The CHL's statistical guru Efraim Larsson is at it again. He uses PDO to explain why the enormous success by Swiss teams during the group stage isn't likely to continue at such a high level in the playoffs.
by Efraim Larsson
During the first two editions of the Champions Hockey League, teams from the Swedish Hockey League and Finnish Liiga basically dominated the tournament. While they again posted solid numbers in this year's group stage – with all teams from both leagues except for Lulea Hockey qualifying for the Round of 32 – neither finished on top when summarizing the statistics and comparing the results between the leagues.
Teams from one particular league had problems during the two first seasons. In 2014-15, four of the six participating teams failed to make it out of the group stage, and last season only one team made it to the knockout stage. This season, all six participating teams made it to the playoffs, and five of them won their respective groups. By now you might have figured out which league I’m talking about – the Swiss National League A, of course!
Three of the Swiss teams – EV Zug, SC Bern and ZSC Lions Zurich – won all four of their group stage games, and when taking all six teams into consideration, the teams from Switzerland managed the tournament's highest point-percentage of 81.94% – 59 points out of 72. As you’ll see more detailed below in this article that’s an extremely solid percentage that will hold the Swiss in lead for a while.
The rankings below are sorted by the previously mentioned point-percentage, which is a ratio of how many points the teams of the specific league have taken, divided by the highest maximum points they could have taken. For example, the NLA has 24 games played, which means that they could have taken 72 points. They took 59 points in the group stage, and 59 out of 72 points is a quotient of 81.94%.
I think you’ve got the hang of it. Here are the rankings:
| # | League | CT | TS | GP | GF | GA | GD | GFA | GAA | SO | GSO | RW | OW | OL | RL | Pt | Pt/Team | W-% | Pt-% |
| 1 | NLA | SUI | 6 | 24 | 78 | 40 | +38 | 3.25 | 1.67 | 4 | 1 | 17 | 4 | 0 | 3 | 59 | 9.83 | 87.50% | 81.94% |
| 2 | SHL | SWE | 8 | 32 | 107 | 67 | +40 | 3.34 | 2.09 | 6 | 0 | 18 | 1 | 6 | 7 | 62 | 7.75 | 59.38% | 64.58% |
| 3 | Liiga | FIN | 8 | 32 | 92 | 63 | +29 | 2.88 | 1.97 | 5 | 2 | 17 | 3 | 2 | 10 | 59 | 7.38 | 62.50% | 61.46% |
| 4 | SVK Extraliga | SVK | 2 | 8 | 19 | 21 | -2 | 2.38 | 2.63 | 0 | 0 | 4 | 0 | 0 | 4 | 12 | 6.00 | 50.00% | 50.00% |
| 5 | CZE Extraliga | CZE | 6 | 24 | 66 | 68 | -2 | 2.75 | 2.83 | 1 | 3 | 8 | 4 | 2 | 10 | 34 | 5.67 | 50.00% | 47.22% |
| 6 | DEL | GER | 6 | 24 | 60 | 65 | -5 | 2.50 | 2.71 | 3 | 3 | 10 | 0 | 1 | 13 | 31 | 5.17 | 41.67% | 43.06% |
| 7 | BLR Extraliga | BLR | 1 | 4 | 10 | 13 | -3 | 2.50 | 3.25 | 0 | 1 | 1 | 1 | 0 | 2 | 5 | 5.00 | 50.00% | 41.67% |
| 8 | EBEL | AUT | 4 | 16 | 37 | 55 | -18 | 2.31 | 3.44 | 0 | 2 | 4 | 1 | 0 | 11 | 14 | 3.50 | 31.25% | 29.17% |
| 9 | DEN Ligaen | DEN | 1 | 4 | 10 | 22 | -12 | 2.50 | 5.50 | 0 | 1 | 0 | 1 | 1 | 2 | 3 | 3.00 | 25.00% | 25.00% |
| 10 | EIHL | GBR | 1 | 4 | 11 | 20 | -9 | 2.75 | 5.00 | 0 | 0 | 1 | 0 | 0 | 3 | 3 | 3.00 | 25.00% | 25.00% |
| 11 | NOR Ligaen | NOR | 2 | 8 | 12 | 28 | -16 | 1.50 | 3.50 | 0 | 2 | 1 | 0 | 1 | 6 | 4 | 2.00 | 12.50% | 16.67% |
| 12 | Ligue Magnus | FRA | 2 | 8 | 8 | 31 | -23 | 1.00 | 3.88 | 0 | 3 | 0 | 0 | 2 | 6 | 2 | 1.00 | 0.00% | 8.33% |
