Breaking Down: The Struggle with Quality

As you may have noticed, the beloved Avs have been a little bit of a mess over the last 2 months. Since the end of November they’ve gone 8-14-4, which is dead last in the NHL over that time period. This isn’t a slump, it’s half the season to date and they’ve completely collapsed after a solid start. The good news is that they’ve still managed to hang on to a playoff spot for now. The bad news is that at their recent pace they’ll end up with around 77 points and miss the playoffs by a mile.

The common assignation of blame is to the Avs goalies, who by no means have been saving the day but aren’t as horrendous as it seems. Nine teams have a lower 5v5 save percentage over the last 2 months including Tampa and San Jose who have around double the points of Colorado in that time. Save percentage has slipped 2% but shooting percentage has fallen by more than 3%. So what’s going on here?

I’m not a big believer in luck over stretches like this so the assumption is that what is happening is either coaching or failure to execute by the players. Determining which isn’t easy but finding the combined effects in key areas is. There’s some evidence that the precursors to this catastrophe started in early November but for the most part the full manifestation began later that month, 26 games ago. I’m going to take 4 situations, 5v5 offense and defense, power play and PK, and break down the differences between the first 24 games and the last 26.

5v5 Offense

The Avs were the 5th best team by points after 24 games with 14 wins and a handful of OTLs. Let’s look a few key offensive metrics and the changes between then and the past 26 games:

Goals For/60: 2.89 –> 2.11  -0.78
Shot Attempts/60: 54.48 –> 59.48  +5.00
Shots on Goal/60: 30.23 –> 32.83  +2.60
Scoring Chances/60: 26.44 –> 28.70  +2.26
High-Danger Shots/60: 10.90 –> 11.02 +0.12
Shooting percentage: 9.57 –> 6.43  -3.06

Seems pretty straightforward, shot quantity went up by about 9% across the board but goals went down by 27%. What? A telltale here is the high-danger shots, which barely went up at all. I don’t like location as a proxy for danger so let’s think of those as shots close to the net and leave it at that. One conclusion is that they added quantity at the expense of quality however, and that explains some of the shooting percentage freefall.

Now let’s look at who is taking the shots. I’m going to use individual shot attempts divided by on-ice shot attempts to get a percentage for 12 different guys then add change in shooting percentage. For reference here, Mack went from taking 27.1% of the shots when he was on the ice in the first 24 games to taking 20.1% of the shots in the last 26 games.

MacKinnon: 27.7 –> 20.1  Sh% -10.99
Rantanen: 14.1 –> 20.5  Sh% +1.05
Landeskog: 19.6 –> 23.5  Sh% +0.02
Soderberg: 27.5 –> 27.6  Sh% -3.17
Kerfoot: 17.9 –> 15.9  Sh% +6.88
Jost: 20.6 –> 23.3  Sh% -2.68
Compher: 18.0 –> 21.3  Sh% -21.86
Calvert: 18.8 –> 20.4  Sh% +0.48
Barrie: 18.9 –> 23.0  Sh% -5.33
Johnson: 19.9 –> 21.9  Sh% +2.83
Girard: 18.3 –> 16.2  Sh% -0.57
Cole: 14.5 –> 14.7  Sh% -4.55

The first thing that jumps out is that there’s a more even distribution of shots taken on the top line and it’s having a profoundly bad effect on Mack’s shooting percentage. He’s gone from 1.79 goals/60 to 0.57. Gabe is taking more shots but his goals/60 has fallen from 1.09 to 0.88. Mikko’s doing fine though, increasing from 0.47 goals/60 to 0.75. Is this a good thing? It sure doesn’t look like it. Tyson Barrie is taking a ton more shots and it’s getting the team nowhere. The defense in general is shooting the puck more and not doing anything positive. I know this was something that the staff wanted to see but it’s really been a downer overall. League-wide the trend is for defensemen to be shooting less so going against the grain is not paying off. In the first 24 games the D-men took 370 shot attempts at 5v5, in the last 26 they’ve taken 505. It’s time to re-think this.

