Stanley Cup Playoff Primer (Part 1)
Tying together the NHL season through the lens of the co-favorites for the Hart Trophy
The Stanley Cup Playoffs are a little more than two weeks away. While being locked-in on football and March Madness, we’ve kept an eye on the NHL all season - putting out moneyline projections for all 1200+ games, monitoring how the betting market has felt about every team, the importance of every skater, and which goaltenders are making the biggest difference when they’re between the pipes. However, as we get ready for the postseason, THIS is the first of two critical pieces to help you understand where we’re coming from when it’s time to breakdown all 15 series in April, May, and June.
Don’t look now, but the Hart Trophy race is lit right now.
Ok, it’s possible that’s an oversell for the NHL’s most valuable player award, but you don’t often get this far into a season with these odds in this market:
Handicapping which player will take home the Hart Trophy for NHL MVP may come down to whether voters are willing to give Connor Hellebuyck both the Vezina Trophy for best goaltender and the Hart. In the past, it’s taken a season above-and-beyond comprehension to take home both. Only two goaltenders (Jose Theodore and Carey Price) have done the double, since this absolute menace went back-to-back with it in 1997 and 1998:
Dominik Hasek’s peak came smack-dab in the middle of six Vezina seasons. We didn’t have Goals Saved Above Expectation (GSAx) back then, but hockey-reference.com has Hasek’s Goals Saved Above Average at 54.5, with the next-best being Tom Barrasso at 23.9. Luckily, all anyone needed was the eye-test to know this guy was a problem, since he did stuff like this, nightly:
This season, Hellebuyck’s GSAA is 37.3, an arm’s length away from Andrei Vasilevskiy’s 30.17. Of course, now we have GSAx, which is an improvement on the improvement - since going back to evaluating goalies by GSAA is better than goals-against average and save percentage, which we were swearing by in the 90s.
Through the end of March, Hellebuyck’s GSAx is 46.39 (Igor Shesterkin is at 30.3), which means he’s prevented 46 more goals than if his backup, Eric Comrie (0.09), was in net.
Meanwhile, Leon Draisaitl has 51 goals (and 104 points) in 69 games, but figuring out what a replacement-level player would do in his exact position is a little trickier, but some advanced metrics sites have Draisaitl scoring around twice as many goals than expected, based on where his chances come from. So, we’ll use that estimate, that his talent has added about 25 goals to the Oilers’ overall tally.
Voters won’t compare 46 goals saved to 25 goals added and automatically assume Hellebuyck is the MVP, but looking at it that way at least sets up a quick discussion important to bettors, about what scoring a goal does to increase a team’s win probability, and what preventing a goal does to do the same.
What’s a goal worth to winning?
Simply put, there’s two ways to help your team to win a hockey game:
Score a goal
Prevent a goal
Scoring a goal is something of a binary result - you either score it or you don’t (though chances are boiled down to fractions of an expected goal).
If there was no goaltender, every shot on goal allowed would go in the net, but simply having a goaltender is going to prevent the vast majority of them. You can shut a team out without actually making a save that would have otherwise been a sure-fire goal. As a result, being better than the average goaltender over 25-35 shots accumulates a full goal saved above expectation.
Through March 31st, NHL games average 6.05 goals per game.
When sitting down to watch a game, knowing that if nothing notable happens, you’re likely headed for a 3-3 game.
How that changes is via outlier results like good or bad goaltending, or unusually high or low scoring chance creation/conversion.
As a bettor on a hockey game, you’re hoping for at one of two things to turn the tables in your favor:
Your team gets to four goals first
Your team only allows two goals or fewer
You can win a lot of hockey games by accomplishing either of those objectives.
It’s going to be really hard to lose if you do both.
The value of a goal scored
Every goal scored puts your team a quarter (25%) of the way to four goals.
Every time you watch a playoff game, and your team scores first, think “quarter of the way there!”. The reason 2-0 is the “worst lead in hockey” is because you’re only half-way to being the likely winner.
