I work with a few numbers to judge goaltenders, usually in the modern era dealing with even strength save percentage, quality starts and blow-ups, looking for the right balance of overall value and consistency.

There is a bit of an issue with doing that over a year-to-year span, however. NHL save percentages have crept up since 1984 when the league started recording shots on goal, with better goaltending and better overall defensive schemes, keeping goals off the board.

So how can we stack Roberto Luongo up against his Canuck kin? Not by overall save percentage, since Kirk McLean’s .901 in 1992 is probably much more impressive than Dan Cloutier’s .901 a few years later. Luongo owns six of the top seven save percentage seasons in franchise history, with Cloutier’s 2004 mark coming the closest to him.

Our friend Bruce Arthur at the National Post has suggested over Twitter several times about the prospect of coming up with an “era adjusted save percentage” statistic mirroring OPS+ in baseball, but I’d like to go a little further. What I really want to know when judging a goalie is “how valuable was he,” as in, “how many goals did he save the club?”

For giggles, I adjusted save percentage for every Canuck goalie since 1984 by year. The adjustment was simple, like the OPS+ formula, for SV%+, it was ( [ save percentage / { league combined save percentage } ] – 10% ) so the average goaltender would have a .900 save percentage.

Here are the 15 seasons where a Canuck goalie who faced 1000 shots was better than league average:

This tells us two things and doesn’t tell us another. It tells us that the Canucks have had very lame goaltending throughout their history and that Roberto Luongo was definitely the best. It doesn’t tell us how many goals a goalie prevented.

For that I turn to something I’m developing I call “SOAG” which is “saves over average goaltender”. It tells us the value of a goalie, by calculating how many saves he made compared to what an average league goalie would make: (  goaltender saves – [ league combined save percentage x goaltender shots faced ] ). It’s simple because it adjusts for both era and for the number of shots faced.

Look at Cloutier from ’04 and McLean from ’95. Since Cloutier had a longer season to work with, he would up making nearly 400 more stops than McLean, saving an extra number of goals. Similarly, Luongo from 2010 was probably more valuable to the Canucks than Cloutier’s ’04 because he had to make more saves.

I generally work with this for even strength save percentage alone, but those only go back to 2002, so we’ll have to do. Here are the ten best seasons by a Canuck goaltender for SOAG:

Not surprisingly, Luongo’s two Vezina-nominated years score very highly. I don’t know who Wendell Young is, but he had a .929 SV%+ in 1989 and having faced 536 shots, that was enough to give him 13.5 goals saved versus the league average that season.

Also, concerns about Cory Schneider tending the Vancouver net? Put them to rest. He has improved every year, with his SV%+ going from .865 to .905 to .917 to .925. He’s also second in team history:

Combine the numbers, and, since 1984 when shots began being counted, the Canucks have allowed 116.5 extra goals compared to league average goaltenders. Yet since the lockout, they’re 129.0 above. Kirk McLean’s low number is compounded by his final three years in Vancouver, which were awful. From 1996-98, he cost the team 64.3 goals.

Also, don’t worry, Wendell Young never really panned out as a journeyman backup. He had a good first season (and was also strong in 1994 with Tampa Bay) but he was pretty league average for most of his career, including his second year with the Canucks. Steve Weeks was also a journeyman who had a real strong 1989 season and nothing else in his career.

So, I hope some of this was illuminating. Draw from this what you will, be I’m pretty fine with knowing that the team is in good hands when Schneider takes over as the starter next season. I just don’t see Luongo staying after what he’s been put through in the last two weeks.