in the second half against the City Islanders. Let’s not do that again, well, ever, until we’ve got a different crop of players that suit that style. This roster was and is fine doing the things it was doing in the first half and the previous three or four matches.

COME ON, CITY.