A picture of a Predator and a Wild(?) in which neither one of them is possessing the puck. That's called "foreshadowing."

In case you don’t recall, the Minnesota Wild got off to a hot start this season. By November 19th, they were first place in the NHL and at or near the top spot in the league for a full month. After 30 games, the Wild had a sterling 20-7-3 record and their fans were justifiably thrilled.

What reasons were given for the unexpected success? New coach Mike Yeo had a history with the players from his time in the AHL and had them buying in to a new system and a culture of winning. This system focussed on blocking shots and making sure that what shots did get through were from the outside and easy to see for their goaltenders. The team was working hard and working for each other. The Church of Yeo was accepting new converts daily (sorry, Ms. Conduct).

The proponents of advanced statistics said that was all horse-twaddle.

Gabriel Desjardins started the parade-raining before the Wild even hit first place, pointing out that their shot totals at even-strength weren’t substantially different than their previous season, when they finished 12th in the West. Derek Zona continued the assault with an in-depth look at their underlying possession statistics, which not only didn’t paint a rosy picture, but also labelled the Wild as the worst team in the entire league at puck possession. Even I chipped in, pointing out how often the Wild started shifts in the defensive zone as compared to the offensive zone.

The golden statistic for most advanced stat lovers early in the season is Fenwick Tied. A team’s Fenwick number is basically just their shot differential at even-strength, including goals, shots on net, and missed shots. “Tied” indicates that the statistic is only looking at shots when the score is tied, because teams tend to attempt more shots when they’re losing and allow more shots when they’re winning.

Bruce Kluckhohn/NHLI via Getty Images

The reason Fenwick Tied gets so much attention is that it has been shown to be a very good indicator of future success (or failure). Teams with terrible Fenwick Tied numbers tend to lose more games as the season goes on and the Minnesota Wild had the worst Fenwick Tied number in the entire league. Considering they were first in the NHL at the time, pretty much every single hockey blogger outside of Minnesota said that the Wild were due for a fall.

To sum up: the Wild were getting lucky and were bound for regression to the mean, which is just a fancy way of saying that the Wild were going to start losing games. A lot of games.

Now, no fan of a team wants to hear that their team is terrible. Fans tend to get a bit defensive about such things and when said team happens to have first place in the NHL in their corner, they tend to fire back with both barrels. Over at Hockey Wilderness, SB Nation’s Minnesota Wild blog, “regression” became a punchline and they began a series of joke power rankings to poke fun at the Wild’s low ranking throughout the league despite their position at the top of the standings.

Then the regression actually happened and it wasn’t so funny any more.

When the Minnesota Wild experienced their first big losing streak in December, it was easy for Wild fans to point at the team’s injury troubles and excuse the slide on that basis. But the team hasn’t won more than 2 games in a row since December 10th. On Sunday, they managed to halt a 7-game losing streak with a victory over the Boston Bruins and now have a 26-24-9 record.

The Wild won 20 of their first 30 games. Unless they win tonight against the Panthers, they’ll have won just 6 of their last 30 games. They went from 1st in the league to 21st in just 53 days. Over at Hockey Wilderness, they’re not joking about regression anymore. Now they’re talking about how high their draft pick will be. Poor Ms. Conduct has stopped worshipping at the Church of Yeo and has turned to Zen Buddhism.

The collapse is so complete that there’s nothing satisfying about saying, “See, we were right all along.” The Wild have been even worse than the stat-heads imagined. It’s not fun to kick someone when they’re down, especially when they’re as far down as they are right now. It’s just sad, awkward, and mean.

There is at least one positive for the Wild: they’re no longer the worst puck possession team in the NHL at even-strength. No sirree. They’re second-worst.

Hal Gill won't help the Predators possession. Frederick Breedon/Getty Images

Most of the teams near the bottom of the Fenwick Tied rankings are teams in a similar position to the Wild: out of the playoffs or struggling on the bubble. The one exception is the only team below the Wild: the Nashville Predators. The Predators are currently sitting 5th in the Western Conference and are 10 points clear of the 8th place Los Angeles Kings. They actually have 7 more points than the 3rd place San Jose Sharks, who are in first place in the Pacific Division.

