OGA Believes…

All over the airwaves and digital sphere right now you are hearing it. Everyone is providing their Kreskinesque predictions of who in the NHL will win it all. Cool. Fun.

Crap we say. Picking where NHL teams will finish the regular and post-seasons before the first puck is dropped and none of the Hockey world’s dynamics have come into play is no more accurate than saying you know the next set of Powerball numbers.

Here at On Goal Analysis, we break down the season into 10-game segments and analyze the ability of teams to make the playoffs based on our dynamic Playoff Qualifying Curve. But we do it during the season and need at least 10, and more like 30, games for a good portion of this season’s ‘tell’ to come through.

And yet, we are compelled to cave a bit and try something new to start this season. We have been looking at the fatigue factor in the NHL this off season in an attempt to see how it feeds into the overall ‘W.’ With that in mind, we want to lay our reputation on the line and predict how fatigue potentially defines each team’s first 10 games. After that, we stick with OGA’s PQC for end-of-season predictions.

What The Heck Are We Doing?

For our attempt at translation, we looked at nine fatigue factors we believe positively or negatively affect each team’s ability to win or lose. Dipping our big toe into the theory pond, we offer that the NHL season has a definable maximum. We believe it takes x, y and z effort to win, and each team’s output is subtracted from the overall maximum the teams and season can produce or provide. Think here that your car can only go so fast and far based on vehicle maintenance and amount of gas in the tank. One huge point here: None of us will not know what x, y and z actually are until the end of the season shows us, so everything until then runs the gambit of someone’s wild-ass guess to another’s best analysis model.

Why so wishy-washy? Because the NHL season as a whole is defined by its macro-level, dynamic and at-speed nature. A perfect example is defining fatigue and predicting how it equals a win or loss. We start before the season with our nine criteria points and measurements. But any way we slice it before the puck drops, we only know today’s dynamic of injuries and not the ones that will be present after Game 1 and beyond or their influence on team play. Does a Daniel Sedin receive a minor injury that sidelines him for five games and takes his 1.268 points per game off of the table (if he even averages 1.268 PPG)? If those points are gone, how many times in those five games is VAN going to need that one more goal to win the game? How could we possibly know that?

And add to the fatigue calculation that all teams ‘slide’ along the schedule until they come together for a night’s contest at different fatigue and capability levels before dynamically moving on to the next contest. That’s eighty-two times each over 185 days, and not a whole lot unlike the Cub Scouts’ Soap Box Derby.

How Do We Call It?

Everywhere teams match up on the schedule, we calculated our fatigue factors to give teams a defining plus or minus goal differential. We then performed simple addition and subtraction to determine a projected winner and loser for each contest. From that, we fearlessly give you our 10-game prediction of the wins each team produces.

The overall winner and losers’ plus or minus goals differential equals an average of approximately .54 goals separating opponents. This math means we apply a scale to our predictions. Any difference of +/- 1 goal or less equals a toss-up with each team having a 50-50 chance of a win. Both teams therefore get a .5 W credit for that game. A +/-1.5 differential presents one team an advantage over the other and makes them more likely to win (+1.5) or lose (-1.5). Above a +/-2 is what we consider a solid call for a win or loss. This scale is projected in the Game 10 Call Key below and to the right of the Eastern and Western Conference breakdown charts below each match up:

Also expressed on the left of our Conference charts is the historical range of Hi, Average and Lo Game 10 wins each team has produced since the Lockout. We also offer this season’s Hi, Projection and Lo win count based on our fatigue criteria. We will provide our report card of the range of wins and by-game predictions in early November when the 10-gamers are complete.

Note here that in the spirit of the x, y and z we know the maximum Conference wins that can be had for the 15 Eastern teams by Game 10 is 118 wins. What? But there are 15 teams x 10 wins for 150 wins, right?

