Brian Elliott’s career-high in save percentage is .909 from his sophomore season with the Ottawa Senators in 2009-10. That stellar number had him tied for 25th in the NHL with Dan Ellis and Antero Niittymaki.

Even at his best, Elliott is a very average goaltender and he was definitely not at his best last season, when he put together a 15-27-9 record with the Senators and Avalanche, with an .893 save percentage that was 43rd in the NHL and a 3.34 goals against average, 44th in the NHL. Sure, the Senators and Avalanche were not very good last season, but Elliott was one of the worst goaltenders in the NHL.

This season, however, it’s quite possible that he’s the best.

Elliott currently leads all eligible goaltenders in save percentage, goals against average, and win percentage. He’s also tied for the league lead in shutouts while having started 6 and 11 games fewer than the goaltenders he’s tied with, Tim Thomas and Jonathan Quick. His save percentage is currently .944, his GAA is 1.52, and his record with the Blues is 13-3-0.

Sometimes goalies can get a boost to their save percentage when they’re playing for a team with a great penalty kill, but that’s not the case for Elliott as the Blues have a below average penalty kill percentage at 81.2%. Elliott is also the league leader in even-strength save percentage, which is generally considered to be a better indication of team-independent performance.

Elliott is also leading the Blues in points. Tru fakt.

While Elliott is having an absurd season so far, one that completely flies in the face of his previous performance, it is even more absurd that Elliott isn’t even technically the Blues’ number one goaltender. The team is currently deploying Elliott and Jaroslav Halak as a tandem, using the two goaltenders as 1a and 1b. Halak has the better pedigree, but currently has a .903 SV%, 2.37 GAA, and a 6-7-4 record. There is no doubt who has been the better goaltender for the Blues this season.

The difference is that Halak signed a 4-year, $15-million contract in 2010, while Elliott signed a one-year, two-way deal for $600,000 this past off-season in order to bring in competition for the backup job with prospect Ben Bishop. Instead, he’s battling for the starting job with the guy with the multi-year contract.

To understand how incredible Elliott’s turn around truly is, I think we need to get a better picture of how awful he was last season. For that, I’m going to turn to two interesting statistics that might give us a clearer picture: Quality Starts and Goals Versus Threshhold.

The idea of Quality Starts is simple: in a given game, did the goaltender perform well enough for his team to win the game. Using wins, a team statistic, as an individual statistic for goaltenders is misleading and using only long-term metrics like save percentage and goals against average doesn’t account for the game-by-game nature of goaltending. Quality Starts as a statistic was introduced by Robert Vollman at Hockey Prospectus as a way of measuring how well a goaltender performed on a game-by-game basis apart from whether his team scored enough goals to actually earn the win.

A Quality Start is defined as any start in which a goaltender saves an above average number of shots (a save percentage above .912) or, with a save percentage between .885 and .912, a start in which he allows 2 or fewer goals. Back in 2009, it was found that the actual win percentage for Quality Starts was 77.5%, while non-Quality Starts resulted in a win percentage of 32.5%.

Last season, Brian Elliott was second-last in the NHL in Quality Starts, ahead of only Nikolai Khabibulin, who has seen a similar resurgence this year. Only 17 of Elliott’s 51 starts were Quality Starts, a percentage of 33.3%. had Last season, Vollman also introduced the idea of the Really Bad Start, where a goaltender fails to stop even 85% of the shots he faces during a game, not even giving his team a chance. Elliott had the dubious honour of leading the league in this statistic, with 15 Really Bad Starts in 51 starts.

This season, however, 13 of Elliott’s 16 starts have been Quality Starts and the Blues have won all 13 of those games. Elliott has given up more than 2 goals against only twice this season and 81.2% of Elliott’s starts have been Quality Starts, which is a higher percentage than any goaltender in the NHL last season, admittedly in a much smaller sample size. In comparison, 9 of Halak’s 17 starts this season have been Quality Starts, a percentage of 52.9%, which would have been 6th worst last season.

This is a thing that happened.

When Elliott starts, he consistently gives the Blues a chance to win.

The other statistic is Goals Versus Threshhold or GVT, introduced by Tom Awad, also of Hockey Prospectus. Awad explains that GVT is “the value of a player, in goals, above what a replacement player would have contributed.” Essentially, it is a measurement of how much a player helps a team win compared to the next best guy.

Elliott did not fare well by the GVT measurement last season. In fact, he wasn’t just bad; he was the worst player in the NHL by the measurement. I don’t think it’s possible for Elliott to have had a worse season. I am also have trouble understanding how it’s possible for Elliott to follow up such an awful season with such an incredible start to this season.

Is it possible for Elliott to continue this performance? Probably not. It just doesn’t make any sense that Elliott could improve that much or experience such an absurd swing in performance from last season to this season. As a comparison, Halak played 18 games during his 2010 playoff hot streak with the Montreal Canadiens, where he posted a .923 save percentage and a 2.55 GAA, with 13 Quality Starts in his 18 games. It’s possible that Elliott is currently just experiencing a similar hot streak to that of Halak, albeit with better statistics and fewer consecutive starts. That said, Halak’s numbers on his hot streak were comparable with his numbers from the regular season that year.

Is it possible he could still win the Vezina? Not if he continues to split starts with Halak. However, if the two goaltenders continue with their current levels of performance, it’s entirely likely that Ken Hitchcock will have no choice but to run with Elliott as his starter.

Of course, now that I have written this, Elliott will crash and burn and end up in the ECHL to end the season.