Most fans watch the game of hockey from above, with moves and plays developing before them like a game of chess.  When viewed from this angle, Corsi makes perfect sense:  the more shots attempted, the more shots go in.  It’s simple, straight forward, and to the point, plus it’s been shown to correlate numerically with success.

Because of this simplicity, it’s extremely difficult to watch the talented Avalanche struggle so greatly with the concept, and it’s almost unfathomable why Patrick Roy hasn’t bought in yet.

While it’s easy to toss Roy under the bus as a no-nothing idiot, a terrible coach, and the source of all this team’s problems, that’s not actually the case.  However, in order to understand Roy’s misgivings with Corsi or the philosophy behind his entire offensive system, you have to go back to the beginning.

You have to put yourself in a goaltender’s state of mind.

Understanding Roy’s Offensive Zone System

Goalies hold these truths to be self evident: that not all shots are created equal, that they are endowed by their location with certain distinct qualities, that among these are speed, angle, and how much reaction time the brain and muscles have to make the stop.  For better or worse, this is where Patrick Roy’s mind lives, and this is where the Avs current offensive philosophy was born.

Roy’s system is based on the fact that on average, 91.7% of shots taken in the NHL don’t hit the back of the net.  Instead, they either go careening off in random directions or result in a stoppage of play and a faceoff.  Since it’s impossible to tell which way the puck is going to bounce in either situation, every stopped shot opens the offensive team up to a roughly 50% chance of a lost possession.

To complicate matters, any goalie worth his salt is going to stop an unscreened shot from the point.  Sure, one might slip by on occasion, but it’s a very low probability since he has the time and space to make the save.  It’s far more likely shots from that distance are going to end up in a faceoff/rebound situation, which often results in no goals and possession headed the other way.

However, chances from the home plate area are much harder for the goalie to stop.  He doesn’t have as much reaction time, and the angle the shot takes as it’s headed towards him is far more unpredictable.  Add in a screen or other traffic in front to block his view, and the probability of a puck going in is even higher.

If every shot is a gamble likely to result in the other team getting possession, it therefore makes sense to increase the number of high percentage shots and limit the number of low percentage ones.  This is the logic under which Roy’s system operates, and it’s why he doesn’t like any stat that doesn’t take into account how difficult it is for the opposing goalie to make the save.

To this end, the Avs have made up their own in-house metric.  Unfortunately, since it’s based on criteria from one of the most unique goaltending minds in history, it’s difficult from an outside standpoint to quantify.  This leads to frustration and misunderstanding, which is a battle Roy has been fighting since he implemented the system over two years ago.

There are some stats that do try to approximate the criteria.  War on Ice tracks High Danger Scoring Chances (HSC), which takes into account shot location and whether or not it was a rush or a rebound.  It unfortunately doesn’t look at traffic or other situational data like Roy’s metric undoubtedly does, but it does offer at least a starting point for understanding what’s happening with the Avs right now.

During the successful ’13-14 regular season, the Avs accumulated 726 of these high danger scoring chances for, which equates to 11.0 per 60 minute effort.  These numbers placed them 12th in the league overall.  However, when the overall number of shot attempts the team took are taken into account, the Avs were 4th at getting the most bang for their buck, so to speak.  It can be argued that while luck did play into the team’s high shooting percentage numbers that year, it may not have been just an illusion of success.  It may have been Roy’s system functioning correctly.

Their time of possession numbers supported this as well.  Even though the numbers aren’t publicly available, in NHL Network’s 2014-15 season preview of the team, it was mentioned that the Avs were actually 5th in the league in 5-on-5 offensive zone possession time per game with a 5:17 average.  This recipe of holding the puck in the zone while looking for a high percentage shot is a picture perfect example of how Roy’s system is designed to run.

Now, skip to last season.  The Avs high danger scoring chances for plummeted by nearly 100 attempts over the season to 632, good for only 27th in the league.  Instead of averaging 11 of these high quality chances per game, the team only managed 9.6.  To make matters worse, when the total number of shot attempts are taken into account, the Avs were only 11th in the league in picking their spots.  They shot fewer times and from worse locations, so it’s no wonder the system fell apart.

This year doesn’t appear to be going any better.  Although the numbers this early in the season are still too small to draw big conclusions from, the Avs are currently 25th in high danger scoring chances and a lowly 17th in shot selection.  With an offensive system based so strongly on these two principles, something is very clearly wrong.

Why the System is Broken

Part of the problem is that this philosophy is very cerebral.  It relies on players making very good decisions on when to shoot, where to be with and without the puck while in the offensive zone, and how to read the situation to judge when it’s time to crash the net to make a play.  It also strongly leans on a solid net front presence screening the goalie and knocking home rebounds, as well as good passing to establish an effective cycle.

At the moment, the Avs are failing in each of these categories.

The biggest problem is the refusal to shoot by many of the team’s top players.  Duchene, Landeskog, MacKinnon, and Iginla’s rate of shooting (shots/60) are currently the lowest they’ve been in their entire careers. Wayne Gretzky famously said, “you miss 100% of the shots you don’t take”, so even if every shot is viewed as a gamble in this system, they’re still a basic requirement for it to function.  Maximizing shot selection doesn’t work if shots aren’t being taken at all.

