Expected Goals are a better predictor of future scoring than Corsi, Goals

newplot

This piece is co-authored between DTMAboutHeart and asmean.

Introduction

Expected goals models have been developed in a number of sports to better predict future performance. For sports like hockey and soccer where goals are inherently random and scarce, expected goals models proved to be particularly useful at predicting future scoring. This is because they take into account shot attempts, which are better predictors of a team and player’s performance than goal totals alone. 

A notable example is Brian Macdonald’s expected goals model dating back to 2012, which used shot differentials (Corsi, Fenwick) and other variables like faceoffs, zone starts and hits. Important developments have been made since then in regards to the predictive value of those variables, particularly those pertaining to shot quality.

Shot quality has been the subject of spirited debate despite evidence suggesting that it plays an important role in predicting goals. The evidence shows that shot characteristics like distance and angle can significantly influence the probability of a certain shot resulting in a goal. Previous attempts to account for shot quality in an expected goals model format have been conducted by Alan Ryder, see here and here

In Part I, an updated expected goals (xG) model will be presented that accounts for shot quality and a number of other variables. Part II will deal with testing the performance of xG against previous models like score-adjusted Corsi and goals percentage.

Continue reading

A Look into Alex Ovechkin’s Elite Power Play Abilities

"Alex Ovechkin2" by Keith Allison. Licensed under Public Domain via Commons.

Alex Ovechkin2” by Keith Allison. Licensed under Public Domain via Commons.

I don’t know if we’ll ever see a power play quite like that of this decade’s Washington Capitals. We can’t attach a firm date to it because it could extend as far as the end of Alex Ovechkin’s career at this rate, but we know that its peak of power began with the hiring of Adam Oates as Caps head coach back in 2012. Oates had run a successful 1-3-1 power play for the New Jersey Devils with Ilya Kovalchuk as his trigger-man, but nothing even close to the heights he managed to achieve with the man advantage in his two seasons in DC. Barry Trotz, to his credit, has kept the same formation — what’s that old adage about things that ain’t broke? — with only minor tweaks, and last year the power play continued to succeed.

Now there’s a lot to discuss about the formation and its success — I like to think of the Caps’ PP as a work of art more than anything else — but for the sake of this post I’m going to focus in on Alex Ovechkin. Never has there been a more criticized future first-ballot Hall of Famer, nor arguably a more controversial elite goal scorer. It should already be a given that Ovechkin is the best power play goal scorer of all time — he sits fifth overall in PPG/g despite playing in a significantly lower scoring era than his contemporaries like Mike Bossy and Mario Lemieux — but I would argue by the time he retires, he will also likely be the greatest goal scorer of all time period. It’s the man advantage recently, in the latter stages of Ovechkin’s goal scoring peak, that has been the sniper’s bread and butter. Since Oates brought the 1-3-1 to town, Ovi has scored 48% of his goals on the power play, compared to 33% prior to that. He scored 25 power play goals last year, six ahead of the next highest total in Joe Pavelski’s 19. You have to go back another five to reach the player who is in third — Claude Giroux with 14 — indicating how great of a season the Sharks’ center/winger had, but that’s a story for another day.

Continue reading

Paul Bissonnette is Wrong and Right

Photo by Michael Wifall, via Wikimedia Commons; altered by author

Photo by Michael Wifall, via Wikimedia Commons; altered by author

From the outset, I want to say the Player’s Tribune, conceptually, is a wonderful thing. To have players guest post or answer questions without the emotions of a post-game presser or rigid formality of a journalist interview provides great insight to their personalities. And just like anybody we’d encounter in daily life, they say things we agree with, things we don’t agree with, or things we might’ve worded differently. Take, for instance, today’s “Mailbag” with Paul Bissonnette. A majority of the interview, which were questions from readers, were your general enforcer interview questions: best fight, worst fight, scary fight, do you like to fight, etc.

But then there was this final question, which I can only assume came from Mark Spector:

Bissonnette Players Tribune II

Bissonnette’s response, his longest of the interview, was chock full of wrong, with plenty of right on the side.

Continue reading

Practical Concerns: Why Alfy should do analytics

Yesterday, the Ottawa Senators announced the hiring of Daniel Alfredsson as the team’s Senior Advisor to Hockey Operations. Alumni of the Hockey-Graphs blog Emmanuel Perry (who is a Senators fan) took advantage of the situation to come up with this (obvious hoax): https://twitter.com/MannyElk/status/644648872682242048

alf

Now, the more I think about it, the more I believe that having someone like Daniel Alfredsson lead an NHL analytics group is actually a wonderful idea.

Continue reading

Exceeding Pythagorean Expectations: Part 5

“Bryz-warmup” by Arnold C. Licensed under Public Domain via Commons.

Bryz-warmup” by Arnold C. Licensed under Public Domain via Commons.

