Oilers Hackathon: That’s a Wrap

I just submitted my submission to the Oilers Hackathon question. Rather than wait and use the early season information, I opted to simply be done with it. I broke down and made one edit from my previously published figures on goal differential. I originally had the Flyers with the highest goal differential and the Oilers with the lowest. I took 0.55 goals per game off the Philly estimate and gave it to the Oilers. What can I say, that Yakupov goal last night was fantastic!

Below is my actual submission for Question 4. The one page restriction made it impossible to demonstrate all the work that went into the predictions. Oh well.

I guess the Oilers will have some sort of leader board after the contest closes. Rather than use my usual jonathanjoyce37 username, I went with islandfever37. I’m sitting in the Cayman Islands currently and it is beautifully hot, hence the name.

Can Special Team Performance Predict Regular Season Results?

The majority of advanced statistical analysis focuses on even strength performance. I wondered if it was possible to ignore even strength performance and still be able to predict team performance. If so, a team might be able to make roster decisions and game strategy decisions to optimize special team results and drive team success.

My study looked to answer this question. I used my findings to make four predictions for the upcoming season.

For the purposes of my study, I defined team success as making the playoffs. Once in the playoffs, there is a fair bit of luck needed to make it all the way to the Stanley Cup.

Data used

My study was completed solely using the “teamyear.csv” file, inserting columns to compute certain numbers.

Methodology- Season by season review

My study looked at the past 15 NHL regular season to determine how many of the top 16 ranked teams in each statistic made the playoffs each year. My baseline statistic is goal differential; on average, 14.5 of the top 16 teams in goal differential make the playoffs.

I did not expect to exceed the predictability of goal differential, but wondered if there was a statistic that would beat shot differential, itself a reasonable predictor of playoff qualification at 12.1 teams per season on average.

I computed a variety of statistics defined in Appendix A: Power Play differential (“PPd”); Power Play Goal differential (PPGd); a weighted 50%PPd / 50%PPGd differential; a 33%PPd / 67%PPGd differential; and a Power Play % less Penalty Kill % differential (PP%d) (PP% – (1-PK%). In Appendix B, I computed a couple of weighted PPd / PP%d differentials.

Insights gained

Appendix A and Appendix B contain the year by year and average results for each of the statistics computed. The PP%d metric was a slightly better predictor of playoff qualification than shot differential. From Appendix B, the metric comprised of 25%PPd and 75%PP%d (“wPP%d”) increased the playoff qualification predictability to 12.3 of the top 16 ranked teams.

Prediction #1

For the 2012-2013 season, 12 of the top 16 teams in the wPP%d metric will make the playoffs.

Methodology- Team by team review

I then looked at all thirty teams over the past 15 years to determine if there was a predictive element to the wPP%d statistic. Was it a predictor of teams about to make the playoffs or teams about to drop out of the playoffs.

Insights gained

Having been through the thirty teams, I did not see anything in the special team metrics that consistently suggested a team was on the way up or on the way out. Generally speaking, an excellent wPP%d does not appear to cause a team to make the playoffs, though there may be instances where this is the case. My predictions aim to test the notion wPP%d can have a future predictive quality.

Predictions #2, #3, #4 (see Appendix C)

The Detroit Red Wings will finish out of the top 16 in wPP%d and will miss the playoffs.

The New York Islanders will finish in the top 16 in wPP%d and will make the playoffs.

The Edmonton Oilers will make the playoffs if they are around league average in PP shooting percentage (no more than 1% higher) and in the top 16 in wPP%d.

Appendix A: 

  • Power play differential (power plays less penalty kills) (“PPd”)
  • Goal differential (“GF-GA”)
  • Shot differential (“SF-SA”)
  • Power play goal differential (power play goals less short handed goals) (“PPGd”)
  • A couple different blended weightings of PPd and PPGd
  • Power play percentage differential (Power play percentage less penalty kill percentage) (“PP%-(1-PK%”) or (“PP%d”)

Comparison of metricsAverage by metric

Appendix B:

weighted PP percentage differentialAverage PP percentage differential

Note: I refer to the .25PPd + .75PP%d as wPP%d. The .5PPd +.5PP%d case was discarded as an inferior predictor.

Appendix C:

+ = Top 16 in the metric and made playoffs

= Top 16 in the metric, did not make playoffs

& = Not top 16 in the metric and made playoffs

# = Not top 16 in the metric, did not make playoffs

A team on the way down- the Detroit Red Wings

Red Wings

The above table is about as pretty as it gets. Fifteen years of near excellence. Four Stanley Cup wins.

There isn’t a whole lot in the 2012 season to suggest the Red Wings are on the way down. They finished sixth in goal differential and third in shot differential.

Yet there may be some cracks beginning to show. After being top 4 in PPGd differential in nine of ten years, the Wings fell to 17th last year. A freefall to 23rd in PP%d accompanied the power play goal differential drop. Those two stats combine to a wPP%d ranking of 21st.

I predict the Red Wings are on the way down, missing the playoffs this season. Their wPP%d will again be outside of the top 16. 

A team on the way up- the New York Islanders


Since the 2005 lockout, the Islanders have been the Oilers of the Eastern Conference. One trip to the playoffs and six seasons spent near the basement of the NHL standings. Several high draft picks later and folks are wondering when it might come together for the team.

Like the Oilers, the Islanders appear to be slowing trending the right direction. Their PPGd and PP%d have both been barely top 16 the past two seasons. Their wPP%d was 14th last season. Their shot differential was still slightly negative, but in the top 16.

Based on their improvement in special teams the past two seasons, I predict the Islanders will have a top 16 wPP%d this season and sneak into the playoffs.

A team on the way somewhere- the Edmonton Oilers


If I were to predict solely based on wPP%d, I would be more enthusiastic about the Oilers than the Islanders. The Oilers were 3rd in PP%d, 9th in wPP%d and generally horrible everywhere else. If wPP%d were the signal of future prosperity, the Oilers would be my chance to push my chips to the centre of the table.

Last season, the Oilers were dynamite on the PP, converting on about 20% of their opportunities. Their shooting percentage was 16.7% as compared to a league average of 12.6%. With league average shooting, they would have scored twelve less PP goals and fallen to about 25th in PPGd and around 20th in PP%d. That would combine to about 25th in the league in wPP%d.

With those numbers in mind, I would be crazy to suggest the Oilers’ last season wPP%d is an indication of playoffs this season. All else equal, regression is coming and it is a bitch.

For my prediction, I will lean back on the historical average of slightly more than 12 of the top 16 rated wPP%d teams making the playoffs. The Oilers will make the playoffs if they are around league average in PP shooting percentage (no more than 1% higher) and in the top 16 in wPP%d. 


Leave a comment

Filed under Edmonton Oilers

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