NHL standings point model and alternatives

Over the holidays, I watched the IIHF World Junior Championship (Go Team USA!) and attended a PWHL Takeover game in Seattle between the Boston Fleet and Montreal Victoire. Both events provided an opportunity to watch overtime and shootout hockey.

One of the biggest upsets in the World Juniors was Latvia’s win over Canada in an eight-round shootout. If you haven’t watched it, it’s an impressive display of goaltending dominance, with only one goal scored in 16 shootout attempts. Another great shootout came in Czechia’s bronze medal win over Sweden. The Kraken’s 2023 first-round draft pick, Eduard Šalé, clinched the win in the 14th round, the longest shootout in IIHF history. The full PWHL game is on YouTube as well.

Looking at the standings after these games, I noticed that both the IIHF and PWHL use a three-point standings model, which differs from the NHL’s two-point model. This got me thinking: is one system better than the other, or could they be improved further—perhaps with a five-point model? Let’s review the basics of each standings point model.

NHL two-point model

Most fans are familiar with this model, as the NHL has used it since the “loser” point was introduced in the 1999-2000 season. The system is straightforward:

• 2 points for any win (regulation, OT, or SO).
• 1 point for a loss in OT or SO.
• 0 points for a regulation loss.

Why a loser point?

Before the shootout was introduced in 2005, NHL games could end in a tie, with each team receiving one point. This often led to conservative play in OT, as teams aimed to preserve their single point. To address this, the NHL began awarding one point to any team that reached OT, with an additional point for the game’s winner. The oddity of this system is that regulation wins award two points, while games that go to OT award three points in total (two to the winner, one to the loser).

IIHF/PWHL three-point model

Both the IIHF and PWHL use a model that ensures an equal number of points are awarded in every game. The breakdown is as follows:

• 3 points for a regulation win.
• 2 points for an OT/SO win.
• 1 point for an OT/SO loss.
• 0 points for a regulation loss.

This system places greater emphasis on regulation wins while still rewarding teams that push games into OT or a shootout.

Overtime format

The NHL, IIHF, and PWHL all use a five-minute OT period played at 3-on-3 to open up the ice and increase scoring chances. However, 3-on-3 hockey is rare during regulation play. While it may offer some thrills, the frantic chaos of overtime has a certain charm that feels more like a skills competition than actual hockey. This is why the NHL playoffs stick to 5-on-5 hockey for OT, continuing until a goal is scored. Playoff hockey, in my opinion, is the best in all of sports. Since the NHL values 5-on-5 play in the playoffs, it would be good to prioritize it in the standings model as well.

Five-point model

As I thought about the three-point model, I wondered if it could be improved. By keeping the total points awarded per game consistent, I developed a five-point model:

• 5 points for a regulation win.
• 4 points for an OT win.
• 3 points for a SO win.
• 2 points for a SO loss.
• 1 point for an OT loss.
• 0 points for a regulation loss.

This tiered approach rewards regulation wins the most, followed by OT wins, and finally shootout wins, while still giving some credit to teams that push games into OT or shootout.

Applying the models

I tested these models on NHL standings data from the 2021-22, 2022-23, and 2023-24 seasons. Surprisingly, the same 16 teams made the playoffs in all three models and across all three seasons, though some playoff seedings shifted, potentially altering first-round matchups and possibly rewriting history. Here are the highlights:

2021-22 season

Under the three-point model, Nashville and Dallas swapped wild-card positions in the Western Conference. While both Nashville and Dallas lost in the first round, the three-point model altered the Central Division standings by elevating St. Louis to second place over Minnesota. St. Louis’ stronger record in regulation wins (43 to Minnesota’s 37) supports this ranking.

The five-point model produced similar standings as the two-point model but narrowed the gap between Minnesota and St. Louis to one regulation win, highlighting the increased value of regulation victories.

2022-23 season

The three-point and two-point models produced identical standings in 2022-23, but the five-point model created some shifts. In the Eastern Conference, New Jersey claimed the top spot in the Metropolitan Division, dropping Carolina to second. This adjustment feels accurate, as New Jersey had more OT wins (11 to Carolina’s nine).

In the Western Conference, the Pacific Division saw a reshuffling: Edmonton (1st), Vegas (2nd), Seattle (3rd), and Los Angeles (WC-1). A Vegas-Seattle matchup in the first round would have generated significant hype, though it’s hard to trade away the Kraken’s first-ever playoff victory over Colorado.

2023-24 season

The three-point and five-point models elevated Winnipeg to the top spot in the Central Division in 2023-24, pushing Dallas to second. Winnipeg’s league-high 46 regulation wins contrasted with Dallas’ 21 OT appearances, suggesting the three- and five-point models better reflected team performance.

Conclusion

Is one model better than the others? Personally, the three-point model used by the IIHF and PWHL seems the most accurate and would be a minor improvement over the NHL’s two-point system. The shifts in playoff standings are minimal, but the three-point model’s consistency (equal points per game) provides a more accurate points percentage. For an example, consider a 10-game series where every game ends in OT wins for one team. The two-point model produces an inflated points percentage of 150 percent, while the three-point model delivers a more logical 100 percent, mirroring the fixed points allocation.

Worth noting, the upcoming 4 Nations Face-Off tournament, which is hosted by the NHL (not the IIHF) will feature the three-point model during the round robin portion of the tournament. So, the NHL may actually be considering such a shift for the future.

If you are interested in looking at the expanded set of all the standings, they are available in Google Sheets here: NHL Standings Models.

What do you think about standings point models? Should the NHL adopt the three-point model, or are there other improvements you’d suggest? Let me know your thoughts in the comments below!