While 2-1 and 1-0 games - where a team loses despite allowing fewer than three goals - are occasional, only two times in the entire month of March did a team take a lead of 4-2 (or better) and lose, and just four more times did a team score to take a 4-3 lead and not win.
When we broke down the 4% Club in February, some might have been surprised that we estimated Connor McDavid’s availability as only worth 8% to the Oilers’ chances of winning, or that the Hart co-favorite, Draisaitl, was only worth 6%.
More recently, those estimations were put to the test, when both were out last week.
On March 20th, Draisaitl had to sit out a game with the Jets due to injury, and the market flipped from the Oilers being -120 on the moneyline (54.5%) at home, to the Jets being a -118 favorite (54.1%). That’s an 8% shift for Draisaitl’s absence.
The next game, McDavid joined Draisaitl on the injured list with a lower body injury. As we “noted,” the Oilers’ win probability shifted from around 69% (-213) against the Kraken with both players, to 55%/45% in favor of the Oilers, as we addressed on Substack’s notes:
The Oilers’ supporting cast stepped up to beat the Kraken 5-4, proving that not all hope was lost without Edmonton’s superstars.
At least for one game, since they lost to the Stars and at Seattle in the rematch, 6-1.
Without McDavid and Draisaitl, the Oilers were set to by lined around -120 (54.5%) to beat the Flames last Saturday, but Draisaitl’s return moved the Oilers to -175 (63.6%). That 9% shift is higher than we’ve been giving the Hart trophy candidate credit for, but it’s easily explained once we circle back to Draisaitl’s goal production above replacement.
Scoring 25 more goals than an average player in 69 games played means Draisaitl has added 0.36 goals per game.
Since we know that 1 goal equals 25% to win probability, 0.36 goals would equal… *drumroll*… 9% win probability!
The value of a goal saved
From the point of view of a goaltender, saving a full goal above expectation in a game makes you roughly 33% more likely to win.
As we look toward the playoffs, without watching a game, if the only information we’re given after the fact is that a goalie gets credit for 1.00 GSAx, that goalie has single-handedly improved his team’s chances by 33%.
Ie. A 50/50 (-110/-110) game, becomes 66.6/33.3 for his team.
If a goaltender stands on his head, with something like a 3.00 GSAx, that team is pretty much assured of victory.
A week or so after looking at the skaters, we dug into the netminders, trying to figure out which goaltenders matter heading into the playoffs, and you may recall that the list of goalies that force you to apply something meaningful to their team’s chances of winning is not long, but Hellebuyck is atop it.
Saving 46.39 goals over 3378 minutes equates to 0.82 GSAx per 60 minutes.
If 1.0 GSAx is worth 33% of win probability, Hellebuyck has been adding a whopping 27% of win probability every time he starts.
Need more proof? Hellebuyck’s record is 43-13 (76.8%), while Comrie’s record is 8-10 (44.4%).
Conclusion
A couple weeks away from the Stanley Cup Playoffs, this sets the foundation for two pieces of the hockey handicapping puzzle:
If Draisaitl, McDavid, or any other team’s star is forced to miss games with injury, we have a comprehensive understanding of what changes we need to make to the win probability of that player’s team
The adjustments we’ll be willing to make for a team like the Jets because of the goaltending they’ve gotten this season
Meanwhile, Hart voters still have a few weeks to rationalize between skater versus goaltender as to who they’re willing to pick for most valuable player. Which, is anyone’s guess.
NHL Betting Guide (Apr. 3rd)
This is the moneyline betting guide for Thursday’s slate in the NHL. Check back on Notes each day, as the projections for games on subsequent days will be posted in a thread with this article as home base.
The two columns on the far right are the target moneylines for a bet on either team should you be interested in backing one side or the other. For a better understanding on how moneyline betting works in the NHL, click here.
Quick reminder:
As we venture forward at THE WINDOW, please acknowledge that you’ve found value in this content by liking (❤️) and/or restacking (🔄).
It will help others on Substack find THE WINDOW. “When I get it right, great! I want you to be able to see my work and understand how I got there, when I don’t.”