3/25/18: Correlation is Not Causation: The Dallas Stars Catastrophic Moment

The STAAR test for Freshmen taking it for the first time is rapidly approaching. I’ve been working to get my particular group of students ready since, I don’t know, August 24th. One of the things we do is go over released exams from previous years.

Question 9 from last year turned into 20 minutes of discussion when we hit it recently.

Causation

Kids hate the word causation, but as soon as you tell them “cause and effect” they lose their minds trying to eliminate answers that don’t make any sense. Obviously B is bullshit. A and C caused them a lot of difficulty because those can be argued.

You can make an argument in favor of a causation relationship between an increase in bus stops and a decrease in car sales. People are buying less cars because they have more access to buses. Duh. Ice cream sales increase, so sunburn does too? Well, sure. People eat ice cream when it’s hot outside so they’re getting sunburned.

What they have a hard time realizing initially is that ice cream didn’t directly cause the sunburn and that the number of bus stops didn’t directly cause a decrease in car sales. When it rains the water level increases. An increase in rain (water falling from the air) leads to an increase in water levels (fallen water, now on the ground). D is the answer.

Yesterday Bob Sturm threw this out on Twitter:

Bob isn’t wrong. These things did happen after that even took place. Did the Dallas Stars collapse because Ben Bishop got injured, or did the a collapse merely happen around the time Bishop got injured?

Bob mentioned the save percentage over that ten game span so let’s look a little closer at that. I took out the empty net goals, which bumps the save percentage up to 89.7%. It may not seem like much to us, but it probably is to Kari Lehtonen.

I pulled the expected goal totals from the last ten games and the Corsi % to see what has been going on.

Last10

In the last six games in particular the goaltending has been rough. Lehtonen has allowed an extra goal per game over the expected total in each. Over this 10 game stretch Stars goaltending (Bishop did play some) allowed five more goals than they were expected to allow.

Over an 82 game schedule that would be giving up 41 more goals than expected which will get a lot of people fired and/or released.

Here’s the rub though: the goal scoring has been worse. The skaters have scored six goals less than they were expected to by expected goals for.

When you throw those six together with the five extra goals the goaltenders have allowed….yikes. We’re talking a swing of eleven goals in a ten game span. If the games were played just by expected goals the Stars would have gone 6-4 over that stretch.

Our old friend PDO tells the same story.

PDO

Over a long enough timeline PDO tends to 100 because all shots either go in the net or they don’t. The Stars have been trailing 100 during this ten day period on the shooting and goaltending end.

And it isn’t for lack of trying by the top guys. Game Score attempts to quantify the contributions of each player to the outcome of each game. Over the ten game period in question these are the cumulative Game Scores of the big four and the rest of the roster.

GS

Notice Current Whipping Boy Jamie Benn and Tyler Seguin are both at 13. The big four cumulatively dropped a 41 game score over these games compared to 23 from the rest of the roster combined. Those four accounted for 64% of the Game Score contributions during this stretch!

Spoiler alert: the Stars are paper thin. This isn’t just the last ten games obviously. Part of the problem is they do have players who have shown they can play who don’t get much of an opportunity. This chart shows expected goal plus minus per 60 minutes at even strength.

xGF601

I know it’s “hip” to hate on Brett Ritchie, but he isn’t nearly as bad as he’s made out to be and his delta per 60 minutes is double second place. Gemel “Fucking” Smith has been criminally underused and ya boy is at number four (looks for Marc Methot). Jason Spezza has the same expected goal differential per sixty minutes as Tyler Seguin.

I’ll say it louder for the people in the back: ” JASON SPEZZA HAS THE SAME EXPECTED GOAL DIFFERENTIAL PER SIXTY MINUTES AS TYLER SEGUIN.”

I think it’s fair to blame goaltending for this stretch. It’s also fair to say the goal scoring has let them down severely and coaching decisions all season have kept the optimal lineup from ever being utilized.

Either the Stars are getting unlucky, they’re tired, or the chickens are finally coming home to roost at the most inopportune time possible. Ben Bishop did get hurt, and the Stars haven’t been producing over that stretch.

I’m about to dig into some ice cream, but I’m pretty sure I don’t need suntan lotion to do it.

All info pulled from Corsica.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s