Silver linin(xG): Underlying numbers paint a better picture for MNUFC

I’ve got something shocking to tell you: according to the underlying “Expected Goals” numbers, the Loons just had their best stretch of the year.

I can hear you now: Game’s not played on a spreadsheet, mate! Duly noted. Also noted: the only statistic that counts in the standings is the one that shows that Minnesota’s earned one point over its past four games, tied for its worst four-game performance in what has been an extremely streaky season.

But one of the things I’ve learned, from enough reading about numbers in sports, is that my eyes are going to lie to me. I’m going to remember the bad much more than I’m going to remember the good. I’m going to remember Luis Amarilla hitting the post against LAFC on Tuesday and missing an unmissable header against Portland last week, instead of remembering that he was in position to score in the first place, which is not nothing.

This is why I try to keep tabs on xG. The frustrating thing about xG is that it’s not obvious how it’s calculated, nor is it consistent from source to source. The MLS website will tell you one thing, FB Ref will tell you another (its data from the service StatsBomb), and American Soccer Analysis will tell you a third, and I’m sure there are other sites and apps that will give you other numbers as well.

That said, I don’t think the inconsistency means that the numbers are useless. They do tend to mostly agree with each other, in terms of direction and magnitude. For example, ASA had the Loons beating LAFC 1.88-0.89 in Tuesday’s xG battle; at FB Ref, it was 1.2-0.8; and for the MLS site, it was 1.6-1.0.

Is that enough caveats? Okay, now let me throw some numbers at you.

According to the numbers on all three sites, the Loons’ loss at Portland and draw with LAFC was the team’s best two-game stretch of the entire year.

According to the numbers on American Soccer Analysis, the last four games – the one-point-out-of-twelve run – was the Loons’ third-best four-game stretch of the season. (On the MLS site and on FB Ref, it’s sixth.)

Even though the Loons got absolutely hammered at Real Salt Lake, they won the xG battle in the following three games (again, looking at ASA’s numbers). Amazingly, that’s just the second time all season they managed that.

Why mention this? It’s partly as a silver lining, but also partly to contrast MNUFC’s latest skid with the team’s struggles earlier this season. Their previous worst run – loss at LAFC, loss against Cincinnati, loss at Seattle, draw with the Galaxy – also coincided with their worst run of the season in the underlying xG numbers. Not only were they losing, but they deserved to lose.

That’s not true on this run. They punted the game in Utah, and were the victims of “five minutes of madness” against FC Dallas, to quote Adrian Heath, but they could and maybe should have won against both Portland and LAFC.

“As I said to the players after the game, we’ve had three really big games, and I think we’ve played really well in all three,” said Heath. “We didn’t get what we deserved, I don’t think we did in Portland… [and against LAFC], to play the way we did and come out with only a point, it’s disappointing. But I couldn’t be any more than pleased with the players.”

Obviously, xG isn’t a guarantee. It is notable that the team’s two worst xG performances this year, away against New York and against the LA Galaxy, both resulted in wins, back when Dayne St. Clair could be counted on to steal a point or three every match.

But it’s something to take heart in, as the Loons enter their final three-game stretch of the year. It’s a difficult run-in, despite what the standings say; Sporting KC is unbeaten in its last five games, San Jose has won its last two home games (including beating LAFC), and Vancouver just beat the brakes off the Galaxy on Tuesday.

It feels strange to say after one point in four games, but if the Loons are going to push their playoff bid over the line, they’ll need to keep playing like they have in the past three games.