Should we trust the Ottawa Senators to understand what makes a player good at defence?

The idea for this article sprung into my mind after seeing this data visualization from the invaluable Corey Sznajder, data tracker-extraordinaire, plotting historical goals for and goals against on a per-game basis from 2000 to 2009:

It's definitely "random historic dataviz" time of the off-season. pic.twitter.com/I3kMeDBGPI

— Corey Sznajder (@ShutdownLine) September 2, 2021

These are the Ottawa Senators that I fondly remember falling in love with.

It is still wild to me that despite posting historic marks in 00/01, 02/03, 03/04, 05/06, and 06/07, the team only made it into just one Stanley Cup Final. I also didn’t realize until viewing this chart that the decline of this generation was so quick. The team went from elite in 06/07, to all-offence, shoddy-defence in 07/08, to middling in both by 08/09.

Bless his heart, Corey has gone ahead and plotted this data for 1970 to 2021. Here’s the last decade for your Ottawa Senators:

It’s been a drastic change.

• The 2012-13 Senators played 48 games in a lockout-shortened season — losing Erik Karlsson 17 games in — and battened down the hatches like Darryl Sutter’s LA Kings.
• The 2016-17 Senators had a magical run, with players buying into Guy Boucher’s system for a year before imploding under the same system the following season.
• The 2014-15 year is the only year that makes me pause, but even that featured a historic run with Andrew Hammond, Mark Stone, and Mike Hoffman leading the Senators down an improbable run at the end of the year to get into the playoffs.
• Otherwise, this organization has been a mess on the ice like it has off it, with the period of 17/18 to 20/21 ranking among the worst single-season performances in the league. /

Another noteworthy event happened right as the team started to decline in 2007: Pierre Dorion, the current General Manager of the Senators, was hired to be the team’s Director of Amateur Scouting. While I don’t think you can necessarily fault him for the NHL team’s on-ice performance until he became assistant general manager midway through the 2013-14 season, we’ve seen over the past few seasons that the players your scouts identify have a big role in forming the identity of your squad. What stands out to me the most about this second set of data, covering 2010 to 2021, is that I don’t think Pierre Dorion has overseen a Sens team that was “good” at defence.

Now, plotting goals might as well be plotting team shooting and save percentages — statistics that naturally fluctuate and can mask true talent. Let’s use some modern day measurements of “defence” — shot attempts against (CA/60) and expected goals against (xGA/60) — while noting that these still have the caveat of splitting the game into a binary of “offence” and “defence” when all of these metrics intersect as we try to determine if any team is “good”. Evolving-Hockey has data stretching back to 07/08 season; here is how the Senators fare in both metrics, with league-wide ranks indicated using brackets for each season:

I recall the Senators being ‘meh’ when I compare them to the team in my childhood memories, and it’s reflected in the data:

• In terms of shot attempts against (CA/60), the Senators rank in the top-half of the league in prevention for six of the 14 seasons — consecutively from 07/08 to 12/13 — and only in the top-10 once, the Clouston-led 09/10 team that lost to a Pittsburgh Penguins team in their prime during the playoffs.
• In terms of expected goals against (xGA/60), the Senators only rank in the top-half of the league in prevention for five of the 14 seasons, again, only ranking in the top-10 during the 09/10 year.
• It’s interesting to compare these data to to earlier plots showing raw goals for and goals against. I now hold the defensive acumen of the 09/10 squad in much higher regard even though they didn’t stand out when you looked at goals. It’s also fascinating to analyze the 16/17 squad that bled shot attempts but protected the front of the net, only to have expected goals against go from a respectable 13th to a bottom-feeder 25th the next season. /

The 09/10 season stands out to me for one other reason. Last year, I used the Weekly Question to explore when last the Sens’ had four NHL-quality top-four defencemen? Using Evolving-Hockey’s Wins Above Replacement model to classify if a player was above-average relative to their peers — in this case, a season WAR greater than 1 — I found the following (emphasis in bold):

“the closest the team came to icing four “good” NHL defencemen, point blank, was 2009-10 where Phillips and Volchenkov were still playing quality hockey in top-four minutes, Erik Karlsson was a slick rookie mentored by Filip Kuba, and Matt Carkner was delivering quality minutes on the back-end. There’s a chance that the teams from Ottawa’s mid-00s glory years might have featured an elite defensive corps, but we don’t have WAR data for seasons before 2007-08.”

