Possession Obsession, or How I Learned to Stop Worrying and Drive the Play

When the Wild entered this season, few expected them to be an elite team. Good, certainly, but cup hopefuls? not so much. Then the season began, and the Wild began dominating opponents like few had seen before; the team was immediately in the discussion as one of the best in the league.

Since then, we've seen practically everything but the Biblical Plagues, and the Wild's playoff chances are a measly 26% (or thereabouts). So, what went wrong? Many things. The blame spreads from poor defensive play, to poor goaltending, to Vanek and Koivu simply not being the leaders we need them to be.

Take, for instance, this chart of the Wild's Save and Shooting Percentages.  All the stats here are from War on Ice, and are at Score-Adjusted 5v5. The lines are based on 5-game averages. (also, all these stats are from prior to the SJS game on 1/6).

Here the green pips and line is the goalies' 5v5 score-adjusted Sv%, and the blue is the team's sh%.

Looking at the green line for a moment; it's clear why the Wild struggled so much from games 12-30; the goalies threw out individually good performances, but on average they were below a .900 sv% for a good long while. Only in the past few games has the average moved back into normal territory. Of the Wild's 37 games, only 21 have seen over-.900 sv% from the goalie. That's not good.

When a goalie is playing poorly, teams need to shoot the lights out to have a hope of winning, and as you can see that didn't happen. The Wild's sh% was very low at the same time as the sv% was at its lowest. In other words: as the goalies started to play poorly, the pucks stopped going in. That creates losses, and lots of them. Fortunately, the sh% not only was not QUITE as awful as the sv%, but it normalized more quickly as well. As the goalies have heated up, however, the shooting has cooled off slightly; this is why we've lost games where the goalies have played well lately.

Because Sh% and Sv% are so random and vary so wildly, stats guys have started looking at puck possession as a measure of a team's play. The standard stat for measuring possession is Corsi.

So, to look at the Wild's level of play, let's look at their #FancyStats. The following chart is Corsi For per 60, Corsi Against per 60, and CF%, all at 5v5 score-adjusted. (again, these only go through the Dallas game on 1/3)

Here, the green is CF/60, the yellow is CA/60, and the blue is the CF%. You'll notice that the green and blue lines follow each other, while the yellow is their inverse.

The Wild started off the season generating a large number of Corsi events per 60 minutes. This continued throughout the season until around the 27th game (which was against the Sharks). After that, the Wild stopped generating shots at the same rate, and we have since them started to see a decline in the Wild's ability to control play.

The Wild's shot suppression has been suffering from about the 15th game on; the yellow line seems to climb endlessly. The Wild have been slowly allowing more and more shots as the season has continued, and began being outshot on average at the 30th game.

Since then, the Wild have averaged below 50% CF%. Whether this is Mike Yeo abandoning the system that found the Wild leading the league early on, or the players not executing that system can't be known. What is known is that some of the roster decisions being made can't be helping.That said, Stu Bickel being in the lineup isn't enough to make this a sub-50% possession team; his 3 minutes on the ice each nice don't have that much power. The Wild have certainly had struggles with injuries and illnesses, but not to the extent of other teams, and the possession statistics don't line up with missing their players, at least not enough to indicate a strong correlation.

In other words, it's not a great argument to say that Yeo's lineup decisions or injuries are hurting this team significantly; the Wild have been trending in the wrong direction for most of the season, and no single player or line change has been able to stem the tide. This means that the problem is not with who is on the ice, but what they are doing.

Back to Basics

What likely happened was: the Wild's offense wasn't producing early on (because of their horrendously low sh%). As their process seemed to be failing, the Wild- either by design of the players or coaches- stopped playing with a focus on driving possession, instead trying to focus on defense. Defensive shells have been shown time and again to fail in the NHL, and we've seen the results. Factor in two struggling goalies, and you have a season on the brink of being a lost cause.

So, what's the solution? The Wild need to return to their style of play from the beginning of the season. Aggressive forechecking and crisp passing has been lacking recently, and they need to return. The Defensemen jumping to join the attack, when appropriate, needs to come back. Maybe most importantly, the Wild need to understand that nothing they do in front of Kuemper or Backstrom can make them play better; players have little-to-no effect on their goalies' Sv%, and therefore they shouldn't worry about it.

Either way, the Wild need to turn things around quickly, before their playoff hopes die out completely.

--------------------------------------------------------------------------------------------------------------------------------------------------------------

All stats from War on Ice, prior to the 1/6 San Jose game.