The Predators have an impressive 35-19-6 record and a plus-13 goal differential, while boasting one of the lowest payrolls in the league. It just makes me want to throw the Jack Adams at Barry Trotz and be done with it, but there’s the troubling issue of their puck possession statistics at even-strength. How in the world are they so comfortably in a playoff spot when their Fenwick Tied is literally last in the league?

Nashvile Predators Possession

Statistic Percentage League Rank
Fenwick Tied 44.34 30th
Fenwick Close 45.01 30th
Fenwick Tied Home 44.65 30th
Fenwick Tied Road 43.91 29th
Fenwick Up 2 46.32 8th
Fenwick Up 1 42.33 25th
Fenwick Down 2 55.23 19th
Fenwick Down 1 47.80 30th

The only thing I can point to is the Nashville powerplay, which is second in the league at 21.7%. Combine that with the solid goaltending of Pekka Rinne and you have a formula for a lot of close games. The Predators have been fortunate to win a lot of those games and have an 18-8 record in games decided by 2 goals or less. They are also second in the league in empty net goals, which has helped them seal victories in tight games.

Is it just luck and a great powerplay? Are the Predators due for a fall? How good will Pekka Rinne need to be to give Nashville a long playoff run?

Comments (9)

  1. Can you tell me the Fenwick of the Wild before and after the “regression”?

    I can tell you some fun stats. Here are the the stats after the collapse (which coincided almost simultaneously when Bouchard, Latendresse, and Koivu all got injured) followed by the stats beforehand in parentheses.

    G/G 1.79 (2.63)
    GA/G 3.10 (2.13)
    PP% .132 (.170)
    PK% .806 (.855)
    Save .902 (.934)
    Shooting .065 (.102)

    Clearly both the offense and defense fell in the tank. You could say goaltending too but I argue save percentage is much more a function of defensive system than goaltender performance.

    Yet your theory, because you are using the Fenwick to predict success, is that the Wild were routinely outshot and therefore lucky to win those games and should not have won those games, because they were out shot.

    Now I don’t have access to advanced stats and will admit mine are not going to be accepted, but I have gone through game by game and noted shots, missed shots, and blocked shots, and here’s what I’ve come up with. Granted this is going to have “home bias” and “score effects” and etc, etc, but this is the closest I can get:

    Fenwick 0.47 (0.44)
    Corsi 0.46 (0.43)
    SF/G 27.4 (25.8)
    SA/G 31.5 (32.1)

    What I’m seeing is that all of the shot based totals have improved. According to your theory, the Wild should have a better team. Yet:

    Win% 0.31 (0.72)

    Clearly that is not the case. And aside from the statistics, anyone watching the games can clearly tell you it is night and day in terms of how they play.

    If these stats can’t tell the story about the Wild’s collapse, what can? Why is it they “improved” statistically in every advanced stat category yet got worse statistically in every traditional category?

    I firmly believe that all shot based statistics are an incredibly flawed “tool” as it violates the rule of ceteris paribus. Specifically, all teams do NOT run the same system. Coaches like Yeo and Trotz and Hitchcock are more than happy to give up as many low percentage shots as the other team would like, because it is often equivalent to a turnover. If all coaches ran that system, Fenwick/Corsi would be a very valid measure and would likely have far more correlation than goal differential statistics.

    I’d challenge you to explain with your statistics WHY the collapse occurred. Because I would counter that the statistics showed the exact opposite.

    • They have improved their Fenwick and Corsi…barely. It’s such a slight improvement and they were playing so far over their heads at the beginning of the season that it’s not making a huge difference now. The improvements that you are noting there minuscule.

      Like I said, no one predicted the kind of slide the Wild are currently on. In fact, their record over the last 29 games indicates that they have actually been extra unlucky. In a way, they regressed to the mean and then sped right past it.

      I understand the argument regarding different systems, but it doesn’t add up. Those who were tracking scoring chances, which are shots from in close rather than low percentage shots from the outside, were finding that the Wild were also getting badly outchanced as well as outshot. So far, no one has reliably shown that teams can influence shot quality. Puck possession metrics that use shot totals like Corsi and Fenwick have repeatably been shown to be reliable.

      I’ve watched the Wild. They absolutely look worse now than they did before. But is that because they’re playing worse now or is it because their shooting percentage and save percentage were unsustainably high during the first 30 games, making them look much better than they actually were? The eye test isn’t very reliable for that. Maybe you’re seeing a team that doesn’t have the same optimism or some other intangible, but it’s likely because they’re not getting the bounces they did at the beginning of the season and they’re getting frustrated by losing so many games.

