He doesn't mention what he's using for scoring chances, if it's a custom measurement or the default one you sometimes see on intermission reports, but either way it seems to be a very subjective measurement.
As far as the underlying idea, if it could be measured accurately and consistently, it does seem interesting. Adds an element of shot quality to corsi while also (hopefully) filtering out some of the inherent flukyness of +/-
Anyway, the Oilers results are interesting. A net -.3 scoring chances per game this year as opposed to -3.1 last year, although that hasn't reflected in improved results as they're currently playing at a .360 points percentage compared to .468 last year through the same number of games. Granted that's a little marred by their terrible 5-15-2 start (.273 points percentage) and their record since then is much improved although still not good (8-10-2 for a .452 points percentage).
Although I'd say the more concerning piece is that, according to the article, that improvement is coming from the non-core areas of their team with RNH, Hall, Eberle and Gagner actually showing worse numbers.
Will be interesting how it plays out as Eakins continues to get his hands on them.
I'd be interesting in seeing some sort of hybrid where shots are weighted by quality, but I do like this approach in general (agreed about the fuzzy nature of what the reporter calls "scoring chances").