The Power Play

Last year the Avs PP was 8th in the league at 21.9%, not amazing but plenty effective since they led the league in opportunities by a country mile. This year is a different story, sort of. Right now they are 5th at 25.1% and again lead the league in opportunities by a lot. In the first 24 games they were the #1 unit in the league at 32.1%, since then they are a pretty pedestrian 19.8% and 13th in the league. They’ve gone from better than than a PPG per game to just 21 in 26. I’ll use the same metrics as with the 5v5 offense to uncover some reasons why.

Goals For/60: 11.57 –> 7.54  -4.03
Shot Attempts/60: 96.12 –> 95.83  -0.29
Shots on Goal/60: 56.96 –> 47.92  -9.04
Scoring Chances/60: 49.39 –> 49.99  +0.60
High-Danger Shots/60: 24.03 –> 20.34  -3.69
Shooting Percentage: 20.31% –> 15.11%  -5.20%

It’s pretty easy to say they were on a shooting percentage bender and it’s regressed to a sustainable level. Their Sh% mirrors their PP% relative to the league even. Shot attempts and scoring chances have stayed steady through both periods but the shots on goal and shots from close to the net have dropped dramatically. Shooting percentage is a result, we need to find the process here. I’ll go back to who’s taking the shots and individual Sh% for some insight.

Mackinnon: 32.9 –> 33.0  Sh% -5.22
Rantanen: 18.2 –> 23.5  Sh% +6.43
Landeskog: 17.2 –> 17.6  Sh% -2.28
Compher: 20.0 –> 10.4  Sh% -25.00
Barrie: 22.4 –> 17.0  Sh% +14.29

Kerfoot: 15.3 –> 12.2  Sh% -33.0
Soderberg: 13.0 –> 14.3 Sh% -25.0
Wilson: 19.3 –> 13.8 Sh% -3.08
Jost: 17.4 –> 23.9 Sh% -20.0
Girard: 13.0 –> 19.5  Sh% +0.0

Some interesting things to add here, the 2nd unit players have been getting about the some PPTOI all year but Mack/Mikko/Landy have gotten around 50 seconds more per game in the last 26. Barrie’s minutes have skyrocketed from 3:34/gm to over 5. Compher has gone from 1:18/gm to 4:03.

For the first couple months of the season PP2 more or less carried the whole power play at times. They generated lots of shots and even better got a lot of shots from the slot area. There isn’t a great way to extract pure PP1/PP2 numbers, the best way I can think of is to use MacKinnon and Kerfoot’s on-ice rates as proxies.

CF/60: 95.77 –> 102.18  +6.41
GF/60: 10.28 –> 9.92  -0.36
Sh%: 18.32 –> 19.23  +0.91

CF/60: 107.02 –> 73.90  -33.12
GF/60: 14.86 –> 3.02  -11.84
Sh%: 22.73 –> 8.70  -14.03

The 1st unit is slightly less productive lately despite attempting more shots. The big takeaway here is that the 2nd unit has gone from highly effective to non-existent. The style that they used, more east-west movement and shots from the slot, has been neutered and conformed to how the 1st unit plays. This is backwards. The 1st unit should have taken on the characteristics of the more successful 2nd. Instead we have more shots from the defenseman and not enough raw talent on the perimeter to make use of how the top unit plays. It’s mind-boggling.

Penalty Kill

Last season the Avs had the 4th best PK in the NHL at 83.3%. Since they were one of the more penalized teams they gave up more goals per game than we might like but it was still a positive. In the first 24 games of this year they were competitive at 81.1% which was 11th in the league at the time. Since then they’ve been dead last at 72%, not good. I’ll use some similar metrics to above to shed some light on why.