      Again, like I said, they have improved their possession numbers ever so slightly, but it’s not a big enough change to suggest that they have significantly improved as a team. Their winning percentage over their first 30 games and their last 29 games won’t correlate perfectly with their Fenwick numbers over those periods of time; that’s not how it works. But over the course of time (like a full season or two), those wins and losses will even out and head towards a reliable indicator of team talent, such as Fenwick Tied.

  2. The Wild were definatly playing above their heads early in the season, but were still a playoff team had they stayed healthy. The drop off the cliff is largely because of injuries.

    The Wild were not a deep team, and that is the problem. Them being succesful had so much to do with them being healthy it’s not even funny. They had 6 players on their team, that are actually top 6 players, and 3 of them are borderline(Cullen, Seto and Latendresse). While Fletcher has done a fine job accumulating good offensive prospects, unfortunatly none of them are in the AHL yet to help the big club when injuries like this occur.

    Look at the rosters that the Wild have been forced to trot out there because of the lack of depth, i contend no team would be able to sustain the success they had with their full team trotting out 5 guys a game that have no business in the NHL.

    • I think there’s something to this, actually. The Wild had a good enough start that even with regression they could have still been a playoff team. The injuries, particularly to some of their core group, really exacerbated things and made the slide a lot worse.

      The issue is that contending teams have the kind of depth that can handle those kinds of injuries. The Wild aren’t there yet, but, quite frankly, weren’t expected to be. They still have time to rebuild. Hopefully, the early success of this season doesn’t take them off track and convince management that the team is closer than they really are.

      • But that’s the rub isn’t, most Wild fans knew they weren’t a legit Stanley Cup contender(Baring Backstrom getting Thomasesque hot). But we did know we were a playoff team when healthy, 5-8 seed type of team. The stories that got Wild fans hot, were ones saying they are the worst team in the league.

        The team from the first half of the season, was a slightly above average team getting lucky. The current, injury riddled team, is quite likely the worst team in the league because they are fielding almost two full lines of players that belong in the AHL.

        So saying they regressed because the advanced stats said they would is disingenuous, it’s only the same team in uniform.

        • You knew that when exactly?

          When the wild were in first place because of a hot start? If Minnesota had 15 wins in their first 30 games instead of 20 and were in 9th instead of 1st, would Wild fans have known that they had a legit playoff team?

  3. Predators fans, don’t bother. Mr. Wagner is a Canucks fan/blogger. This is all nothing more than confirmation bias for his own agenda (pro-Canucks/anti-Canucks peers e.g. Wild, Predators).

    My assertion is that Mr. Wagner is hardly a legit source for impartial or non-Canucks information, as evidenced by his non-Score.com affiliation and refutation of facts that do not support his argument (first comment) because they’re only “slight” changes instead of, presumably, “large” or “gargantuan”. Inconvenient, perhaps, is the better word.

    If a statistical argument is based on “the numbers don’t lie”, then the numbers better support your argument. If you have to sweep contradicting numbers under the rug, then it’s not much of an argument.

    This round goes to Jarick.

    Here’s the thing: I’m a Wild fan. I admit it. If, Mr. Wagner, your inclination is to discount what I’m saying because of it (because it’s been established that the data doesn’t support your argument), then QED.

    Like I said, I wouldn’t get too jacked up about this if I was a Predators fan. It’s just your day in the Canucks fan angst barrel apparently.

    • *sigh*

      Ad Hominem attacks are so much easier than actually addressing the argument. I’m sorry if I find a change in possession statistics that moves the Wild from 30th to 29th in the league relatively insignificant.

      I actually quite like the Predators. I wrote this post not because I dislike them or want to discredit them: I’m actually curious as to why and how they’re achieving their current success despite their objectively terrible possession numbers at even strength. I would love to hear an explanation that doesn’t involve fallacious reasoning involving intangibles. If Fenwick Tied is faulty, I’d really like to know.

  4. How in the world are they so comfortably in a playoff spot when their Fenwick Tied is literally last in the league?
    *******************************************

    Pekka Rinne! End of story!

Leave a Reply

Your email address will not be published. Required fields are marked *