Hockey Season 15/16 - One Last Thing... ;-)

As the 2015-16 season is slowly coming to an end even in North America, our columnist Thomas Roost decided to look back and write "one last thing" about this memorable season. (Ok, it’s one last thing plus 10… ;-) ) 

1. Skills, speed, creativity, grit, and offense beat trapping, destructive defense, and no-risk-strategies. To win, you don’t need to slow down or trap the game anymore. Examples: Pittsburgh Penguins, San Jose Sharks, Tampa Bay Lightning, SC Bern, and HC Lugano

2. Big data is finally entering hockey analytics and charismatic old-school hockey populists are dying out. Examples: Most NHL-teams hired analysts in various positions… and, in my opinion, it won’t be long until some of these guys will be general managers of NHL teams.

3. These days, a modern hockey team is not divided into two offensive, point-producing lines and two checking lines A perfect modern hockey team has twelve skilled, speedy, gritty, smart and reliable forwards, and eight mobile, puck-skilled, smart and composed defensemen. Examples (by tendency): Tampa Bay Lightning, Pittsburgh Penguins, San Jose Sharks

4. The size factor in skaters is fading away to a certain point. Examples: The Pittsburgh Penguins are just the 25th biggest team in the NHL out of 30. SC Bernwon the Swiss championship with the second smallest team overall.

5. The size factor in goalies is becoming more important: Hockey-goalies on an international level, which are shorter than 6.02 ft (188cm), are disappearing. Example: Nowadays, NHL-teams seldom draft goalies smaller than 6.03 (190cm) .

6. Quite often in the playoffs, unexpected things happen. Examples: In the NHL, the number 4 and 11 are facing each other in the Stanley-Cup-Final, and in the Swiss league, it was number 5 vs number 8 in the final

7. The importance of high-end goalies might still be a bit overrated. Examples: No goalie-expert would have predicted that in this year’s Stanley-Cup-Final, the (more or less) rookies Martin Jones and Matt Murray would face each other. SC Bern became champion with solid but not high-end goaltending

8. High-end 18-year-old players can dominate good level pro-leagues. Examples: Patrik Laine in Finland and Auston Matthews in Switzerland

9. The top 6-hockey-nations widened the gap between them and the group of followers. Examples: Various international tournaments on both the senior- and junior-level.

10. It’s possible to win the Stanley-Cup without defender-depth. Examples: Pittsburgh Penguins, San Jose Sharks

But then again:

There is simply no “best way” to play hockey, there is no “perfect way” to build a hockey team because next season, some of these ten points might have already changed … ;-)

And one last thing:

Unfortunately, it’s not impossible that in the 2016 NHL-Entry-draft, neither a Swiss nor a German player will be drafted. I still hope for the one or the other surprise but…. this is indeed alarming!

Cheerio, enjoy the sun, the barbecue, enjoy the Euro2016-soccer-tournament (or football tournament, if you like) and the Olympics in Rio. I definitely will… but I will most definitely miss hockey.

NHL Draft - Some knowledge and some nuggets of wisdom

In just a few days from now, the 2016 NHL Draft will be history. In the days, hours and minutes leading up to the draft – or even during the draft – there are usually hectic speculations, rumors, and real trade offers.

Time to look a bit deeper into the not exact science of judging the value of possible assets in trade talks and the, also, not exact science of predicting the career of talents.

Despite the fact that drafting is an inexact science at best, draft rights and picks are treated like known quantities and qualities in trade talks. The real value of these assets, however, is rarely discussed in concrete terms. Everybody seems to know that a first-round pick is worth more than later round picks. But the interesting question is: how much more are they worth? Some data analyses will help to answer this question.

The analyses show us that the clearly highest value in NHL draft picks lies in the overall top three selections; the largest drop-off in pick value lies between picks three and four. Between picks four and fifteen, there is not a big difference; these picks are worth roughly a bit more than 33% of a 1-3 overall selection.

A sixth-overall pick is more or less twice as valuable as a 30th overall selection. The analyses also show that late first-round picks are just a little bit more valuable than 2nd- or 3rd-round picks.

This means the difference in value between a first-round-pick of a contending team and a 2nd- or 3rd-round pick is significantly smaller than the difference between a late first-round-pick and a top-five-pick of the same round.

So the idea of “quantity over quality” in NHL-drafts in terms of collecting a couple of 2nd- and 3rd-round picks compared to holding on to a late-first rounder is probably a good idea. In the end, the analyses show that, by tendency, the value of middle- and late 1st-round-picks are a bit overrated and the value of 2nd- and later-round picks might be a bit underrated.

I already mentioned more than once that hockey scouting is not an exact science but we still have to try finding some tools to project the success expectancy of NHL-prospects. Some analysts found that the best predictor of future NHL-production is the amount of primary points per game played in the draft year. They compared these numbers with comparable players in their draft year and, of course, there is much more to it (era, league-adjustments, marginal goals by league and so on).

But with this model, the chance of predicting future NHL-success of prospects is clearly higher than without such a model.

As a concrete example, I can present our Swiss forward Timo Meier: The analysts found a group of 43 players with comparable numbers in their draft year. Of those 43, 24 became successful NHL players (NHL-success is determined by them playing 200 NHL games or more). The result shows that Timo Meier’s chance of becoming a successful NHL-player is 55.8% and his projected NHL points per game played will be 0.57. Although this system is helpful it’s important not to fully rely on it; it’s an additional tool, not more and not less.

Thank you to Stephen Burtch and Zac Urback for their great insights.

Horgen, 20th June 2016