| 13 | POL Ekstraliga | POL | 1 | 4 | 7 | 24 | -17 | 1.75 | 6.00 | 0 | 1 | 0 | 0 | 0 | 4 | 0 | 0.00 | 0.00% | 0.00% |
| CT = Country, TS = Total teams in league, GP = Games Played, GF = Goals For, GA = Goals Against, GD = Goal Difference, GFA = Goals-For Average, GAA = Goals-Against Average, SO = Shutouts, GSO = Games Shut Out, RW = Regulation Wins, OW = Overtime/Shootout Wins, OL = Overtime Losses, RL = Regulation Losses, Pt = Points, Pt/Team = Points-per-Team Average, W-% = Winning Percentage, Pt-% = Point Percentage | |||||||||||||||||||
To be honest, there are no really shocking numbers here except the brilliant ones by the NLA. Looking at the six Founding Leagues, which all have at least six participating teams, the stats are pretty similar to the two previous seasons, with the SHL and Liiga winning between 60 and 65 percent of their games.
Looking at the challenge leagues, which are represented by only one or two teams, we can see some differences. This obviously isn’t weird at all, as one team’s stats from four games can be quite different than four games the following season. With more teams, that will average out more.
If we look at the case of Slovakia, they were represented by the same two teams this season as last season – HK Nitra and HC Kosice – but their results in this year's group stage were much better, increasing from 30 to 50%. Why? Let's look at the teams they were up against. Last season they faced Skelleftea AIK (Sweden), Bili Tygri Liberec (Czech Republic), Red Bull Munich (Germany) and KAC Klagenfurt (Austria). This year their opponents were Black Wings Linz (Austria), Stavanger Oilers (Norway), HC Pilsen (Czech Republic) and SC Bern (Switzerland). The quality of the opponents had a big impact.
However, the quality of competition does not always need to render a negative impact. For example, the Norwegian Ligaen dropped huge in this tournament compared to last season. They went from third-highest point-percentage (54.76%) in last season’s tournament to third-worst after this group stage (16.67%).
The main reason?
I would say Storhamar Hamar’s extremely good showing in the 2015-16 tournament, where they won their group and advanced to the Round of 16, was an aberration, and we probably won’t be seeing that kind of success from a Norwegian team for a while.
I would also like to point out the Belarusian Extraliga, as newcomers Yunost Minsk beat Neman Grodno’s 25.00 point-percentage from last season by quite a margin. They won two of their four games and ended up taking 41.67 of the potential points in the group stage.
Let's get back to NLA ...
During the group stage, the Swiss teams not only ranked second in offence with 3.25 goals per game, but they also collectively had the CHL's best defence, with only 1.67 goals against per game. All together, the Swiss teams had the group stage's best save percentage at 93.98. However, I have a feeling that they won't continue with that level of success in the playoffs.
Why? Let me show you!
There’s a metric called PDO, which combines a team's or league's shooting percentage and save percentage (eg: 8.50 Sh% + 92.45 Sv% = 100.95% PDO).
The PDO is well known to eventually find its way back to near 100% over the long-term. The larger the sample size, the less deviation there will be. Therefore, we can expect that each team's PDO will migrate closer to 100% the longer they remain in the competition.