Since Erik Karlsson’s departure, the Senators have only had three defencemen post single-season WAR totals greater than 1: Thomas Chabot and Dylan DeMelo in 18/19, and Artem Zub last season.

Looking back, I’m frustrated that the team had Thomas Chabot’s defensive partner in Dylan DeMelo, but split them up for much of 19/20 — hindering them both — before letting DeMelo walk for a four-year deal at $3M that looks sweet considering he put up a single-season WAR of 1.5 last season. It’s miraculous that Chabot’s likely new defensive partner, Artem Zub, exists, given that there was no precedent for an undrafted 24-year-old free agent defenceman from Russia to come into the NHL and have the impact he’s had.

Regardless, pairing Chabot and Zub doesn’t mask the fundamental issue that has plagued the Senators for much of the last 14 seasons, which is that they don’t have enough good defencemen. So, when I ask myself this article’s titular question “should we trust the Ottawa Senators to understand what makes a player good at defence?” My answer is “no”.

Thinking about the future, I think there are reasons for both pessimism and optimism. Let’s start with the former so that we can finish this article without banging our head against a wall.

Pessimism

If I had polled you in advance to ask whether the last season’s team was better defensively than the 2019-20 squad, I imagine that most would’ve said “yes”. I would’ve, too! The data... doesn’t bare that out, which is worrying for a squad that did not add a top-four quality defenceman so far this offseason and instead, brought in competition for the young, highly-routed Erik Brännström and Victor Mete. Want to feel frustrated with me? The data for the 20/21 season are the team’s full season ranks. Here’s how the team fared in CA/60 and xGA/60 post-trade deadline, with the small bois in the lineup:

53.17 CA/60 (19th); 1.98 xGA/60 (8th)

Recall that in fourteen seasons, the Senators have not had a full season xGA/60 rank lower than 10th. 14! FOURTEEN! This team was awful until April! And given the signings of Del Zotto and Holden, there’s a chance we never see a Chabot-Brännström-Mete left side?!?!

Now, it’s absolutely possible that Brännström and Mete can’t keep up their performances for a full season. They’ve also never had a chance to. Again, it comes back to whether you trust the Senators to be good talent evaluators of what makes a good defenceman, and again, my answer to that remains a “no”.

Optimism

Remember how I said that Dylan DeMelo’s single-season WAR this year with Winnipeg was 1.5? Well, Michael Del Zotto’s was 1.1 with Columbus. Now, what’s important to note is that defensive contributions are what fueled DeMelo’s WAR whereas Del Zotto’s were fueled by his offensive contributions. Yes, since signing, we’ve heard him and Senators management talk about Del Zotto’s improvement defensively, but it remains to be seen how he’ll be used this season. He could be of legitimate help in a bottom-six role.

To me, the biggest reason for optimism is basically the motto of the Sens Sickos: we have to save the Senators from themselves. Artem Zub? Popped out of nowhere to be a top-four defenceman. Erik Brännström and Victor Mete? Still with the team, and clearly put their best foot forward to end last season. Despite the team’s offseason priorities, all I can do is hope that their play to end the year helped them gain trust in D.J. Smith’s mind. To D.J.’s credit, he wasn’t afraid to sit veteran players last season in favour of young talent.

Finally, we’ve also seen a shift in how the team’s amateur scouts have identified defencemen since Trent Mann took the helm. Even if the organization missed in Jonny Tychonick, taking him and Bernard-Docker represent a puck-moving skillset that the team lacked. Lassi Thomson? An offensive profile. Max Guénette? A smart, two-way skillset.

When Jake Sanderson joins the squad at the end of his sophomore season, there’s a chance that for the first time since 09/10, the Senators will have three above-average defencemen in their lineup. If Brännström and Mete are allowed to take a step forward this season, they could actually have four. After a decade of mediocrity, that’s reason for celebration.

Note: You can support Corey’s work on Patreon and get access to the full suite of tools on Evolving-Hockey through a site subscription.