No. Of the 150 possible contests in the Eastern Conference by Game 10, there are 106 games played within the Conference and another 44 against Western Conference teams. And we know all games will not end in a Win and OT/SO Loss for all contests. In fact, 24.15% off all games last year ended as 3-pointers. Somewhat painfully, maximum wins that can come in the Eastern Conference by Game 10 are therefore:

106 games x 1/2 (to remove the redundant game count) = 53 x 1.5 possible wins in-Conference, or 79.5 (relative) wins;

Plus 44 x .7585 wins cross-Conference that are not 3-pointers = 33 wins;

Plus 11 x 1/2 win for cross-Conference OT/SO losses = 5.5 additional wins;

Equals a total of 118 maximum wins/roughly 7.5 maximum wins per team for the Eastern Conference.

Keeping this in mind, below is the Eastern Conference in rank order based on (projected) Hi range, our prediction, and the Lo range in order where tie-breakers are needed:

Our maximum prediction in the East based on fatigue factors accounts for a total of 90 wins earned out of the possible 118, or 6.0 wins per team average. (Actual post-Lockout Eastern performance has been 5.504 wins per team by Game 10.)

Are you surprised the highest projected wins is 8-of-10 for the NYI? Or that the lowest is 4.5-of-10 for PIT? Neither of those teams may capture those numbers. Theoretically, we know no team may make their Hi, Projection or Low number of wins. But what we know of hockey says teams that overachieve have some combination of player chemistry, grit that supersedes personal fatigue, and some opponents who are more fatigued than the general rule when they meet up. Those that underachieve are experiencing the opposite. But we believe fatigue factors will put all of them somewhere between the Hi and Lo limits.

The Western Conference looks like this:

Of the 150 possible contests in the Western Conference by Game 10, 110 games are played within the Conference and another 40 against Eastern Conference teams. Taking into account the same calculations as above, the West produces 117.5 maximum wins / roughly the same 7.5 maximum wins per team as in the East. (Their actual average since the Lockout is 5.511 wins by Game 10.)

For the Western Conference, EDM has the highest projected wins with 8.5-in-10. The lowest is PHX with 5-of-10. While the Hi is higher than, and the Lo is not as Lo as back East, the interesting factoid here is that only six Eastern teams rank below a max win projection of 7 wins or higher while 10 teams do so in the West. What’s that equate to? It will be easier to project overall playoff contenders in the East than the West. Again this season. It is, as this slice of the season clearly shows, the fatigue factor which makes the West seemingly more competitive and simultaneously more unpredictable.


Predictions of who will be in the playoffs and how they will do before the first puck drop of the dynamic NHL season is nothing more than an analytical distraction. It’s airwave filler.

But from the common starting point of the season’s opening, it is possible to use fatigue factors to determine an initial estimate of wins into the first few games of the season. Team fatigue factors play a part in estimating how many wins NHL teams will secure by Game 10.

The total number of possible wins in the Eastern and Western Conferences is almost equal, as is the projected numbers of possible wins per team at Game 10. What is an interesting contrast is that the East has nine teams whose projected Hi in wins is at or above seven, while the West only has five. It means the East will be easier to predict than the West as the top teams will separate from the bottom quicker.

Keep an eye out for OGA’s report card on fatigue and how it affects predictions on the first 10 games of the season.




  1. Pingback: The Long NHL Weekend « The OGA Blogs - October 11, 2011

  2. Pingback: OGA’s Monday Morning Hockey Notes For 17 October « The OGA Blogs - October 17, 2011

  3. Pingback: OGA StatNip – 8 November « The OGA Blogs - November 8, 2011

Leave a Reply

Please log in using one of these methods to post your comment:

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

Twitter Updates

  • ...This suggests that the LAK will defeat CHI in Game 6 to play the NYR for the Stanley Cup 3 years ago
  • Since 2005/6, when a Round 3 team wins Game 5, they lose Game 6 66.7% of the time; also, Western series are determined by Game 5.889... 3 years ago
  • ...So those stats point to the NYR claiming victory tonight and playing the winner from the Western Conference in the Finals 3 years ago
  • Since 2005/6, when a Round 3 team wins Game 5, they lose Game 6 60% of the time; also, Eastern series are determined by Game 5.778... 3 years ago
  • 4. Top 4 in each DIV play 2 rounds seeded within the DIV, 1 round versus other DIV champ and then the Finals between the Conference champs. 3 years ago

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 27 other followers

%d bloggers like this: