Home/Road Fenwick splits, and the bizarre Calgary Flames season

 

 

I was comparing home and road splits for team score-tied Fenwick percentage earlier and noticed an odd discrepancy for the Calgary Flames: the team is sixth in the league in possession at home, yet 29th in possession on the road.

Score-tied Fenwick percentage, for those unaware, is the rate of shots and missed shots that were fired on the other team’s net. It’s an excellent predictor of wins and future success, and a fairly decent arbiter of team quality. Naturally, it’s the first thing I look at when a team is going through a hot streak or a slump, to look at this statistic to see if the recent play is due to actual talent or luck.

Naturally, this sort of bears out for the Flames: they are 7th in the Conference in points (despite playing only 20 games) at home, yet 12th in points on the road. Clearly, there’s a huge gap in how they play at home and on the road.

Now, this brings up a few pertinent questions. What makes the Flames play better at home? It was suggested when I brought this up on Twitter that it was due to line matching, but I had my doubts. I noticed that, looking back through the last three seasons at Behind The Net, teams didn’t seem to be able to maintain a home and road split in Fenwick percentage from year-to-year:

So I don’t think that this is due to matchups or arena bias. It’s not even a tough road trip issue: On Calgary’s recent 7-game trip, the team played at 47.7%, which is still poor but above their season average on the road.

Perhaps it’s just sampling bias, which brings up the second pertinent question:  “Which teams in the last three years have been the most noticeably different on their home ice compared to road ice?” The 2008/09 New York Islanders were 50.9% at home, so an above-average possession team, but were a horrible 39.2% on the road (if you’re less than 45, you hardly have a hope at the playoffs. Anything in the low 40s is awful).

Also that year, the San Jose Sharks were an astounding 61.8% team at home, but just 51% on the road for a very impressive split. (I checked the toughest part of the schedule for the ’09 Sharks, a 20-game stretch wherein they played just six times at home, and they had a 54.1% Fenwick rate, so travel wasn’t the issue in their case).

So perhaps the high home and road split numbers of the Calgary Flames are a coincidence: there is a very wide disparity with these teams halfway through the season compared to the full season (the correlation between home and road splits [r-squared] is .19 after half the season, whereas in the last three seasons, after 82 games, the correlation is a much-more significant .45).

What this means is that the Flames and the Edmonton Oilers (56.3% home and 44.5% road) are likely to see those numbers come closer together as the season draws closer to a conclusion. For Calgary, in the thick of a playoff race with a very good chance based on how few home games they’ve played, it’s probably more likely to trend upwards, but they should also start playing a bit better on the road as well.

One other thought: for how much press the Winnipeg Jets earn at home, their home/road split is actually 21st in the league. The main difference between the Jets record at home and the Jets on the road is their luck factor, or PDO, which adds even strength shooting percentage and save percentage. On the road, the Jets have a 97.8% PDO, while at home, it’s 101.6%.