This is the fifth part of a five part series. Check out Part 1, Part 2, Part 3, Part 4 here. You can view the series both at Hockey-Graphs.com and APHockey.net.

To quickly recap what I’ve covered in the first four parts of this series, I have updated the work that’s been done on Pythagorean Expectations in hockey, and am looking to find out whether teams that have the best lead-protecting players are able to outperform those expectations consistently.

The first step is to figure out how to assess a player’s ability to protect leads. To do this, for every season, I isolated every player’s Corsi Against/60, Scoring Chances Against/60, Expected Goals Against/60 (courtesy of War-On-Ice) and Goals Against/60 when up a goal at even strength. I then found a team’s lead protecting ability for the year in question by weighting those statistics for each player by the amount of ice time they winded up playing that year. For players that didn’t meet a certain threshold, I gave them what I felt was a decent approximation of replacement level ability. For example, here was the expected lead protecting performance of the 2014-2015 Anaheim Ducks in each of those categories.

Screen Shot 2015-09-12 at 4.04.51 PM

Now let’s look a little closer at our Pythagorean Expectation — derived through PythagenPuck.

Continue reading

Exceeding Pythagorean Expectations: Part 4

“Zdeno Chara 2012” by Sarah Connors. Licenced under Public Domain via Commons.

Zdeno Chara 2012” by Sarah Connors. Licensed under Public Domain via Commons.

This is the fourth part of a five part series. Check out Part 1, Part 2, Part 3, Part 5 here. You can view the series both at Hockey-Graphs.com and APHockey.net.

So now, four parts into this five part series, is probably a good time to discuss my original hypothesis and why I started this study.

As I mentioned in my previous post, baseball has already gone through its Microscope Phase of analytics, where every broadly accepted early claim was put to the test to see whether it held up to strict scrutiny, and whether there were ways of adding nuance and complexity to each theory for more practical purpose. One of the first discoveries of this period was that outperforming one’s Pythagorean expectation for teams could be a sustainable talent — to an extent. Some would still argue that the impact is minimal, but it’s difficult to argue that it’s not there.

What is this sustainable talent? Bullpens. Teams that have the best relievers, particularly closers, are more likely to win close games than those that don’t. One guess that I’ve heard put the impact somewhere around 1 win per season above expectations for teams with elite closers. That’s still not a lot, but it’s significant. My question would be, does such a thing exist in hockey?

Continue reading

Exceeding Pythagorean Expectations: Part 3

“Pythagorus Algebraic Separated” by John Blackburne. Licenced under Public Domain via Commons. The 2006 Red Wings may have been the best hockey team since the lost season.

Pythagorus Algebraic Separated” by John Blackburne. Licenced under Public Domain via Commons. The 2006 Red Wings may have been the best hockey team since the lost season.

This is the third part of a five part series. Check out Part 1, Part 2Part 4, Part 5 here. You can view the series both at Hockey-Graphs.com and APHockey.net.  

Since the last post was getting a little long, I decided to hold off on releasing the full Pythagorean results.

Continue reading

Exceeding Pythagorean Expectations: Part 2

“Pythagorus Algebraic Separated” by John Blackburne. Licenced under Public Domain via Commons. The 2006 Red Wings may have been the best hockey team since the lost season.

Pythagorus Algebraic Separated” by John Blackburne. Licenced under Public Domain via Commons.

This is the second part of a five part series. Check out Part 1, Part 3, Part 4, Part 5 here. You can view the series both at Hockey-Graphs.com and APHockey.net.

In Part 1, I looked at some of the theory behind Pythagorean Expectations and their origin in baseball. You can find the original formula copied below.

WPct = W/(W+L) = Runs^2/(Runs^2 + Runs Against^2)

The idea behind the formula is that it is a skill to be able to score runs and to be able to prevent them. What isn’t a skill, however — according to the theory — is when one scores or allows those runs. Teams over the course of weeks or months may appear to be able to score runs when they’re most necessary, to squeak out one-run wins, but as much as it looks like a pattern, it is most often simple variance. If you don’t fully buy into that idea, or you don’t really understand what I mean by variance, read this and then come back. Everything should be a lot clearer.

When applying Pythagorean Expectations to hockey, there are a couple of factors that complicate the matter.

Continue reading

Exceeding Pythagorean Expectations: Part 1

Screen Shot 2015-09-13 at 9.13.44 PM

Nashville Predators vs Detroit Red Wings, 18. April 2006” by Sean Russell. Licensed under Public Domain via Commons. The 2006 Red Wings may have been the best hockey team since the lost season.

This is the first part of a five part series. Check out Part 2, Part 3, Part 4, Part 5 here. You can view the series both at Hockey-Graphs.com and APHockey.net.

The 2015-2016 NHL season is almost here, and our sport has come upon a new phase — arguably the third — in its analytics progression.

Continue reading