One of the key questions about the Dallas Stars 2014-15 season that has proven fairly vexingly difficult to answer is just how bad the team's defense was.
In looking at the end results, specifically the goals allowed, it seems like the defense was nothing but a problem all year. If that was the case, especially with the Stars forward core entering the "win now" portions of their careers, it would be reasonable to expect that Stars general manager Jim Nill would be looking to make significant upgrades to the bluelilne.
Thus far this offseason, those moves haven't come (though it's worth noting that some big free agent names remain on the market). Does that mean the Stars defense wasn't as bad as everyone once thought?
In order to answer that question, I took a look at the Stars possession and scoring chance numbers and compared them to those of the San Jose Sharks for both the full regular season and the calendar year of 2015.
Why San Jose?
Well frankly, this started as an offshoot of the Kari Lehtonen - Antti Niemi examination. After all, it would be notable if Lehtonen put up significantly worse numbers against a much worse defense - that would imply Niemi could expect to see his numbers drop overall in Dallas.
In a larger context, the Sharks were a disappointing team that didn't see their defense take the brunt of the criticism. The finished 24th in team goals against at 226, just seven goals behind 18th, right in the middle of the mediocre pack. Of note, the Sharks allowed the most empty-net goals against in the league. Essentially, San Jose was seen as more of an overall team failure than a single driving issue.
But as we've seen before, total number of goals allowed inexorably ties up two very separate things - goaltending and team defense. We looked at how the starting goaltending of the Stars and Sharks compared last weekend. So let's start this analysis with the overall possession-based, 5-v-5 defensive statistics for the Stars and Sharks last season, as complied by War-on-Ice.com.
Both teams get a little boost from only looking at 5-v-5 goals against, though it still is pretty terrible for the Stars. The Corsi percentage, or the number of positive total shot attempts as compared to the total number of shot attempts in the game, makes each team look much better, Dallas especially.
The total Corsi events row is included to point out the obvious - these are two high-event teams. The Stars did a much better job of converting that into offensive zone starts, however, and both team's PDO indicates they are slightly unlucky.
The final two rows are probably the most damning overall, though to some extent it goes with being a high event team. Both teams were on the wrong side of midway in terms of both scoring chances against and high-danger scoring chances against. The number that points to this being more a feature of the high-event style of play is the chances for category. Over the season, the Stars and Sharks were both top five in terms of scoring chances and high-danger scoring chances for, and both ended up on the positive side of the relative percentage in both categories.
But that includes the whole season, including, as I've written before, the tire-fire defense the Stars rolled out for the first two months. How do things look from a shots perspective when evaluated from January 1 onward?
Simply put, the Stars were one of the best possession teams in the entire league at 5-v-5 play for the second half of the season, a number that only underscores how much their goalie problems cost them, as they actually dropped down the list of goals against as the defense got better.
Dallas improved in most measureable possession categories except Corsi event total, offensive zone start percentage and PDO, and all three are close enough to the full-season averages that it may just be a distribution pattern effect rather than a significant swing.
The Sharks, meanwhile stayed basically the same with mediocre defense overall.
Both remained high-event teams, but the number of Corsi events against per game did have a fairly meaningful swing. In the first half, the Stars were 26th in Corsi events against per 60 at 58.8, while the Sharks were 20th at 55.3.
In the second half, the Stars moved all the way up to 16th at 55.1 while the Sharks slipped down to 26th at 58.1. If we relate this back to the goalie statistics from last week, Niemi struggled slightly more as the defense in front of him gave up significantly more chances while Lehtonen struggled more as the Stars defense gave up less.
This shift was evident in scoring chances for the Stars but not the Sharks, indicating the Sharks were likely giving up more from the outside rather than things in tight. For the season, the teams finished 22 and 23 in the category.
So what's the moral of this story?
While the overall Sharks defense remained pretty consistently mediocre, and even got a little bit worse, throughout the season (something reflected in the point totals as well as the goaltender statistics), the Stars defense actually got better. What didn't change was the results - the Stars actually gave up more goals in the second half of the season. Given how the overall defense improved, that's pretty hard to blame on anyone but the goalies.
The defense is a convenient scapegoat for the Stars, and make no mistake, it was not good for the first 8-10 weeks of the season. But in the context of the team as a whole, it was more than adequate in the second half - 16th in its own right and 4th when the Stars' ability to create chances was considered. There's nothing wrong with that for a high-event team that relies on players like Tyler Seguin, Jamie Benn and Jason Spezza to lead one of the league's most dangerous offenses.
But despite that offense and the more-than-adequate defense, the Stars were 29th in goals against in the second half and finished the final 46 games on a 94.5 point pace. That's not good enough to get into the playoffs most years, even without the team defense as a scapegoat.