the systems numbers - a tale of 2 stories

The other day I found myself bored in a lecture (classic). So I decided to play around on naturalstattrick.com This is a great website, and even though it doesn't have all the bells and whistles of Corsica.hockey, I find it more user friendly and also freezes less. Anyways, I decided to look at the Senators advanced numbers. As many know, and many in the Sens community don't agree with, the Sens never have great advanced statistics. Many Sens fans actively dislike advanced stats because the Senators have won without having good stats. Now the counter argument to this is, "Well they are due to crash eventually". And yet, they still haven't. That got me thinking about why this may be, and I read an article from the Athletic's Graeme Nichols (Better known as @6thSens on twitter) about how The Senators Succeed by Protecting the Front of the Net. This is a great piece and I highly suggest checking it out. I decided to see how the theory was standing up, in terms of stats this season. I wanted to see if the Senators 5v5 stats have improved at all or suggest that they can keep playing well, and not having to rely on strong special teams.

What I did first was look at team stats as a whole. The Sens have not faired well in this, and currently sit at 3rd last (29th) in the NHL in CF%. They are also 7th in Corsi Against, meaning they give up more attempts then a lot of teams. Now the Sens had some early season success by leading early and having score effects catch up to them, so I thought maybe score adjusted CF% would look better. It did not... Sens rank dead last in Score-Adjusted CF% at 45.85% and still at 7th (worst) in CA. That does not seem very ideal. However, as the article said, the Sens do give up shots, but a lot of them are from the outside. This means that CF% may be affected in a negative way.

What I then did was look at High Danger CF%, to see if the numbers held true to the Sens giving up less high danger chances. From here we do see an improvement. The Sens move to 17th in HDCF% at 49.27%, and ALL the way up to 26th in CA. This means overall, they give up the 6th least high danger Corsi attempts. This is a huge improvement for the Sens and a massive reason they seem to be doing so well. When switching to Score-Adjusted CF%, Ottawa falls to 21st overall at 48.61%. and fall to 23rd in HDCA (Meaning they are actually the 9th best in that category). Now that is some massive improvements for the team as a whole, and I think helps explain a little bit the reason behind the Sens success. 

I decided I wanted to take it one step up, and look at individual players for all of these categories as well. What I looked at was players at 5v5 this season with at least 100 minutes of ice time, to get out the crazy outliers with super small sample sizes. This still left 514 players, 327 forwards and 187 Defensemen. Lets start with the forwards. I looked at the worst 20 forwards in CF%, and here were the results:
​

Ottawa Player	CF% Rank / 327	Score Adjusted	Score Adjusted Rank / 327
Tom Pyatt - 38.80CF%	2nd Worst	39.10CF%	2nd Worst
Jean-Gabriel Pageau - 39.37%	3rd Worst	39.66CF%	3rd Worst
Nate Thompson - 42.50CF%	13th Worst	42.93CF%	15th Worst

Those clearly are not good looking numbers. Having 3 players in the bottom 6% of any category is not something good to look at. However, once again lets look at the forwards in High Danger CF%.

I don't even need a chart for this one because once again, all 3 players that were in the bottom 6% of the league in CF% are now no where to be found. In fact, the worst HDCF% forward the Sens have is Zach Smith, and he is sitting at 53rd worst in the league. They have 1 forward in the bottom 16% of the league for HDCF%. Pageau, Pyatt, and Thompson all jump out of the worst 100. To 113th, 159th and 160th worst in the league. This again is a huge change, as it shows even though the Senators have some players giving up huge amounts of shot attempts while on the ice, they are really limiting the amount that are high danger chances.

Now lets take a look at the defence and see if it is the same type of trend or not. 
​

Ottawa Player	CF% Rank / 187	Score-Adjusted CF%	Score-Adjusted CF% Rank / 187
Cody Ceci - 41.72CF%	6th Worst	42.08CF%	5th Worst
Dion Phaneuf - 42.86CF%	11th Worst	42.60	8th Worst
Fredrik Claesson - 43.80CF%	18th Worst	43.68CF%	17th Worst

Once again, the Sens find themselves with 3 defensemen in the bottom 20 of the league, or 3 defensemen in the bottom 10% of the league. That really is not a good look for the Senators, but maybe they can once again turn it around in HDCF%. 

When changed to HDCF% the Sens still have one Dman in the worst 20, and one at 22nd. Cody Ceci sits at 14th in HDCF% with 39.56%. Phaneuf falls to 22nd, and Claesson all the way down to 172nd! Score Adjusted puts Ceci as the 13th worst and Phaneuf as the 14th worst Dmen in HDCF%. 

What to takeaway from this all? Well, I think it does show that the Sens do a good job limiting High Danger chances in their own end. They still give up a ton of shots, but lots of them are not very high risk. The other thing I took away from this, is the need to get a new shutdown defensemen/pairing. Because Ceci and Phaneuf are not effective in a shutdown role. I get their jobs are tough playing against the best competition, but being 2 of the worst 10% in the league at shot attempts against AND High danger attempts is just not good enough for a team who is looking to be a cup contender. I'm not sure how the Sens address that issue, whether they do it internally and give someone in the org. a shot at it. Or if they go make another big deadline move and say pick up a guy like, Chris Tanev? Either way what is clear is that the Sens could use a new shutdown pair, but also that the system and theory of limiting good chances is still holding up for the Sens.