With that in mind, welcome Linus Ullmark to the chat.

One of the many faces of the Boston Bruins’ incredible resurgence was relatively unknown just six months ago. It was the summer of 2021 when things first shifted for the former sixth-round draft pick. As he blew out the candles on his 28th birthday, Ullmark had never topped 34 NHL starts in a season. His career to date had been spent with the lowly Buffalo Sabres, the root canal of goalie gigs at the time.

That July, Ullmark signed with Boston, an aging but competitive team hoping to keep its contention window open a little longer. He was joining a potentially crowded crease. Prospect Jeremy Swayman looked ready for prime time, while long-time stud Tuukka Rask was expected back from off-season surgery mid-season.

Few could anticipate that two years later Ullmark would be the betting favorite for the Vezina Trophy. Fewer yet could imagine he would headline a story about all-time great goalie seasons. But here we are.

Adjusted Goals Saved Above Average (GSAA)

On our quest to rank Ullmark’s 2022-23 performance in goaltending history, let’s meet Goals Saved Above Average. GSAA is the number of goals saved (or allowed) vs. league-average save percentage over the same number of shots.

Robin Lehner’s 2021-22 is the perfect example to illustrate the stat. The NHL’s average save percentage last season was .907. So was Lehner’s. On the 1,288 shots Lehner faced, an average NHL goalie would stop 1,168, for a .907 save percentage. Lehner stopped exactly 1,168. We would say Lehner saved zero goals above average. Had he saved 1,178, he’d earned 10 GSAA. Had he stopped 1,158, he’d have -10 GSAA. Simple enough.

Vezina winner Igor Shesterkin (.935 save percentage) saved an incredible 45 goals above average, the best in the league. Phillip Grubauer, meanwhile, welcomed the Seattle Kraken to the NHL by “saving” 26 goals below average, the league’s worst. While it doesn’t account for shot quality, GSAA allows us to go back to 1955-56 to compare goaltenders—the first season shots were tracked.

So, how does Ullmark rate historically by GSAA?

As Bob Cole would say, “Ohhhh, baby!”

Using my version of Adjusted GSAA—which neutralizes scoring environment and schedule length—Ullmark is in rarified air. By virtue of sharing the cage with Swayman, Ullmark’s total GSAA count is limited. If “limited” is the 18th-most GSAA since the 1950s, I suspect he’ll be satisfied.

Ullmark’s puck-stopping efficiency, however, is the second-best in NHL history. Sports science and roster construction has evolved over time. Goalies don’t flirt with 70 games played anymore like Bernie Parent or Tony Esposito. When we slot Ullmark’s work on a per-game basis, his incredible year sparkles. By saving 44 adjusted goals above average in just 47 games played, Ullmark is stopping about one more goal per game than an average NHL goalie, says Carolina’s Frederik Andersen.

Incredibly, Ullmark’s .937 save percentage is .034 better than the NHL’s collective goaltending union. This is the biggest gap since Dominik Hasek in 1993-94 (+.035), among goalies with at least 41 outings.

Adjusted Goaltending Records

Wins are a team stat. Applying them to goaltenders is greatly overstating the goalie’s importance in these wins. Mediocre goaltenders win many games on great teams, while great goaltenders lose many games on mediocre teams.

The amount of winning that Ullmark has done this season, however, is jaw-dropping. In 45 decisions spanning six months, he’s lost seven times (38-6-1). Seven! By comparison, poor John Gibson has lost seven straight games in a three-week stretch for Anaheim.

Adjusted goalie records are a measure I’ve developed to level the inequity created by three-point games.  The “loser” point was introduced in 1999 for any loss after regulation. As a result, the concept of .500 for teams and goaltenders is skewed worse than the late Al Arbour’s vision without his trademark glasses.

Adjusted goaltending records correct three important things:

• Overtime losses are losses, not ties. Yes, the team got a point. But as a goaltender, you lost the game.
• Shootout decisions are scrapped. Yes, being a good shootout goaltender helps the team, and being a bad one hurts. But goaltenders prior to 2005-06 did not have access to shootouts, and the NHL doesn’t even count these results in individual totals. So, if a game is tied after overtime, both goalies get a tie.
• Decisions are adjusted for an 82-game schedule.

The result? Every goaltender in history has access to two points per game. Most importantly, we can evaluate goalie records more fairly than the current method where 25 of 32 teams can claim to be “.500” at the time of this writing.

Now, back to Ullmark…

Even after his three shootout wins are downgraded to ties, Ullmark’s 2022-23 is the fifth-highest adjusted winning percentage. With one more point earned, it would be second-best. To take nothing away from Cecil “Tiny” Thompson, his feat was 93 years ago. It was so long ago that his nickname was ironic because he was a huge dude (5’10”, 160 pounds) in Depression Era times. Thompson would be the smallest goaltender in the NHL today.

Vezina Trophy

Now, Ullmark’s workload has been managed gorgeously and he tends goal for one of the most dominant squads in hockey’s long history. It can be argued that he doesn’t face the same difficulty of shots as some of his peers. Or that if he made 60+ starts, we might see some dents in his shiny stat line. Both are valid thoughts.

In fact, there are two other goaltenders having world-class seasons that should join Ullmark as Vezina Trophy finalists—Ilya Sorokin (New York Islanders) and Juuse Saros (Nashville).

One of the best data points available post-2007 is Goals Saved Above Expected (GSAx). The concept is similar to GSAA, only shot quality is taken into consideration. Using shot quality (i.e., location, type, pre-shot movement), analysts can determine the number of goals an average netminder should expect to allow based on their specific in-game situation. GSAx is the difference between actual and expected goals allowed.

Here, using Evolving Hockey’s model, we see that Sorokin and Saros are not only ahead of Ullmark, but that the trio’s performances are among the best of the modern analytics movement.

Quick takeaways?

• Any of the three are viable Vezina choices by this measure.
• Ullmark’s lighter workload and degree of shot quality limit his impact.
• With three seasons in the top six, Henrik Lundqvist was more than just a pretty face.

When it comes to voting for the Vezina, however, Ullmark is a near-lock. The award is voted on by the NHL general managers—not the PHWA. Naturally, the GMs overvalue wins in their voting habits. It’s almost as if wins keep them employed long-term.

You need to go back 14 years (Tim Thomas, 2008-09) for the last time a goaltender won the Vezina when finishing more than five wins back of the NHL lead. Saros (31) and Sorokin (29) lag well behind Ullmark’s NHL-best 38 wins. It would take a monumental shift in voting trends for anyone but Ullmark to walk away with the hardware.

Closing Thoughts

By any statistical measure, Ullmark’s performance this year has been special.

Relative to the NHL average, he is saving pucks at the second-best rate since shots on goal were tracked. His adjusted winning percentage is a few decimal points from the best record in 93 years. Despite the lesser workload, when factoring in shot quality via GSAx, his performance holds up extremely well.

Ullmark is a wonderful story. He’s diligently risen from obscurity to the top of his profession. It’s a historic season of goaltending on a historic team, both worthy of celebration.

Paul Pidutti is the creator of the Adjusted Hockey project, which brings context to hockey statistics. His Hockey Hall of Fame methodology, the Pidutti Point Share (PPS) system, is the first comprehensive measure of HHOF worthiness. You can find his work on Twitter (@AdjustedHockey) and adjustedhockey.com.

Adjusted GP & Pace stats from Adjusted Hockey; all other data from Hockey Reference.