Analyzing how the Stars were able to upset the favoured Golden Knights

After dispatching the heavily favoured Colorado Avalanche in a back and forth seven-game series, the explanations as to why it was possible were readily available. “The Dallas Stars only won because Colorado lost Philipp Grubauer and Erik Johnson in Game 1.”

Well, maybe. It’s possible that competent goaltending would have turned the tide for the Avalanche, but after tossing aside the Vegas Golden Knights in just five games in the Western Conference Final, the Dallas Stars are now on to the Stanley Cup Final for the first time since they were attempting to repeat as champions way back in the 1999-2000 season.

Coming into the playoffs, the Avalanche and Golden Knights were easily the favourites to come out of the West, so no one can say that the Stars had an easy road, but what exactly happened in this series?

By the eye test, the Golden Knights were the superior team. By the numbers, that seemed to be true as well. So how did the Stars not only win the series, but win it quickly? Let’s break it down game by game.

By expected goals, there are two narratives you can take from this series. The first is that the Stars played great in Game 1 and Game 3, with Vegas blowing them up in Game 2, then Dallas stole Game 4 before winning a tightly contested Game 5. That’s a description of how the series played out at even strength only, but when you add in special teams, things get a little crazy.

In all situations, Vegas produced far more offence than Dallas in three straight games, but the Stars played much better in Game 5 and took the series with an overtime win.

At even strength, the total expected goals in the series were 8.6 to 9.5 in favour of the Golden Knights. That one-goal difference for the series along with one outlier game in Game 2 doesn’t look to me like the Stars lucked their way into a win here. Hockey is a random sport, and over five games the difference of just under a single goal in expected offence isn’t as dominant as some shot metrics may make the series appear.

The story then should be about how the Golden Knights’ power play wasn’t able to make good on the offence it produced. Did the Stars simply get lucky while shorthanded? Did Anton Khudobin steal the show? Or were the Stars able to implement some effective defensive measures that may not appear in expected goal metrics, but might explain why they were able outperform expectations? Let’s dig into it.

Comparing what Vegas was able to accomplish on an average power play in the post-season to what they were able to do against the Stars, it should first be pointed out that Vegas' power play was among the most lethal of any team’s that played in these bubble playoffs.

The Stars’ penalty killing was able to shave little bits off of a few areas, but there isn’t much significant change from earlier in the playoffs.

The Golden Knights were still putting more than a scoring chance on net per power play opportunity, which is much higher than league average, and connected on almost two slot passes per power play, so the passing lanes were not closed off.

What’s even more interesting is that the Stars weren’t tightly checking the Golden Knights either. The number of shot attempts Vegas was able to direct at the net without pressure being applied to the shooter was ever so slightly higher than in the rest of the playoffs. The same goes for screened shots. The Stars weren’t breaking those screens and letting their goaltender see the puck either.

That means that the Stars skaters were able to create a relatively level playing field at even strength, at least close enough that they could scrape together wins, but on special teams the skaters weren’t the difference.

By essentially every measure Vegas’ power play was vastly superior to Dallas’ penalty killing, and by the numbers Dallas’ power play wasn’t that great either. However, the Stars’ power play managed to score just as often as the Golden Knights’ did.

At even strength, the Golden Knights got goaltending nearly as strong as the Stars did, with Robin Lehner and Marc-Andre Fleury combining for a .943 save percentage, but between five shots on Jake Oettinger and Anton Khudobin's series, the Stars’ goaltenders put up a .955 save percentage. Still, that was recoverable considering Vegas was outshooting Dallas, but the same didn’t hold true on special teams.

While shorthanded, Khudobin continued to stop pucks at a rate of .935, while Vegas saw their goalies stop shots at a rate of just .769. From the inner slot, the Golden Knights weren’t able to beat Khudobin at all on the man advantage, whereas the Golden Knights were beat on one-third of the Stars’ chances from there.

When the Golden Knights put a power play shot on Khudobin, there was no rebound to speak of 39.5 per cent of the time. When the Stars did the same, Lehner and Fleury managed to control the rebound just 20 per cent of the time.

Even when screened, Khudobin was able to control rebounds on shots he faced 37.5 per cent of the time, while the Golden Knights’ goalies gave up a rebound on every screened shot they faced.

No matter what way you slice it, Khudobin was brilliant when the Stars were down a man, and Lehner wasn’t. Are a few minutes of shorthanded time over five games something to worry about for future performance? No, not at all. But in this series, special teams goaltending was the biggest difference maker.

When submitting content, please abide by our  submission guidelines, and avoid posting profanity, personal attacks or harassment. Should you violate our submissions guidelines, we reserve the right to remove your comments and block your account. Sportsnet reserves the right to close a story’s comment section at any time.