In Further to The Big Mistake, I opened with stating I read a lot, and large chunk of my reading includes research articles conducted by the blogging community. In many of Arctic Ice Hockey's comment sections, there has been quite a few interesting debates and discussions. Often these have led to requests about whether or not there has been research done of certain topics.
I recently thought: why don't I highlight a few of my favourite articles in order to bring some of our community up to speed on a few things…
This is just a very surface level introduction. There are many, many more articles that I would suggest if I could, but I wanted to start small.
I'd suggest starting with the articles that interest you the most. I'd also suggest in clicking the links they reference to if you want to go further down the rabbit hole.
Shot Metrics: Corsi and Fenwick
Title | Site | Author | Notes |
Shots Directed at Net | Irreverent Oiler Fans | Vic Ferrari | The idea of looking at shot attempts as a hockey statistic in approximating possession is introduced for the first time |
Oiler Scoring Chances and Reasonable Expectations | Irreverent Oiler Fans | Vic Ferrari | Rolling averages of scoring chance% closely resemble Corsi% |
Zone Time | Irreverent Oiler Fans | Vic Ferrari | Corsi and it’s close relationship to zone time |
Loose Ends – Part I: Predicting Future Success | Objective NHL | J. Likens | Team Corsi% with score-tied predicts future wins better than past win% or goal% |
Shots, Fenwick and Corsi | Objective NHL | J. Likens | When are particular shot metrics better at predicting future goal%? |
Two Graphs and 480 Words That Will Convince You on Corsi | mc79hockey | Tyler Dellow | Corsi% and it’s ability to predict future goal% at a player level |
Context and Usage: QoT, QoC, OZS and TOI
Title | Site | Author | Notes |
The Importance of Quality Competition | NHL Numbers | Eric Tulsky | A player’s Corsi% is affected by their opponents, but QoC by Corsi is flawed |
A Competition Metric Based on Ice Time | NHL Numbers | Eric Tulsky | Quality of Competition improved using TOI as opposed to Corsi |
On Zone Starts | Driving the Play | Jared L. | How zone starts affect possession |
Shift Analysis | Hockey Numbers | Java Geek | The conceptual idea of WOWY’s and quality of teammate is introduced |
More on Corsi & Context | Hockey-Graphs | Garret Hohl | Using TOI and Corsi% to estimate player’s impact in terms of goals |
Regression: PDO, On-ice sh% and On-ice sv%
Title | Site | Author | Notes |
On-Ice Shooting Percentage as a Player Talent | Arctic Ice Hockey | Gabriel Desjardins | Four year split sample of on-ice sh% regresses by about 80% to the mean |
Real Effects and Team Shooting Percentage at Even Strength | Irreverent Oiler Fans | Vic Ferrari | Repeatability of even strength sh%, sv%, and shot differentials are substantially different |
Even Strength Outshooting and Team Quality | Objective NHL | J. Likens | 65% of non-luck goal differentials is due to even strength shot differentials |
Shot Distance Allowed as a Team Talent | Arctic Ice Hockey | Gabriel Desjardins | Shot location allowed as a skill does not appear to be a major driver of sv% |
Luck versus Shot Quality in Shooting Percentage | Arctic Ice Hockey | Gabriel Desjardins | Skill in shot quality/location is tiny sliver (10%) of shooting percentage differences |
Shooting Percentage Regression | Broad Street Hockey | Eric Tulsky | Relationship between past and future on-ice shooting percentage is almost nil |
Factoring Regression into Analysis | Broad Street Hockey | Eric Tulsky | Regression of shooting percentage improves projected goal rates |
Fooled by Randomness: How to Evaluate Defensemen | Broad Street Hockey | Eric Tulsky | Defensemen on-ice save percentage regresses 85% back to mean |
Shot Quality Revisited | NHL Numbers | Eric Tulsky | Quantity of scoring chances and shots are closely linked + no persistent sustainability in scoring chance/shot |
Shot Quality Matters, But How Much? | NHL Numbers | Eric Tulsky | Shot quality factors are real but small enough that they don’t significantly alter what is understood by quantity alone |
Goaltending
Title | Site | Author | Notes |
Fooled by Randomness: Goaltender Save Percentage | Arctic Ice Hockey | Gabriel Desjardins | How large of a sample is needed to evaluate using save percentage? |
Judging Goalies: Should We Include PK Sv%? | Broad Street Hockey | Eric Tulsky | When should all minutes or just EV be used? |
Projecting Future Goalie Performance: Updating and Improving Hockey Marcels | Hockey-Graphs | Garik | Using weighted history plus ageing curves to improve projections |
Shot Attempts Against | Brodeur is a Fraud | The Contrarian Goaltender | No relationship between team shots against and save percentage found |
Shot Quality Fantasy | Irreverent Oiler Fans | Vic Ferrari | Save percentage when goaltenders change team’s is within what is expected given variance of sv%. |
Save percentage vs League Average | Brodeur is a Fraud | The Contrarian Goaltender | Using league averages to reduce sv% inflation over different eras. |
Howe and Why: Quality Starts | Hockey Prospectus | Robert Vollman | How save percentage measures a goaltenders ability to improve the probability of winning |
Consistent goaltenders vs inconsistent goaltenders. | NHL Numbers | Eric Tulsky | Goaltender’s seem to be equal in consistency, it is average performance that separates them. |
This is just the start -only the tip of the iceberg- but it is a good start.
Another good source is the NHL Numbers reference library, which contains most of the articles here plus many more.