To make matters worse, the shots that are occurring right now aren’t high danger chances.  In ’13-14, the Avs had four players who averaged over one high danger scoring chance per game (Duchene, MacKinnon, Landeskog, and Stastny).  Currently, they have none.  Iginla and Duchene come close with 9 in 11 games, but players like Cody McLeod, Blake Comeau, and John Mitchell are sitting higher in the ranks than Landeskog, MacKinnon, or Tanguay.

It’s not that they aren’t getting their opportunities.  Night in and night out, most of the Avs top players are passing up prime scoring chances in search of the perfect play.  While Roy’s system does encourage some of this decision making, the entire point is to create high probability situations.  Now, even when these opportunities come around, the players aren’t capitalizing on them, either out of self doubt or the desire to find an even better option.  It’s created a negative feedback loop:  the less the Avs score, the less they look to score.

Their positioning in the offensive zone hasn’t helped.  By and large, they’ve passed the puck around the perimeter of the zone without looking to penetrate into the higher scoring chance areas.  In comparison to the 2013-14 season, the team’s average shot distance has increased by well over three feet.  For a club that relies so heavily on high percentage shots for success, this added distance increases reaction times for opposing goalies, decreases the club’s shooting percentage, and drops the number of goals scored.  It also critically undermines everything the coaching staff is asking the players to do while in the offensive zone, causing the system to implode upon itself.

The general lack of netfront presence is also troubling.  Management did a great job this summer of adding big new forwards like Soderberg, Grigorenko, Skille, and Comeau, but they too have been caught in the perimeter play trap.  The biggest reason Roy still (frustratingly) relies on Cody McLeod during the occasional power play or 6-on-5 situation is his willingness to go to the front of the net and screen the goalie.  Granted, he hasn’t been great at it this year, but no one else has stolen the job from him either.  Under the logic of Roy’s system, screens increase the probability for a goal, so removing this cog has further undermined an already struggling system.

All of this play around the outside has also lead to another less than outstanding side effect:  the increase in unnecessary passing.  Unfortunately, this generation of Avalanche players has never demonstrated the ability to make consistently clean, accurate puck movements.  Poor judgement calls and bad execution have always somewhat limited their ability to cycle in the offensive zone, but now that shots are being sacrificed in lieu of more passes, the problem has compounded.

Stopped shots result in around a 45% chance that the other team gets possession, but flubbed passes and turnovers result in near 100%.  It also often means the leaving the zone without a Corsi event for.  Even when the Avs shot selection system is functioning properly, their Corsi for totals are still likely to remain slightly under the league average;  however, the club is currently averaging 45.0 Corsi events for per game, which is 29th in the NHL.  For a club based on offense, this amount of possession bleeding is far from a recipe for success.

Even if you disagree with the philosophy behind Roy’s system, it’s very difficult to argue that what we’re seeing right now is anything more than a shadow of what it’s designed to be.  Instead, this shot selection philosophy has become twisted into a very poor excuse for players – particularly the top forwards – not doing their jobs. After the Carolina game, Roy called the team “fragile”, and it’s very hard to argue that point when it comes to their shooting, positioning, decision making, and overall play on offense, especially after something doesn’t go their way.

In short:  it’s unfair to blame a system for the team’s failure when it’s not even being properly run.

Righting the Ship

Even though it’s not Roy’s fault his system is failing, as a head coach who has opted for a philosophy that goes against the mainstream logic, it is still his responsibility to get it functioning again.  The Avs’ issues in the other two zones and their desire to be an offensive team just add that much more importance to the matter.

One method for fixing their o-zone play would be a clear focus on shots and shot location in practice.  Reward those that are driving the net and call out those that are only playing around the outside. Remind the players that shot selection doesn’t mean not shooting, and make it clear that getting the puck on net is the number one priority when they enter the zone.  Set a weekly shot total goal if you must, but do something that forces players to think about getting the puck towards the goalie over anything else.

The other would be a focus on positioning and other details in the offensive zone.  Drive home the importance of going into the home plate area and working without the puck to get open.  Also emphasize the critical nature of net-front screens and crisp, tape to tape passes with fewer mental errors.  Take the players’ minds off whether or not their shot will go in or whether they should be passing or whatever other nonsense is polluting their noggins right now and ask them to focus on the smaller details that they’re still able to control.  Give them something concrete to hold on to, something they can actively work on, and take baby steps towards breaking this slump.

Whether the answer is one of these methods or another one entirely, something has to be done by the coaching staff to get the players out of their own heads and playing hockey again.  It won’t be some vague regression or puck luck that does it – if this team wants to save their season, they need to start focusing on what they can actively do to generate more offense and jump start their system. Otherwise, one of the most talented rosters the Avs have been able to ice in the past decade will go the way of the others: to a disappointing early exit and a waste of their young stars’ best years.

It’s time for the Avs to strive for something different, and fixing their offensive zone play is a huge step in the right direction.  This scheme was born from the mind of a goalie; now it’s time for him to rally the team and find a way to save it.

Andi Duroux

As a Colorado native and relative newcomer to hockey, Andi grew up following college basketball before switching sports in 2010. Since that time, she's developed a passion for learning about the icy game and sharing that knowledge with others. Her focus on history, in-depth analysis, and statistical research provides a unique take on both the Colorado Avalanche and the NHL as a whole.

  • Eric C

    Great article