Goals Against/60: 6.59 –> 9.96  +3.36
Unblocked Shot Attempts Against/60: 81.35 –> 58.94  -22.41
Scoring Chances Against/60: 52.68 –> 39.43  -13.25
High-Danger Attempts Against/60: 22.47 –> 14.11  -8.36
Save Percentage: 89.2% –> 76.7%  -12.5%

This is truly baffling. The staff made some adjustments that have worked fantastically at suppressing shots, 25-35% less than before, yet goal rate went up 50% and the goalies are letting in one more  goal every eight shots on net. This area more than any other has been the trigger for the goalie blame so lets take a look at how they’ve changed. For reference, Varly played 2/3rds of the games in the first part of the year and the two have split time in the 2nd.

Shots on Goal Against/60: 60.55 –> 43.01  -17.54
High Danger Shots Against/60: 17.54 –> 9.10  -8.44
Rebound Attempts Against/60: 7.92 –> 5.79  -2.13
Save Percentage: 91.6% –> 76.9%  -14.7%
Goals Saved Above Average/60: +1.42 –> -0.23  -1.65

Shots on Goal Against/60: 62.64 –> 41.82  -20.82
High Danger Shots Against/60: 15.97 –> 11.82  -4.15
Rebound Attempts Against/60: 9.83 –> 7.27  -2.56
Save Percentage: 84.3% –> 76.1%  -8.2%
Goals Saved Above Average/60: -3.09 –> -0.57  +2.52

The smoking gun here is in the goals saved above average numbers. GSAA is a metric that is used as a proxy for quality chances against, taking a theoretical amount of goals that an average goalie would have saved given what the opponent generates. It mainly uses shot distance so it’s an imperfect measure but it is fairly useful for comparisons. Grubauer’s GSAA was pretty horrifying on the PK in the first part of the season so what the Avs staff has done is try to tailor the defensive structure to help him out. The collateral damage is that it’s made Varly much worse. Add to that the fact that Grubi still isn’t very good. This is troubling for many reasons.

That’s the “what” and “why”, the “how” is more difficult to parse out and can only be done anecdotally. The system has definitely changed, what the PKers are being coached to do is up for debate. By my eye test they are more passive and collapsed and focused on preventing shots more than passes to dangerous areas. I’m not a fan. What made last year’s unit great was aggressive pressure, preventing royal road passes and forcing turnovers, now that’s not as evident.

5v5 Defense

Along with the PK, defense at 5v5 has taken a downturn as well. Goals are way up and once again the goalies get the lion’s share of the blame.

Goals Against/60: 2.20 –> 2.89  +0.69
Shot Attempts Against/60: 56.97 –> 56.97  no change
Scoring Chances Against/60: 24.56 –> 25.74  +1.18
High-Danger Attempts Against/60: 10.48 –> 11.32  +0.84
Save Percentage: 92.7% –> 90.6%  -2.1%

There’s been little or no change in shots allowed but goals against has risen 30% and save percentage fallen from outstanding to slightly below league average. Let’s look at the goalies again.

Shots on Goal Against/60: 29.73 –> 30.36  +0.63
High Danger Shots Against/60: 7.41 –> 8.16  +0.75
Rebound Attempts Against/60: 4.24 –> 4.62  +0.38
Save Percentage: 92.9% –> 90.7%  -2.2%
Goals Saved Above Average/60: +0.06 –> -0.04  -0.10

Shots on Goal Against/60: 30.66 –> 31.08  +0.42
High Danger Shots Against/60: 8.38 –> 9.02  +0.64
Rebound Attempts Against/60: 2.37 –> 3.90  +1.53
Save Percentage: 92.3% –> 90.2%  -2.1%
Goals Saved Above Average/60: -0.12 –> -0.13  -0.01

The key point here is that while save percentage has tumbled, GSAA has dropped only slightly for Varly and is basically the same for Grubi. The conclusion is that both guys are still doing very average work compared with what they face in front of them, not exactly what you’d like but not terrible, the quality of chances against has gone way up. One huge red flag is that Grubi is seeing 65% more rebound chances now than in the first 24 games. Part of that’s on him of course but that’s something that the staff should be able to correct with defensive tactics.

