9 Things: On Goalies, On Luck, On Donner & Blitzen

LOUD NOISES! abound in the land.

1. Let's talk about our goaltending feelings.

Semyon Varlamov and J.S. Giguere have just put up out-of-this-world numbers through nine games. How far? Check the save percentages (all numbers from NHL.com/stats and extraskater.com):

Yeah. Good lord. Giguere's overall .981 currently leads the league.

1. Giguere (.981)
2. Tuukka Rask (.954)
3. Jonas Gustavsson (.953)
4. Kari Lehtonen (.953)
5. Varlamov (.950)

With any unsustainably hot streak (I mean just look at how early in the year it obviously is if Jonas Freaking Gustavsson is #3 in the league in SV%) comes people saying it won't last. That's obnoxious. It gets worse when people who use stats use certain words that strike a nerve - "luck" comes to mind.

2. "That can't last forever!" Yeah... we know.

Here's what's going on. Stats are used to make predictions. It's what numbers people do. They look at data and make predictions based on it in order to test their models and ideas. No one is saying not to enjoy the ridiculous run while it lasts. No one (that I have seen) is saying the goaltending isn't good simply because their numbers through nine aren't sustainable. When fans, or even media members, take this too personally as an affront to Fun Itself, obnoxious flips to outright noxious.

That played out both on Hockey Twitter and in the Penguins recap thread here yesterday (among other things, of course). It makes no sense to take offense at someone predicting that an obvious hot streak is going to end. Fandom isn't rational and doesn't make a lot of sense, but that may not be more apparent than when someone makes a prediction we don't like about the team we do. The amount of words and bile that came out of this bordered on absurd. It's a prediction, not an assault on your enjoyment of the game.

But the issue here really isn't that people are saying this is a lucky streak. The prediction that Colorado goaltending will come back down to earth is so laughably obvious that it's said derisively. And that is the problem. It's one of tone. Stats people were quick to gleefully point out that the Sharks' early pace was unsustainable. No shit, Sherlock--Tomas Hertl isn't going to score three goals a game. It's important to make accurate predictions as stats attempt to gain mainstream footing, but come on. Nobody thinks Giguere is going to have 2 shutouts for every 3 games. It's not a big deal. Don't make it one and nobody talks about fun 140 characters at a time for three hours, nobody spends their afternoon slapfighting in the comments section, and we all have a much better time watching this crazy game.

3. Pretty Damn Optimistic

The Avalanche have the 4th-highest PDO in the league at 5v5 right now with 104.0 (behind Anaheim, Tampa Bay, and Boston). Their overall 106.7 is league-highest. So an obvious prediction to make is that something is going to give here. To find out what it is, let's see where that *flameshield* luck is coming from.

PDO is a combination of team SV% and S%, and regresses heavily to 100.0 over very large samples. (Sometimes you see it written without decimals, 1000.) 9 games in, that number isn't particularly meaningful. But we can expect it not to end that way at the end of the year. Even in an abbreviated season (thanks, Gary) last year the league range was 96.5 to 103.2. PDO is driven by team shooting percentage (currently 10.7%) and team save percentage (currently .961). That .961 is tops in the league and it's not even close; again, nobody expects that to stay so high for long. But the shooting percentage, while somewhat high, is hardly astronomical. League average last year was 9.11%, which is fairly normal for post-Lockout 2 hockey. (per QuantHockey).

It's probably unwise to make any predictions based on possession numbers right now because there's just not enough 5 on 5 close data. The Avs have spent 218 minutes that way and are allowing 29.9 shots/60 with a Corsi% right smack in the middle of the league. If the goaltending even falls to average, you're looking at 2 or 3 goals against per game. This offense has a perfectly good chance to be in every single game at that rate. From here on out, even about .500 hockey and you're in good shape.

4. How's MacKinnon?

The rookie is holding his own, you could say. He's only leading the team in even strength points/60 (3.99). It's all assists, and it's driven by an outright stupid on-ice shooting percentage of 13.33%. (Nate's first goal was a power play tally, not even strength.) His linemates Jamie McGinn and P.A. Parenteau are both north of 22%. His Fenwick% is worst among Avs forwards with 50+ minutes of ice time, an unfortunate 40.7%, despite only starting in his own zone 20.9% of the time. He's an 18-year old rookie. This is to be expected. There is going to be a learning curve here.

Most players can't sustain a high on-ice shooting percentage. One of those who can is Alex Tanguay. Let's not get ahead of ourselves on this one; this ability can't really be seen in the numbers for several seasons. But in terms of production, MacKinnon has been really good, and a lot of that has to be seen as thanks to how Roy is using him.

5. Blocks Party

There was some talk after the Pittsburgh game that the Avs had weathered the storm of shots by blocking a lot of them. Whether that's a workable strategy over the long term isn't my concern here, but the Avs definitely did block more shots than normal on Monday night. I compared Extra Skater's Fenwick and Corsi Against numbers, which should isolate blocks per shot attempt; the League blocks 17% to 32% of shot attempts with Colorado right in the middle of the spread. Against Pittsburgh, they blocked 30.5%, which would be the third-highest rate in the league if it were their season average.

Did blocked shots help the Avs win? That's a long-term debate for another day, but they definitely were blocking shots at a higher rate than they usually do.

6. Mmmmmmmatt Duchene!!

In case you missed it, Dutchy scored goals 10 and 1 in the NHL Network top ten goals last week. He's been absolutely unreal after a period of some growing pains with his new, "whippier" stick. No, he won't shoot 23.1% forever. But since switching from 120-flex sticks to 85-flex he's been able to snipe under the bar like never before. It's something else.

7. Usage: Paul Stastny

We'd wondered before the year began whether Walnuts would still be deployed as a defensive center. Many speculated his line with Landeskog and Tanguay would be a scoring line, while others argued his was really the only option to shut down opponents.

Excluding John Mitchell, who has taken Dzone draws in Nathan MacKinnon's place in key situations and is therefore kind of skewed, the three Avalanche PLAYERS with the most defensive zone starts--not just forwards, PLAYERS--are Paul Stastny, Gabriel Landeskog, and Alex Tanguay. Yet the three of them are all within 1.5% of even in Fenwick%. That means they're starting in front of their own goalie a lot (nearly half the time) but playing equally in either offensive zone. It's sneaky, it's quiet, and it's a hugely important measure of how this line is helping the team without necessarily putting up sick goals like this one.

8. Oh my God

There are no words for Duchene's play to get O'Reilly the puck against Buffalo.

9. Scoreboard watching

#ConferenceIII is on fire at the top. The Avs' 8-1-0 start has them 5 points away from being out of a playoff spot entirely. Maybe 4 depending on the Pacific: Anaheim hadn't lost since facing the Avalanche until they finally fell to Toronto last night on the heels of a Phil Kessel hat trick, and San Jose's 8-0-1 start has them leading the league with 17 points. Let's not mess with the wild card spot, boys. Of the top 4 competitors for those 2 spots at this early point, 3 are from the Pacific.

The wins must continue to come when more divisional opponents start to come in. The Avs play eight games against divisional opponents between now and December. We'll know a whole lot more about where the 13-14 Avs will finish by then. Like you keep hearing I'm sure, the exceptions are well-publicized, from 2005-2012, only 3 eventual playoff teams were 4 or more points out of a spot at the beginning of November.