On November 12th, in advance of his team's game ahead of their meeting with the Columbus Blue Jackets, statistician Gabriel Desjardins declared that the then 2-11-1 squad was not, in fact, the worst team in the league.
"Their underlying numbers aren't any worse than they were the last two seasons" wrote Desjardins at the time. "The problem, as usual in these situations, is absurdly low shooting percentage and absurdly low save percentage."
That post was prophetic, as the Jackets would win their game against the Winnipeg Jets that night, and showed a modest improvement in team record. Mind you, their numbers were still below average, but what really killed Columbus in the early going was goaltending: Steve Mason could literally not make a save at even strength.
Enter Curtis Sanford. After Mason, who has been on the complete decline since his Calder-winning season that pulled Columbus into the playoffs for the only time in franchise history, stopped just 89.4% of the shots at even strength, Sanford has come in and has put up a much more respectable 92.5%.
As a result, Columbus has stopped bleeding goals against and find themselves in much more desirable situations. It's funny to see how one player can affect a team's fortunes so much. The best possession teams in the league can be without goaltending (Vancouver Canucks in October) or the worst possession teams in the league can have a red-hot goaltender (Minnesota Wild since October) and run records wildly different to what they've achieved.
Percentages nearly got Scott Arniel, head coach of the Blue Jackets, fired this season. Percentages have already claimed the coaching lives of Davis Payne, Bruce Boudreau and now Terry Murray this season. (Randy Carlyle's job was probably saved by percentages last season). But Columbus has regressed to the mean with help of their goaltender, so let's take a look at how the team has performed since Gabe made that declaration:
|Pts/82 Pace||Corsi Tied||Shot%||Save%||PDO|
|Before Nov 12||29.3||46.1%||7.5%||88.8%||96.3%|
|Since Nov 12||75.2||53.1%||6.3%||93.1%||99.4%|
Oddly enough, the team has actually been ~playing~ better in front of Sanford, who has picked up the majority of the starts of the team and became the regular starter on November 17th. Is that because the team is more confident in their goalie and willing to push a couple more shots forward? I don't know if that effect has been studied yet, but the Jackets are certainly an interesting example. (Also, Kris Russell, a 50.8% Corsi tied player, was shipped to St. Louis for Nikita Nikitin, who has been a 58.3% players since the acquisition.)
What's interesting though is that there are no red flags in the Jackets record since November 12th. There's no real visible reason why the team shouldn't be on a 90+ point pace with that kind of record: a very strong possession team with a normalized PDO. Why are they only 4-5-1?
Two things: More goaltending issues, and specialty teams.
It appears that Sanford, for all the work he's done in getting team's save percentage to a normal state, has quality starts in just 6 of his 12 games, so he's been a little inconsistent. For starters, the team can't seem to buy a save on the penalty kill now, which also tends to have a major effect on play. We can't split up how they've been doing since a certain date, but here's how they've fared throughout the year for goal rates and shot rates in each man-advantage situation, along with the league rank:
|5v4 G/60||5v4 Sh/60||4v5 GA/60||4v5 ShA/60||ST PDO|
|3.7 (28th)||54.1 (7th)||11.2 (30th)||52.1 (20th)||85.3% (30th)|
I calculated Special Team's PDO by adding 5v4 shooting percentage by 4v5 save percentage. It gives us a measurement of which teams are overall lucky on specialty teams.
So while Columbus are a much better even strength team than before and have been seeing results there, the special teams remain brutal, at least the penalty-killing does. The Jackets powerplay is probably due for a regression, but it may not help all that much.
The Jackets PK may be what keeps this team from ever joining the playoff hunt again. They can't keep giving goals away a man up.
Evidently, the Jackets' decision to dress Dane Byers to scare the Sedins with his scariness is a move that may not pay off for Columbus, who really have no business putting themselves into a situation where they'd take more penalties. Not against a team that's coming off a night of being bullied, and making the opposition pay on the powerplay, against Ottawa on Saturday night.