(Mark Buckner, Getty Images)

For my post this morning, I was planning on writing a post about how the St. Louis Blues, contrary to the opinion of The Hockey News, wasn’t going to finish first place in the Western Conference. Then I checked Backhand Shelf yesterday and saw that Ryan Lambert had already done so in his usual acerbic style. He ran down a number of reasons, but for me the most important is that there’s no way the goaltending tandem of Jaroslav Halak and Brian Elliott will repeat their performances from last season.

Halak, at least, has several seasons of good to great goaltending under his belt. Even if he’s not as phenomenal as his .926 save percentage and 1.97 goals against average, it’s safe to assume that he’ll still be good enough to give the Blues a chance to win when he starts.

But Elliott? What can we expect from him next season? Like Halak, he has several consistent seasons under his belt. Unfortunately, those seasons were consistently bad. In his last three seasons, Elliott posted save percentages of .902, .909, and .893. In 2011-12, he posted the best single-season save percentage of all time. To go from .893 in one season to .940 in the next just doesn’t happen in the modern NHL. It’s unheard of. It seems impossible.

That’s why it’s incredibly difficult to predict what the 2012-13 season holds for Brian Elliott.

Let’s underline exactly how unprecedented Elliott’s ascension was last season. To understand that, we need to look back at his 2010-11 season, where he wasn’t just bad; he was arguably the worst goaltender in the NHL. In fact, judging by Tom Awad’s Goals Versus Threshold statistic (GVT), Elliott wasn’t just the worst goaltender in the league, he was the worst single player, period.

Essentially, Elliott was so bad in his 43 games for the Senators and 12 games for the Avalanche that those two teams would have been better served by acquiring an above-average AHL goalie. Having Elliott in the lineup for those 12 games seemed to terrify the Avalanche, as they traded the Capitals a first and a second round pick for the rights to RFA Semyon Varlamov to be their starter.

So there’s good reason why the hockey world was incredulous when Elliott came out on fire to start 2011-12, first out-battling rookie Ben Bishop for the backup job in St. Louis, then forcing himself into a 1a/1b situation alongside Halak with his stellar play. I expected him to regress over the remainder of the season, but he kept going strong, until finally flaming out in round 2 of the playoffs against the eventual Cup-winners. Elliott played in the All-Star Game, won the William Jennings Trophy with Halak, and even received a first place vote for the Vezina Trophy. Just one, but still, one guy thought he was better than Lundqvist, Quick, and Rinne.

By the Goals Versus Threshold measurement, Elliott went from the worst player in the NHL to the 5th best, just behind Evgeni Malkin. It was an astounding resurgence, except it wasn’t a resurgence because it wasn’t a return to a previous level of performance. It was, instead, a surgence. The only problem is that isn’t a real word.

At the beginning of February, Philip Myrland of Hockey Prospectus compared Elliott’s turnaround to other goaltenders of the last 18 years:

Since 1994-95, there have been 67 seasons where a goalie posted a save percentage below .900 with at least 40 games played. The majority (50 out of 67) saw their save rate improve the next season, but the average rate of improvement was just .009.

Three goaltenders saw their save percentage particularly improve: Manny Fernandez went from .892 in 2001-02 to .924 the next season, Evgeni Nabokov improved from .885 in 2005-06 to .914, and Ed Belfour went from .895 in 2002-03 to .922. The biggest improvement was Fernandez, whose save percentage went up by .032. That doesn’t even come close to comparing to Elliott, who save percentage improved by .047.

To make matters worse for comparison, all three of those goaltenders carry a certain amount of pedigree, as Myrland pointed out:

Fernandez was coming off of back-to-back .920 seasons when his play temporarily fell off in 2001-02. Nabokov had played four full seasons as a starter at .915 before struggling to get back into playing shape coming back from the lockout. Ed Belfour had a Hall of Fame track record and was just two years removed from leading the league in save percentage and making the second of back-to-back trips to the Stanley Cup Finals in a starring role. For each of those three, it was a pretty safe bet that they would return to a high level of play.

Can we expect the opposite to be true for Elliott? Will he return to his low level of play? It’s tough to find any comparable for goaltenders with one incredible season after several that are below-average. Dwayne Roloson might be one of the few, going from several seasons hovering around a .900 save percentage to finally breaking out at the age of 33 with a .927 save percentage in 2002-03 with the Minnesota Wild. He improved that mark with a .930 save percentage the following season, then proceeded to hover at or just above league average ever since, at least until his .886 mark at the age of 42 last season with the Lightning. For Roloson, that breakout season seemed to signify that he was an improved goaltender and could be relied upon as a starter.

Elliott just turned 27 and should be entering the prime of his career. He and the Blues, who signed him to a 2-year extension, are hoping that his breakout season means the same for him that it did for Roloson. There are plenty of non-statistical arguments for this to be the case. In St. Louis, he’s playing behind a much stronger defence than he did in Ottawa or Colorado. He’s been credited with making numerous adjustments to his playing style that have made him more reliable. He’s toughened mentally to avoid losing his confidence. He isn’t relied upon to be the sole starter in St. Louis, making it easier to manage his streakiness.

Even taking those into account, Elliott has to regress next season, but what will he be regressing to? What is his mean? After a season like he just had, what is average for Brian Elliott?

Oddly enough, his new career averages are a near-match for what had previously been his career highs. In 55 games for the Ottawa Senators in 2009-10, Elliott posted a save percentage of .909 and a goals against average of 2.57. His average save percentage is a match at .909 and his average goals against average is just slightly higher at 2.60. It’s entirely possible that this is an accurate reflection of Elliott’s talent-level, with his last two seasons simply being extreme outliers, the thinnest portions of his talent bell-curve.

The alternative view is that Elliott has turned a corner. While it is extremely improbable that he will ever have a season like 2011-12 again, it is entirely possible that with the improvements he has made with his goaltending coaches and the benefit of playing for a strong defensive team like St. Louis, he will continue to compete with Halak for the number one job.

I’m still not picking him for my fantasy team.