Rolling the Dice: A Look at Vancouver’s Luck Through 41 Games

With the Canucks having played their 46th game last night, we are over halfway through the NHL season and approaching the home stretch.  Since 41 games is a good benchmark, we can use this point to analyze how Vancouver has been playing and what specific things have been going their favour, and what’s been going against them.

One thing that we cannot predict in hockey is luck.  Luck plays such a large role in determining the outcome of games to the point that the NHL is identical to a league where only 28% of the games are determined by the better team winning and the remaining 72% of games are a coin-flip.  This should come as no surprise, otherwise you would see teams like Buffalo win-less and the Chicago Blackhawks and Los Angeles Kings of the world virtually undefeated.

But what is luck?  People tend to think of it as a vague, undefined, superstitious entity, but rather “luck” is everything that is outside of a players’ control.  Things like injuries, being subjected to variances in shooting percentages, or referees making bad or ill-timed calls are all included in “luck.”  

Continue past the jump to see how luck has been helping the Canucks this season.

Rob Vollman, who writes the annual Hockey Abstract, has a tool on his website that allows you to see how “lucky” each of the team currently is in-season.  We can use this to get an overview of the big picture and see how the Canucks stack up to the rest of the league, as we have passed the midway mark of the season.

On the whole, the Canucks are in the top half of the lucky teams, ranked 11th in luck score, and should be quite happy that all of the factors, outside of PDO, have been working with them.  To get a better picture of the Canucks luck we need to look at the component parts in each of the five fields.

Injuries happen all the time in the NHL, and it is expected that members of your team will get hurt – this is one of the reasons why having a strong farm system is so crucial.  When key members of your core are not able to play, this greatly effects your team’s performance.  We looked at the Canucks and injuries earlier this year before Dan Hamhuis was hurt, and found that the Canucks were in the top half of the NHL in terms of health.

Injuries can be measured by the Cap Hit of all Injured Players (CHIP).  CHIP assumes that dollar value of a players cap hit reflects their talent level, which for the most it does, and adds up the dollars being spent on injured players.  If your CHIP is high, your team is paying a lot of money to key injured players which likely affects the overall talent of the team relative to what it could be.  

Since the last article, Dan Hamhuis was injured and only just returned.  He has been the only real long-term injury for the Vancouver Canucks this year and the Canucks are still one of the luckier teams in terms of not being injured.  Only three teams through the 41 game mark have been less effected by injuries: LA, Montreal and Arizona.

Interestingly enough, we can compare the the rolling possession numbers of the Canucks vs their game-by-game CHIP values.  In the picture above the blue bars of the CHIP values represent Dan Hamhuis.  You can see the team’s possession rolling numbers have fallen without him on the roster, but interestingly enough the possession values have also risen with the return of Zack Kassian and his finger injury.

With Kevin Bieksa out long term and Brad Richardson also now injured, we will see the CHIP value and Man Games Lost for the Canucks increase through the second half – an example of how luck can turn on your team so easily.

The PDO of the Canucks is the only area they have not been lucky in, otherwise their luck value would lead the league.  As a refresher, PDO is a summation of team shooting percentage and save percentage at even strength.  Added together, over a season, they will regress to 100%, plus or minus about 2%.  If a team falls outside of that range, it can be said they are experiencing disproportionately good or bad luck when it comes to scoring goals and stopping pucks.

War-on-Ice has the PDO values for all teams and through 41 games, the Canucks were 24th in the league in PDO at a rank of 98.6%.  However, this has since regressed back towards the mean, largely thanks to the excellent recent play of Ryan Miller.  Broken down into its constituent parts, the Canucks had a shooting percentage of 7.1% (T-28th in the league) and a save percentage of .915 (T-22nd). 

We’ve already seen Miller’s save percentage regress, and while Vancouver is likely to score more goals than they have been of late to have their shooting percentage regress, the decline of both Daniel and Henrik Sedin seems to have left the team without a legitimate percentage-driving forward.  It’s possible that the Canucks’ PDO may not stabilize at exactly 100%.

Special Teams Index, or STI, works similarly to PDO but is the summation of your powerplay and penalty kill efficiency ratings.  In theory these will regress towards 100% over a large sample of games.  Because it takes much longer time to regress we don’t put nearly as much weighting into STI in the luck score but the Canucks have still be up there.

The Canucks are currently in 5th place in the STI ranking with a rating of 106.4%.  Their penalty kill has been one of the very best in the league and it rates in at 3rd place with an 87.0% efficiency.  Their powerplay needs some work (though is better than the Comets), as Vancouver was ranked 12th with an efficiency rating of 19.4%.  If both of your special teams are in the top half you can’t be too upset, however Vancouver’s powerplay has since fallen to 15th in the NHL thanks to a dry spell.

This might surprise a lot of people who are not involved with analytics, but winning one-goal games is not a repeatable talent, suggesting luck plays a large role in the short run.  Winning a lot of one-goal games does not repeat often season to season.  The linked article does a great job of going into depth on the whys. 

While Anaheim was trying to challenge the Canucks in the first half, Vancouver was one of the most successful teams at winning one-goal games too.  The Canucks were T-3rd in the league in winning percentage in one-goal games, but have since fallen to 4th.  With a 68.8% winning percentage, they are keeping pace with the Anaheim Ducks, New York Islanders, and Nashville Predators.

For another example for how winning one-goal games isn’t an indicator of team quality, look at the Buffalo Sabres.  They are 10th in the NHL in one-goal game winning percentage with a 12-6-3 record. However, they are also 2-25 in games decided by 2 or more goals, which explains why they are the worst team in the NHL.

In similar logic, if winning a 1 goal game is lucky, then so is winning an overtime (OT) game or a shootout.  The Canucks were also T-3rd in the NHL in OT-Game records alongside Buffalo, Columbus, Chicago, Montreal, and San Jose.  Especially since the shootout is a bit of a coin flip, we can guess that Vancouver’s record in the extra frame will be around 0.500 going forward.

***

It is interesting to see how much luck plays a role in the results of games.  The Canucks are currently 20th in Score-Adjusted Corsi and are sitting at third in the Pacific division.  If their PDO was in their favour, the Canucks could likely be at the top of the standings and the media would be talking about how Jim Benning and Trevor Linden saved the Canucks. 

But luck is a fickle mistress, and she can turn on you so easily.  The Canucks could have just as easily suffered a rash of injuries and not received as timely scoring and be firmly outside the playoff picture, like the Minnesota Wild.  

As we look towards the second half of the season, the Canucks may lose the luck they’ve been getting and find themselves contending for a top 10 pick. Or luck could work with them and boost them into the playoffs where anything could happen.  We’ll just have to watch the games to see what could happen.