What Grades Are The Loons Making?

It’s safe to assume you’ve read team CEO Chris Wright’s various interviews over the last few weeks asserting his confidence in Adrian Heath, Manny Lagos, and the Minnesota United FC backroom team. If not, here’s the money quote given by Wright last Thursday to MLSSoccer.com’s Sam Stejskal:

There’s a lot to unpack here. There’s the moving parts comment, suggesting the headwinds from transitioning to MLS while in a temporary stadium. There are the “circumstances” he mentions, the injured players, the upcoming window, that all combine to give him optimism for the future.

As someone who briefly wanted to be a college professor, I issued enough course grades to say that Wright’s optimism should call for an “incomplete.” However, it also makes me question what Wright’s rubric looks like, as well as what sort of curve Wright is working with.

So, like any good nerd, I took a trip to WhoScored.com’s page of 2018 MLS team statistics, consolidating them down to 66 metrics, either as season percentages or per game numbers. I then pulled the ranking Minnesota holds in each, as well as the Loons’ relation to the average number in the league, with additional coding to note when a category marks higher frequencies of negative actions (i.e. cards, missed tackles, shots blocked by the opposition). Finally—similar to my prior article this year—I ran a correlation function to see which items were linking most directly to points per game in the league this year. Not surprisingly, many of the same aspects are still getting strong correlations—goals scored continue to be tops, as do corner kick accuracy, shots on target, assists, and overall shots.

Also not surprisingly, the Loons aren’t looking too hot.

With three games to go in the season, Minnesota United FC rank worse to median in 39 of the 66 metrics pulled, and worse to average in 15 of the 20 metrics with the closest relation to points per game.

Goals per game currently reigns king in terms of stats correlated with success; Minnesota ranks seventh from the bottom. The Loons are also third-worst in accurate corners and corner passing percentage, both in the top four indicators. They’re second-worst in shots per game (the sixth-closest indicator), fifth-worst in shots per game on target (the third-closest indicator), fourth-worst in key passes and key short passes per game (fifth-closest and eighth-closest indicators), second-worst in goals per game from shots within the 18 yard box (tenth-closest indicator), and fourth-worst in shots per game from open play (eleventh-closest). Even with the injuries to the team resulting in significant investment up front, the Loons’ vaunted offense remains relatively anemic.

The Loons are also bottom of the league in generating shots from set pieces, bottom in earning penalty kicks, and worst in getting key passes from long balls. Their overall passing accuracy on low-percentage passes (crosses and long balls) is poor, as is their completion percentage on short passes. They’re failing to get accuracy on free kicks, and thus aren’t generating key passes from them.

Where is Minnesota doing well compared to the rest of the league? There are 19 metrics where the Loons are in the top 10, above league average, and where the metric is neither counting negative information nor negatively correlated to success. The team is generally active defensively, attempting and completing the highest amount of tackles per game, and blocking and intercepting a high amount of balls. They’re also taking smart shots when available, taking the highest proportion of shots in the league from within the 6 yard box and getting the third most goals per game in front of the goal mouth, and converting chances well, with good percentages each of their shots being on target and their key passes and crosses converted.

But Minnesota are also above average in nine indicators negatively correlated to success, five (six, counting raw fouls committed) of which come from being negative counting stats. The team has racked up the fourth highest amount of fouls and yellow cards in the league this year, along with the third highest amount of red cards. They’re also top ten in proportions of long and crossed passes, neither of which are resulting in higher than average accuracy rates (second last in long pass accuracy and cross accuracy).

When taking a systematic look, it won’t surprise you that the teams that rank highly in different metrics—and specifically rank highly in the most determinate of metrics—do well in the league. To find this out, I took each team’s ranking in each of these stats, normalized them into a 50-100 score based on an exam grade, then multiplied each metric’s normalized scores by the metric’s absolute correlation value to points per game to put a lower weight on a team’s performance in less meaningful categories, resulting in a metric we’ll call SUM (Soccer Undertakings Metric). The results surprised me in how close they mirror both points per game and my eye test of a team’s performance:

Indeed, certain teams that seem to be performing well on paper are doing well in the standings. The top seven in SUM are all in the top seven for points per game; Atlanta leads in both categories. The teams that diverge all make sense: Montreal have mainly worked through strong defense down the stretch (a limitation of the data, I’ll confess) but seem to be reverting to what their stats suggest, Toronto and Seattle spent lengths of time playing abjectly due to injuries this year but have been transformed teams at full strength, DC United added Wayne Rooney. Nonetheless, the teams that have been poor to watch are generally in the bottom of SUM, and the better ones rank near the top.

And it’s not a shock that Minnesota rank 20th in SUM, only fractions of a point ahead of Chicago. The Loons simply don’t generate enough offense to be competing with the top teams, and while their defensive record suggests effort, it also underlies the cutting point that the Loons simply don’t have the ball at their feet enough to be in the right position to score. A metric I omitted due to a lack of unrounded available data was the amount of time a team spends in each third of the field. Minnesota spends the second most time in its own third in the league. When a team is pinned back like this, it means that “hero ball” chances rule the roost, kicking over the top to an on-rushing winger or striker, lobbing in crosses at will, and hoping for the best. As a result, higher-percentage chances are at a premium.

To say that many things aren’t going well is an understatement. Minnesota is on track to concede one fewer goal than last season’s history-making 70, with only Orlando’s futility to thank for having a worse defense this year. But the offense isn’t firing enough either. The ideal balance in MLS appears to be a team’s ability to assert itself in the opposition’s third and to generate tons of shooting chances, while maintaining enough rearguard composure to stymie the opposition. The numbers show that Minnesota doesn’t do this, and while the metrics slightly understate the team’s performance in the table, it’s not enough to suggest the playoffs are in the offing with this team as the base for next season.

But are the grades on the field truly passing? Let me leave you with this graph that maps where the Loons stand when taking all 66 of those metrics and directly converting that 50-100 score into a letter grade; while the median grade the team gets borders on passing at a C- to D+ level, that mode is frightening:

Data via WhoScored.com