So you read my first article on the NY Islanders forwards and thought, "Well that's all well and good, but how's the team doing? Can Weighted SC SAT be used to evaluate the team as a whole?" The answer is obviously yes or I wouldn't have asked the question.
Through 47 games this season, the NY Islanders' 5v5 GF% is 51.2%. In the same time-frame the team's Weighted SC SAT% is 49.1%, which isn't too far off. Their SAT% is 50.3% and SCF% is 50%, so the advanced metrics agree the Islanders have been lucky so far. But the first quarter of the season tells a different story then the second.
The First 24 Games of the Season
To understand where the Isles held an advantage and where they needed to improve during the first 24 games of the season, I created a table showing the breakdown of the expected goals from the 6 types of scoring chance attempts I track:
I highlighted the categories where the difference in expected goals was greater than 1. In a similar fashion to last year, the Islanders out-chanced their opponents in SC-alone attempts but got destroyed in All Three (SC + 1T + Royal Road) attempts. This has to stop. If we ignore goals scored by either team in this category, the Islanders' GF% would be 1.3% higher, taking the team from 12th to 7th in that category. Defensemen, be more aware of the royal road! Attempts crossing that line have 3x the chance of SC-alone attempts.
To visualize why plus-minus isn't a great way to evaluate a player (which I hope you believe already), picture a defenseman allowing 5 attempts across the Royal Road and none of them score (the odds of that are (1 - 0.234)^5), or 26.4%). Then picture a defenseman who happens to be on the ice for a goal off a harmless shot, 5 seconds after the powerplay, and from a completely random bounce. One defenseman's +/- is 0, and the other's is -3. Who would you rather have on the team?
For the record, the team's Weighted SC SAT% during this time was a paltry 46.6% but their record was 11-8-4. I was very nervous during this time that the team was heading for a crash.
The Next 23 Games of the Season
Did the team correct their tendency to allow more SC + 1T + Royal Road attempts than their opponents during the next 23 games? Nope:
However, we can let that slide because they continued their dominance in SC-alone attempts and widened their advantage in deflection attempts. And who leads the team in SC deflection attempts? Let's say it together, Anders Lee (1.26 per 60) and Matt Martin (0.64 per 60). Wasn't expecting Matt Martin? I wasn't either, but he has 2 deflection goals and 7 attempts in total. John Tavares also has 2 deflection goals but in 207 more 5v5 minutes, which is about a 20.5 game difference for Martin. Before we go and praise Matt, however, his Weighted SC SAT% is 45% and he's 10th among the 12 forwards in Expected Points/60.
The Islanders' Weighted SC SAT% during this most recent time-frame was 51.4%, so now I'm not so gloomy on the team's prospects going forward.
The Islanders: A Working Class Team?
On my last article, RB Long Island wrote, "[If you only track controlled SC attempts,] isn’t what you exclude the Islanders bread and butter? This is not a team of snipers where wingers make passes on their backhand across the entire ice to the other winger waiting to one-time it (see Kane/Panarin video). The Isles are working class heros that grind their way to opponents mistakes and crash the net for rebounds."
Now that was a good point, and I think a lot of us feel that's the identity of this team. However, is this based in reality? After updating my inventory of tracked vs. untracked goals for the 2015-16 season, I saw the following:
I was pretty surprised when I saw how similar the Untracked Goals% was for the Islanders and their opponents. It also validated a bit in my mind that I'm not being too biased towards the Islanders in my tracking. However this I think dispels the myth that the Islanders are more working class than their opponents. Below is a more detailed breakdown of the reason behind these goals weren't tracked:
If you pay attention to the third and fifth columns, you can see the reasons behind the untracked goals are pretty proportionate for the Isles and their opponents. The main difference is that the Islanders scored more rebound goals (14 to 8), and their opponents scored more harmless shot goals (18 to 14). It will be interesting to see if this trend continues over the final 35 games, but I'm optimistic that Jaroslav Halak and Thomas Greiss will stop allowing harmless goals at a 0.472 per 60 pace. As for the rebounds, maybe the Isles do have something here, but a 6 goal difference over 47 GP amounts to one additional win, and I certainly don't want to bank on rebound goals to get to the Stanley Cup. Perhaps RB Long Island was right, or perhaps this is randomness in a 47 game sample size.
One Last Thing
For fun, here's how the Isles have done in Weighted SC SAT% against each opponent since January 4, 2014. Within the division, we own Columbus and New Jersey; do well or even against Pittsburgh and the Rangers; and Washington, Carolina, and Philadelphia are our kryptonite. (Though we've only played Washington twice in the regular season since January 4, 2014.)
Thanks for reading, and stay posted for my article on the defense coming soon!