Skip to main content

Endorsements

Got any good books to recommend?

Greatest Hits from Working the Net:

Harlow Salon - one of the very best hair salons in Nashville TN.

Runnin' Away With The Central

Continuing our tour through the various NHL divisions, it's time to focus on the Central Division, which for the last several years has served as Detroit's personal happy place (five straight 1st-place finishes), but this year is in danger of going Country, as the Nashville Predators currently stand at the top of the overall NHL standings, with the Red Wings four points back. Both teams have flown under the hockey world's radar in light of Buffalo and Anaheim's hot starts, but are legitimate President's Trophy contenders. The rest of the division has generally played the role of the Washington Generals for Detroit and Nashville over the last couple years, but after recent coaching changes, there is at least hope for competitive balance to return to what was once the close-knit Norris Division of old. Unfortunately I don't have time today to post graphs for all five teams in the division, so I'll focus on Nashville & Detroit for now.

NASHVILLE PREDATORS: Over the course of the season so far, the Predators have generally boasted an Actual Goals per Game trend that tracks well above the Expected value, based on the quantity and quality of shots taken. That would appear to indicate that Nashville's team shooting is among the best in the league. Based on shot type and distance, and average opposing goaltending, they would be expected to have 129 goals at this point, but instead have 163 (3rd in the NHL). This phenomenon appears to be broad-based as well, as out of the seven players with at least ten goals, only one (J.P. Dumont) isn't significantly outperforming his Expected Goals value. As a team they've been ridiculously hot of late, but those sage predictions of offensive depth appear to be bearing fruit.

Defensively, we have a picture that "backup" netminder Chris Mason needs to take into his next contract negotiation. Note how the Actual Goals Against per game line consistently glides along more than 0.5 goals per game lower than Expected Goals Against. When starter Tomas Vokoun went down with a hand injury in late November, he was among the league leaders in the various goaltending categories, but then Chris Mason stepped in and didn't miss a beat, posting a .931 save percentage in 30 games. No other team in the league has this kind of depth to take into the playoffs, period. In terms of impact, that half-goal per game difference makes the difference between Nashville being the 7th-best team in Goals Against, and being down at 19th spot alongside the Rangers and Blues.

DETROIT RED WINGS: This is one of the least interesting graphs we'll see in terms of striking swings or gaps in direction or another, but that is precisely what makes it so remarkable. The Red Wings haven't been shut out all year, and haven't had a one-goal game in the last two months. They've climbed about 0.5 goals per game since October/November, but it's been much more of a gradual process (which appears sustainable) than some spike due to a fortunate streak in the schedule.

At their own end of the rink, the results have been much the same. The skaters have kept Expected Goals Against right around 2.5 per game (outstanding), and the goaltending has been average to moderately above average most of the time. Apprently, part of the secret to keeping your 42 year-old goaltenders fresh is to make sure they face the fewest shots in the league...

Popular posts from this blog

How I'm Trying To Make Money Sports Blogging

To kick off this series of articles general sports-blogging articles here at OTF Classic, I think it's best to start with a comment that Brad left here last week, after I shared my goals for 2012, which include specific revenue targets:
I considered diving into the world of internet marketing myself, but I felt that my friends would hate me for bugging them about stuff. I mean, it's pretty low-risk high-reward, so it's tempting. I wouldn't mind reading about tips on how to maximize impact of blogging in general to make it a legitimate income source. Trying to make money at sports blogging can be a very touchy subject - for the vast majority of us, this is an activity we pursue to both exercise our creativity and share our love of the game, whether it's hockey, football, badminton, whatever, with fellow fans. Mixing that personal conversation with a commercial message can turn people off, especially if it becomes too intrusive for the reader.

It's not unreasonabl…

Canadian Baloney, starring James Mirtle

A tireless refrain from the Canadian media is that Nashville is an absolute failure as a hockey market, and failing to move the team north of the border is an exercise in folly by the NHL.

Our latest exhibit comes from James Mirtle, usually one of the more thoughtful hockey bloggers extant:
But Nashville, quite simply, has proven it cannot sustain an NHL hockey team. Even with the lowest ticket prices in the entire league (I know: I've looked into flying there for a game or two) and a ridiculously forgiving arena lease, the team has had attendance issues despite having one of the best records in the league.

It's not a matter of Canadians not wanting teams in the southern U.S.; I've argued time and again in favour of teams like Dallas and Tampa Bay that have supported their teams and really brought something to the table in terms of bringing news fans and new energy to the game. That's a good thing.

The Predators, however, are not that, not in the beginning and certainly no…

Hooray for PythagenPuck

Back in November, around the quarter-mark of the NHL regular season, I wrote a piece looking at Expected Winning Percentages for each team, based on their Goals For/Goals Against ratio, using the PythagenPuck method as outlined in Alan Ryder's "Win Probabilities" article over at Hockey Analytics. Since we're approaching the end of the campaign, I thought it worth revisiting the two assertions I made back in November - that the Ottawa Senators were capable of getting back into the playoff race, and that the Boston Bruins were in danger of a freefall to the bottom of the standings.

Just to review, the basic idea behind win probability models like PythagenPuck is that over the course of a season, the Goals For and Goals Against numbers can be used to derive a team's winning percentage, within a a surprisingly narrow margin of error. For instance, if you only had GF/GA information, you could make a very good stab at projecting what the standings would look like. The w…