The venerable @Woodguy55 asked me a question about the shots against portion of the quality of competition metric - remember that QualComp is the time-averaged shot differential of a player's opponents. I'll spare you the details of the analysis I did before getting to the answer that I'm going discuss next, but suffice to say that I needed to break the data down by home and road in order to answer his question. At which point, I wondered how last change impacts a player's quality of competition.
Before I get to that, let's get you situated with the leaderboard in home and road QualComp:
Notice a pattern? The list is dominated by St. Louis, Columbus and Nashville, which, as I've noted before, is no surprise given how many games they play against Detroit and Chicago. (It also suggests the potential benefits of using relative shot differential as opposed to raw shot differential for evaluating players league-wide.) One question you're probably asking is why the highest QualComp numbers in the league are still negative. Rather than use both home and road shot differential data, I restricted the dataset to road only - this was primarily to remove arena bias from the + and - components of shot differential - and so the average shot differential is negative. Obviously I've used both home and road ice time in order to calculate the separate QualComp metrics.
At any rate, we're not that concerned about Quality of Competition numbers per se (and you can find them at Behindthenet.ca already anyways), but rather how line matching may be different for different players at home and on the road. Here are the players with the largest increase in QualComp when they're at home:
13 out of our top 20 are clearly among the league's best defensive players and their coaches are clearly putting them out against the other team's top line at home, while on the road, the opposing coach is trying to get his top line out against the second pairing or a weaker defensive forward line. But what do we make of Jason Spezza and Marc Savard? Their coaches try to get them an advantage at home...And opposing coaches go even easier on them on the road? It's possible that it's just an anomaly in the data, but it certainly seems like strange strategy.
What about at the other end of the spectrum?
The most notable thing we see here is the other end of the Central Division battles - someone has to be getting soft minutes as a result of line matching at the top end. Other than that, we've got a handful of teams that aggressively shelter a handful of players (some of whom had their defensive failings undersold at the same time) - nothing here seems ill-advised.