Betting Hot Goalies

This post is a chapter that was cut from my new book The Hockey Economist’s Betting Prospectus because the first draft was too long, but there were still some interesting conclusions worth reading. My book is a comprehensive commentary on the last 3 years of hockey betting, broken down by team, by category, by strategy, by season. There is plenty of useful information for bettors of all skill levels. It covers pre-pandemic, peak-pandemic, post-pandemic. What worked, what failed. Lessons learned, market trends, team-by-team analysis. What impact did the pandemic have on hockey betting? The market differences between these 3 seasons are discussed at length, and there's a lot to talk about. To read more, visit the Amazon store.

No player has a greater impact on the outcome of an NHL hockey game than goaltenders, with rare exceptions like a Connor McDavid. The margin of error is so tiny that having part of your body an inch out of position can lead to a goal instead of a save. When a goalie makes a mistake, it’s much more likely to show up on the scoreboard. If you simply tried betting on the team with the starting goaltender sporting a higher save percentage, you could have turned a profit in 2019/20, but you would have gotten hammered in the 2 following seasons (losing -$5,000 from October 2019 to May 2022).

Sometimes simple models can work the best, but not in this case. I decided to try several different models to see what produced positive results using goalie data as the primary input, and there was weak correlation from one season to the next. The pattern that quickly emerged as my book was being written; these 3 seasons being examined here were very different. 2021 is nearly a complete aberration (with a few exceptions), while 2020 and 2022 are often polar opposites. Underdogs dominated one, and favorites in the other. In 2021/22, most permutations of “bet the better team” produced a profit, but the opposite was true in 2019/20.

One logistical problem is that in many cases, the official starting goalie is not known until closer to puck drop. The betting lines that I’m using in my model were often recorded before the official starting goalie was known. Generally, when it’s announced that an inferior back-up goalie will be starting a game, the line moves after that information is released. Betting against back-ups is easier said than done. Even betting on the better goalie may require you to wait until pre-game warm-ups to log your wager. You can go to Daily Faceoff for a projection of who is going to start the game, but even they aren’t 100% reliable.

My two primary goaltending variables are SV% full-season and SV% for the last 5 GP. Though only one of my attempted models that included last 5 GP numbers actually produced a profitable result. One of the pitfalls of using SV% data from recent games and applying it to the 2021 pandemic schedule is that we regularly saw teams play 2-4 consecutive games against the same opponent. Were the Boston goalies only hot because they played 6 games against New Jersey and Buffalo in their previous 8 games?

Do I necessarily want to bet that the goalie will take that same success into a future game against tougher competition? Probably not. If Cam Talbot got blown up for 10 goals in 2 games against Colorado, does that necessarily mean that it’s better to bet against Cam for his next game against Anaheim? It cuts both ways.

 

The title of this chapter was going to be “betting hot goalies”, which beat out the alternate but more appropriate title, “betting really good goalies”. If you bet every goalie with at least a .920 SV% to beat every goalie with a sub .900 SV% (regardless of the team’s quality), you turned a profit in 2020 and 2022, but got killed in 2021. That’s the safer sub-sample of simply betting on the goalie with the higher save percentage, which got crushed in 2022, probably due to the struggles of underdogs in that campaign.

When team winning percentage is introduced to the modelling, then the differences between 2020 and 2022 become even more pronounced. Favorites were scoring more goals, and more goalies league-wide struggled at goal prevention. Hot goalies were stealing fewer games for underdogs, and you can see in one of the charts above that underdogs with goalies greater than .920 against favorites with a starter under .910 was a great bet in 2019/20, but not in the two post-Covid seasons, which looked eerily similar to one another.

Of all these models, the one that produced the most total money across all 3 seasons combined was betting on all goalies with at least a .940 SV% in their last 5 GP versus opposing goalies who were below .910 in their last 5 GP. That’s also the model that I attempted which produced a profit in all 3 seasons, both at home and on the road. However, it produced less profit each year, so the chronological trendline is pointing decidedly down.

Who were the best and worst goalies to bet on during these 3 seasons?

 

Looking at my worst goalies to bet, Curtis McElhinney topped the list for his two seasons as the back-up goalie in Tampa (he was out of the league by 2021/22). All that lost money was not a function of me being a McElhinney believer, but rather it was likely invested under the expectation that Vasilevskiy would be in Tampa’s net. Andrei was by far my best goalie to bet on, almost doubling second place. Few would argue that he was the league’s top goaltender for these 3 seasons, but he only finished 8th best if you bet on every goalie for every start.

Vasilevskiy might have been at the top of my lists in 2020 and 2021, but part of that success had to do with making larger bets on the right games (especially at home). If you were just betting the same amount on every single Vasy start, then you would have performed less well. There were also several games where I had bet Tampa before the starter was known, and surely the lines would be adjusted after that information was public given the disparity between Tampa’s tenders, so waiting for the announcement would lower those payouts even further.

While Vasilevskiy was lower on the best goalies to bet list than you might expect (due to expensive line prices), he did finish #1 on the worst goalies to bet against list. If you put $100 on the moneyline for Tampa’s opponent to win every Vasilevskiy start, you lost -$4,026 in these 3 regular seasons combined. It would have been even worse if you extended that to the playoffs. There is not a worse goalie in the league to stake your money against in an elimination game.

It proves that I should have been avoiding Tampa when the back-up was starting in goal, but I was not investigating the game logs on a daily basis trying to anticipate when McElhinney would get the nod. Ilya Samsonov from Washington was my #2 best goalie to bet in 2021, while his back-up Vitek Vanecek was my #2 worst. What’s strange about that is Vanecek was actually the better goalie, but for some reason was worse when my money was on him to win.

I’m a little disappointed in myself for laying down thousands of dollars on Dave Rittich to win, but most of that came in Calgary in 2019/20 when people were still calling him “big save Dave”. Jordan Binnington also cracked my 5 worst goalies to bet, which helps explain the origin of my betting rule “never trust the St. Louis Blues”. Jordo blowing easy starts against bad teams cost me some very large wagers.

In 2021 I was having a very difficult time predicting the winner of Nashville Predator games until Juuse Saros got hot, after which I was picking the Preds to win/cover far more often, which produced positive results. By the end of the season, he was the #2 most profitable goalie to bet every game, after a slow start to the season, which was followed by an injury. He didn’t catch fire until after returning from the injury. Goalies are prone to both hot and cold streaks, and you need to be proficient at identifying when to jump on and off the bandwagon.

It was amusing to see Mikko Koskinen show up on my 2021 best goalies to bet list, when Mike Smith was actually the #1 best if you bet every goalie every game. Edmonton was playing against weak competition (relatively speaking) in the Canadian division and the team finished #1 in my Power Rankings that season. But rest assured, Koskinen was not a starter that I was specifically targeting, it just happened. I wasn’t writing Tweets encouraging people to bet on Mikko to win (note to self: make sure that’s true).