NHL Stats Shiny App

Alexander Tin
Posted on May 17, 2020

Objectives of this analysis

 

The objective of my project is to examine current performance metrics by NHL team to understand

(a) which team metrics are most indicative that the team will to make it to the post-season, and;

(b) if the regular season team metrics translate to post-season success.

 

The project is assessed via 5 different aspects of the game:

1) Shot Volume - i.e. how often teams are shooting the puck

2) Shot Accuracy - i.e. how often teams are attempting to put the puck on net and how accurate they are

3) Possession of the Puck - i.e. whether "controlling" the game through posession of the puck is important

4) Goalie Save % - i.e. the goalie save percentage

5) Physicality/Grit - i.e. physicality/ grit of the team

 

Now to dig in a little deeper, the 5 aspects of the game are measured by 10 attributes pertaining to the game:

1) Shot Volume:

  • Total shots taken (CF): Metric known as 'Corsi For. It measures all shots taken against the opposition, including goals, misses or blocked shots
  • Total shots faced (CA): Metric known as 'Corsi Against'. It measures all shots taken by the opposition, including goals, misses or blocked shots

2) Shot accuracy/quality:

  • Shots fired towards net (FF): Metric known as 'Fenwick For'. Similar to Corsi For, but only measures shots that weren't blocked.
  • Goals percentage (ioSH): Measures goal percentage - this is taken as goals/shots taken

3) Possession:

  • Offensive Zone Start Percentage (oZS%): Percentage started in the opponent's half
  • Defensive Zone Start Percentage (dZS%): Percentage started in their own half
  • Face-off Wins (FO%): Wins on the face-off spots/circles.

4) Save%:

  • Goalie Saves (oiSV%): Measures the goalie save percentage. Expressed as over 1.000

5) Physicality/Grit

  • Hits: Hits credited on the opposition
  • Blocks: Blocked shots - willingness of players to put their body on the line

Data Exploration

Click this link to access the shiny app and explore the data! 

Regular Season 2010 - 2019

For the regular season, we have two graphs.

On the left, is a season by season comparison of the teams according to each measured metric, classified according to whether they made the playoffs or not. The dotted line represents the historical average for seasons 2010-2019.

On the right is a pairwise stat comparison of all the teams from 2010-2019, which are also classified according to whether they made the playoffs.

*Note that in 2012-13, the NHL season was shortened (only 48 games were played, instead of the regular 82 games), hence the metrics with absolute figures (i.e. CF, CA, FF, Hits, Blocks) will vary from other years.

Not surprisingly, a high goalie save percentage was the most clear indicator of whether a team would make the playoffs - the better the goalie performs, the less goals the opponent scores, and therefore the higher the chances of winning.

However, it was found that there were no other clear indicators on the measured attributes that separated the teams that did make the playoffs or didn't make the playoffs. Generally, playoff teams had a higher goal conversion (oiSH%), but it was not always true for every season. Physicality and possession were not important factors in the regular season.

 

Graph: Goal conversion among different teams and whether they made the playoffs

Graph: Possession (FO%) did not have a clear relationship to generating more offense.

 

Playoffs

How about the playoffs?

In this tab, the data is related only to teams that made the playoffs. 

On the left, we have the season by season playoff teams by each measured metric. It is color classified according to the final results of the playoffs. Again, the dotted line represents the historical average in the playoffs for seasons 2010-2019.

On the right, we have the pairwise stat comparison of all the playoff teams from 2010-2019, split into different facets according to the team's final results (Champion and Runner-up being the two teams that made it to the final, while the rest were eliminated).

One additional metric for the playoff teams is the average age of the players. 

Once again, it was difficult to differentiate the teams based on the currently measured metrics. There was no one factor that could clearly separate the champion from the rest of the playoff teams.

Conclusion

It was difficult to tell on the team level what differentiated one team from another. Therefore, the conclusion seems to be that the analysis on the team level needs to be complemented with further analysis on the individual skaters level to tell the true story.

Stay tuned!

 

 

 

 

 

 

About Author

Related Articles

Leave a Comment

No comments found.

View Posts by Categories


Our Recent Popular Posts


View Posts by Tags

#python #trainwithnycdsa 2019 airbnb Alex Baransky alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus API Application artist aws beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep Bundles California Cancer Research capstone Career Career Day citibike clustering Coding Course Demo Course Report D3.js data Data Analyst data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization Deep Learning Demo Day Discount dplyr employer networking feature engineering Finance Financial Data Science Flask gbm Get Hired ggplot2 googleVis Hadoop higgs boson Hiring hiring partner events Hiring Partners Industry Experts Instructor Blog Instructor Interview Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter lasso regression Lead Data Scienctist Lead Data Scientist leaflet linear regression Logistic Regression machine learning Maps matplotlib Medical Research Meet the team meetup Networking neural network Neural networks New Courses nlp NYC NYC Data Science nyc data science academy NYC Open Data NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time Portfolio Development prediction Prework Programming PwC python python machine learning python scrapy python web scraping python webscraping Python Workshop R R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn Selenium sentiment analysis Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau team TensorFlow Testimonial tf-idf Top Data Science Bootcamp twitter visualization web scraping Weekend Course What to expect word cloud word2vec XGBoost yelp