Here are the shooting percentages and save percentages for all 13 participating leagues:
| League | GP | GF | SG | SGA | Sh% | GA | SA | SAA | Sv% | PDO |
| SHL | 32 | 107 | 1122 | 35.1 | 9.54% | 67 | 688 | 21.5 | 90.26% | 99.80% |
| Liiga | 32 | 92 | 857 | 26.8 | 10.74% | 63 | 917 | 28.7 | 93.13% | 103.86% |
| CZE Extraliga | 24 | 66 | 747 | 31.1 | 8.84% | 68 | 610 | 25.4 | 88.85% | 97.69% |
| NLA | 24 | 78 | 753 | 31.4 | 10.36% | 40 | 664 | 27.7 | 93.98% | 104.33% |
| DEL | 24 | 60 | 717 | 29.9 | 8.37% | 65 | 634 | 26.4 | 89.75% | 98.12% |
| EBEL | 16 | 37 | 459 | 28.7 | 8.06% | 55 | 554 | 34.6 | 90.07% | 98.13% |
| NOR Ligaen | 8 | 12 | 187 | 23.4 | 6.42% | 28 | 260 | 32.5 | 89.23% | 95.65% |
| SVK Extraliga | 8 | 19 | 208 | 26.0 | 9.13% | 21 | 255 | 31.9 | 91.76% | 100.90% |
| Ligue Magnus | 8 | 8 | 158 | 19.8 | 5.06% | 31 | 315 | 39.4 | 90.16% | 95.22% |
| EIHL | 4 | 11 | 74 | 18.5 | 14.86% | 20 | 156 | 39.0 | 87.18% | 102.04% |
| DEN Ligaen | 4 | 10 | 77 | 19.3 | 12.99% | 22 | 143 | 35.8 | 84.62% | 97.60% |
| BLR Extraliga | 4 | 10 | 86 | 21.5 | 11.63% | 13 | 126 | 31.5 | 89.68% | 101.31% |
| POL Ekstraliga | 4 | 7 | 89 | 22.3 | 7.87% | 24 | 212 | 53.0 | 88.68% | 96.54% |
| GP = Games Played, GF = Goals For, SG = Shots on Goal For, SGA = Shots on Goal Average, Sh% = Shooting Percentage, GA = Goals Against, SA = Shots on Goal Against, SAA = Shots on Goal Against Average, Sv% = Save Percentage, PDO = Sh% + Sv% | ||||||||||
So far, we have a pretty small sample size because each team has only played four games. Therefore, the chances of having an abnormally high or low shooting percentage or save percentage are greater in the Champions Hockey League than during a whole season in a domestic league. With that in mind, it is possible that the NLA teams will maintain their high PDO. But I doubt it.
One thing to remember is that if a team is able to collect a lot of points with a low PDO, chances are their success will continue when their PDO gets closer to 100%. If you have a low PDO and don’t manage to win that many games or earn a lot of points, things will get better with time. On the other hand, if your success is based on an extremely high PDO, you’ll likely run into trouble as your shooting and save percentages get worse. Got it?
The NLA teams collectively have the CHL's highest overall PDO at 104.33%, which by itself isn’t tremendously high. However, I doubt that number will be as high after 12 games from the Round of 32 are factored in.
CHL facts from the group stage
- A total of 517 goals were scored in the 96 group stage games, which is an average of 5.39 goals per game. 54.75 % of those goals were scored by the home teams.
- A total of 5534 shots on goal were recorded for a tournament shooting percentage of 9.34 and, obviously, a save percentage of 90.66.
- A total of 19 shutouts were recorded in the group stage, which is exactly the same total as last season. 15 of them were by the home teams. Three teams shut out their opponents twice – Djurgarden Stockholm, ZSC Lions Zurich and Karpat Oulu. You might remember that Karpat goalie Sami Aittokallio had 3 shutouts in last year's group stage, and 2 more in the Round of 32.
- On 26 August, Sparta Prague fired a CHL-record 65 shots on goal (!) in their game against Comarch Cracovia. Cracovia goaltender Rafal Radziszewski let 5 goals pass by, but his record of 60 saves netted him a save percentage of 92.21 for the day.