Rolling dice on Vegas repeat might be risky
Numbers suggest Golden Knights could have trouble duplicating expansion success
At a bar in San Francisco on a weekend afternoon in June, a couple of sportswriters asked for a television to be switched to the Stanley Cup Final.
The barkeep obliged, looked at the screen and then asked who was playing. Washington and Vegas, he was told.
“Las Vegas has a hockey team?” he said.
Told it was the first season for the team, he acknowledged this, then started to move down the bar. Then he stopped and looked back.
“They made the final in the first year?”
By mid-June, the Vegas story was a familiar one to anyone who paid attention to hockey, but there in Northern California it was funny to get a reminder of how bizarre it was. That incredulity from the bartender was the correct response. Expansion teams do not make the final. It can’t happen. I mean, what the hell?
Four months later, and with the Golden Knights getting ready for Season 2, no one is still quite sure how it all came about.
Was it luck or the product of favourable expansion-draft rules? Was it coaching or brilliant roster construction?
And if it was some of all that, what lessons can be gleaned by bottom-tier teams that want to leap into contention?
Regarding the Knights, we know at least a few things to be true. They weren’t overly lucky, at least not by the way such things are normally considered. They didn’t win a pile of close games — Vegas had a 10-9 record in one-goal games — and they had seven overtime losses, below the league average of 10, so their absurd 109-point season wasn’t padded by loser points.
Their special teams were good, their save percentage was just barely above league average and their shooting percentage was slightly above average.
But there were some weird things going on. The Knights as a team scored on 10 per cent of their shots, but leading scorer William Karlsson potted a ridiculous 23 per cent of his attempts. He shot at a six-per-cent rate a season earlier in Columbus. How anomalous was his conversion rate? Among players who scored at least 20 goals last year, the 4.2-point gap between Karlsson and Anders Lee, in second place with a 19.2 shooting percentage, was the same as the distance between Lee and Connor McDavid, who was 31st in the NHL with a 15.0 shooting percentage.
That’s at least in part how the Knights found themselves with a dominant top line out of seemingly nowhere and it’s likely not sustainable, which is why the expectations for the Golden Knights have been at least somewhat tempered.
Two other things stand out from a roster perspective: team speed and solid goaltending.
But whether that tells much to other GMs is less certain.
You’d like to think most NHL executives are aware that fast is better than slow and that good goalies are more desirable than bad (the evidence suggests, admittedly, that some GMs are still learning these lessons).
The broader way in which the first year of the Knights was instructive, though, was in underscoring just how unpredictable NHL results can be.
Of the four major leagues in North America, it has the fewest scoring opportunities, which creates the most statistical noise. One only has to look at the up-and-down results of various teams, even those with little roster turnover, to see how that variance plays out.
Three teams — Chicago, Montreal and Ottawa — had point totals that dropped by more than 30 between the 2016-17 and the 2017-18 seasons, while Edmonton and the New York Rangers both had 25-point plummets.
The Canadiens are shining examples of hockey’s uncertainty, with the following divisional results over the previous four seasons: Sixth, first, sixth, first.
And while the Knights were a wild story, 8-to-1 pre-season long shots just to make the playoffs last year, the Colorado Avalanche, coming off a 48-point season, were 7.5-to-1 just to make it beyond mid-April.
The New Jersey Devils were about 6-to-1, the fourth-least likely to make the playoffs. All three teams made the playoffs.
It’s not just that an expansion team can make a startling run in today’s NHL, then, it’s that performance in one season is not necessarily a good predictor of results in the next. If last year’s pattern holds, three of the four expected bottom feeders, by post-season odds, would make the playoffs this year.
Those teams: Ottawa, Arizona, Vancouver and Detroit.