Speaking of defense, we should take a quick look to see if there are any outliers. Shots rates are pretty similar through the season for most guys on the team so I’ll use goals per hour and on-ice save percentage.

Johnson – GA/60: 2.91 –> 2.84  -0.07, Sv%: 90.7 –> 90.4  -0.3
Girard – GA/60: 2.78 –> 2.53  -0.25, Sv%: 91.4 –> 91.8 +0.4
Cole – GA/60: 2.06 –> 3.53  +1.47, Sv%: 92.5 –> 87.9  -4.6
Barrie – GA/60: 1.43 –> 3.04  +1.61, Sv%: 94.8 –> 89.9  -4.9
Zadorov – GA/60: 1.37 –> 2.67  +1.30, Sv%: 95.6 –> 92.2  -3.4
Nemeth – GA/60: 1.85 –> 2.75  +0.90, Sv%: 93.6 –> 92.1  -1.5

Welp, looks like we have a problem here. Ian Cole has gone from solid defensive zone performer to nightmare and the staff are giving him 1:10 more ice time during the slide than before. His partner most of the time, Tyson Barrie, has similar issues and he’s getting 43 seconds more time per game at 5v5 these days. Nikita Zadorov also seems to be bitten by the GA bug but he missed 10 of the last 26 games so I’m on the fence about him for now due to sample size issues. Sam Girard has actually improved but his usage has changed a bit lately so it could be a matchup deal or the fact that he’s amazing at zone exits, not sure about that yet. The thing that is clear is that the staff might have to cut back on Cole’s minutes pronto.

 +          +          +

The scoring and defense have flip-flopped from good to bad at the same time so there are a lot of discrete issues for the staff to tackle. Coach Bednar likes to say that the offense feeds off the defense so that’s probably the best place to start from his POV but there’s nothing preventing them from tweaking both at the same time.

In the offensive zone, the first order of business is to get Nathan MacKinnon back on track. He needs to be the guy controlling play and shooting first whenever possible. I realize opponents very badly want to stop that but very few should be able to. If it takes splitting up the top line for a while then go ahead and try, there’s nothing to lose anymore. Getting the Avs defensemen involved offensively is a noble pursuit but it’s outdated and not helping. Focusing on quality over quantity worked fantastically last year, don’t fall into the trap of being a safe dump & chase team just because of problems in the defensive zone. The Avs greatest weapon is speed and it needs to be used in every way possible. Playing a slow system guarantees they’ll be average at best.

The total disappearance of the 2nd power play unit has had multiple effects. They aren’t helping the team on the scoreboard anymore and without that threat at the end of PP’s it makes the 1st unit less able to generate easy chances. Ray Bennett has to roll back some tape from the fall and try to re-create the magic, or better yet use the strategy the 2nd unit used successfully with the 1st unit too. Again, the shooting from the point by defenseman deal sounds smart but isn’t helping the team.

Neither goalie has been more than average on the PK but the current setup has done more harm than good. Even if the plan is to cater to Grubauer’s faults and leave Varly to deal with it however possible, there has to be some change. Using a more aggressive approach to suppress puck movement rather than just shots will result in fewer of both which hopefully is the goal. Like with the offensive issues, the answer is to use speed.

Without knowing what the players are being coached to do in the defensive zone it’s impossible to say what exactly needs fixing. We can see the results but have little insight into the process. The results say that opponents are getting much better scoring chances than they were earlier in the year. That could mean any number of problems to solve. If that means changing tactics or strategies so be it. If it means re-allocating personnel use, same thing. Doing the wrong thing for the right reasons won’t keep the club in playoff position. Bottom line is that they can’t continue to allow those kinds of opportunities so frequently and using the strengths that they do have as much as possible should be part of the plan.

The good news is that the goalies probably aren’t that bad, just average. The bad news is that the rest of the roster on average is in the same boat. We begin 32 games in 64 days to find out if that’s going to be good enough.


Scoring LW, punchy climber for the Ardennes classics, spirit guide

%d bloggers like this: