|
Post by trueblue on Dec 8, 2023 10:50:21 GMT -4
The other thing that I think is killing Dumais is his playing time. The kids has multiple 2 min shifts so he is conserving his energy so he can do this, he only works on offense. If the coaching staff stop allowing him to do this you would see more on both ends. But this is the problem in the Q, they don’t develop how top point getters to defend and they don’t hold them accountable defensively because you need the scoring IMO, Midge is killing his energy making him play on the PK.
|
|
|
Post by fabi1309 on Dec 8, 2023 11:35:23 GMT -4
I liked it but I do have 2 questions. 1 how does Halifax have an above 50% chance with Milner starting? I see that I immediately think this is going to be a loss. Not saying the math is wrong I just want to see why the math says it. Does that make sense? Question 2. What the hell is all the matrix stuff in the middle of the post? Thank you! To address your questions: Question 1: The model currently doesn't account for starting goalies. It just uses the average goalie strength. The reason is simple, I just don't know how to get the data before the game starts. The only place I could find the starting goalie is the pre-game report sheet from the CHL website ( this thing). But I haven't found a reliable way to scan the PDF for the right text and also it doesn't seem like there is a consistent time pre-game that the PDF is updated with all information. Maybe you have an idea ? Question 2: It's a table created with the built-in tool for tables that this forum offers. It turns out that when you are on mobile the table tries to fit into the width of your phone - thus creating the matrix effect. Maybe I'll figure out a better way for the next time. The winning % is active at any time really but it updates constantly once more information is available. To get back to question 1 from Statsman, I just don't have the line-ups available as there doesn't seem to be a reliable way for me to access them (no API, it's not on the gamecentre website, and reading this PDF automatically is extremely unreliable). But I am open to ideas how I could get the data! Also, I do have the winning % throughout the game. I also have a graphic visualization of all shots taken + their scoring likelihood as well as an expected goals counter. Here is the data from last nights game: Win Probability After 1st Period (2 Halifax, 1 Drummondville):
Halifax Win Probability: 65% Drummondville Win Probability: 21% Overtime Probability: 14% Win Probability After 2nd Period (3 Halifax, 2 Drummondville):
Halifax Win Probability: 75% Drummondville Win Probability: 8% Overtime Probability: 17% Expected Goals (xG):
Halifax: 2.24 with 23 Shots Drummondville: 2.53 with 31 Shots Visualization of Shots: (let's hope the image quality is good)That's all the data I have so far. I am working on improving and making estimates and predictions more accurate with everyone's help and feedback!! Thanks so much again
|
|
|
Post by hfxfan09 on Dec 8, 2023 22:15:22 GMT -4
Halifax played a solid game last night to bounce back at home, Milner played another solid game he's made strides and no longer makes me nervous when he's in net his confidence seems to be picking up.
I don't even know what to say about the big line anymore, Vidicek with the hat trick. It's scary any time that line is on the ice.(Nice to see Phillips get a goal top shelf too)
|
|