Loading goals…
hockeygamebot
← Articles

HGB Research

The stress index: which teams play the most stressful games

We added up every move a game's win chance makes, up and down. Montreal, Buffalo and Toronto games move the most this season, 162 to 163 swing points a night. Rangers games move the least, 117.

By Matt Donders4 min read

On November 5, Montreal trailed Utah 2-0 in the first period, 3-2 in the second and 4-3 midway through the third. Montreal tied it with five minutes left and won 30 seconds into overtime. If you watched as a Canadiens fan, your team’s win chance went from 45% to 13%, back to 50%, down near 10% and finally to 100%.

Add up every move the line made that night, including the small ones between goals, and ignore which way each one went. We count those moves in swing points: win chance going from 50% to 51% is one swing point. The first drop alone, 45% to 13%, is 32. Montreal 5, Utah 4 came to 294 swing points.

Eight days earlier, Edmonton beat Calgary 4-0. Edmonton led early, and the win chance never came back toward the middle. The game moved 42 swing points.

A game’s total is its stress index. A team’s stress index is its average per game.

How a game becomes a stress number

1 · Add up every moveMTL 5, UTA 4 (OT) · Nov 5 · MTL's win chance0%50%100%P1P2P3OT0204060game minutes5-4 MTL OT winner: win chance moved 52 swing points52 · MTL OT winner4-4 MTL tying goal: win chance moved 39 swing points39 · MTL tying goal3-4 UTA go-ahead: win chance moved 30 swing points30 · UTA go-ahead294 swing points2 · Minutes in doubtTime with win chance between 25% and 75%0%50%100%0204060game minutesin doubt: 41 of 60.5 min3 · A calm game for contrastSwing points added up over the game, same scale01002003000204060game minutesMTL–UTA 294EDM–CGY 421 · Add up every moveMTL 5, UTA 4 (OT) · Nov 5 · MTL's win chance0%50%100%P1P2P3OT0204060game minutes5-4 MTL OT winner: win chance moved 52 swing points52 · MTL OT winner4-4 MTL tying goal: win chance moved 39 swing points39 · MTL tying goal3-4 UTA go-ahead: win chance moved 30 swing points30 · UTA go-ahead294 swing points2 · Minutes in doubtTime with win chance between 25% and 75%0%50%100%0204060game minutesin doubt: 41 of 60.5 min3 · A calm game for contrastSwing points added up over the game, same scale01002003000204060game minutesMTL–UTA 294EDM–CGY 42

Montreal 5, Utah 4 (overtime), November 5, and Edmonton 4, Calgary 0, October 28. The tags give each goal's move in swing points and who scored it. Panel 3 adds up the moves over each game on one scale.

Source: HGB win-probability model, NHL play-by-play. Home team's win chance after each play. One swing point = win chance moving one percentage point.

Show the numbers
GameSwing pointsMinutes in doubt
MTL 5, UTA 4 (OT)29440.9 of 60.5
EDM 4, CGY 0427.8 of 60.0

How it’s counted

Our win-probability model gives the home team’s chance to win after every play. For each game we take that line from puck drop to the final horn and add the size of every step, up or down.

  • A goal in a close game late moves the line a lot. Utah’s 4-3 goal in the third moved Montreal’s chance from 45% to 15%, 30 swing points. A goal in a 5-1 game barely moves it.
  • Direction doesn’t count, so every reversal adds to the total. A game that heads one way and stays there scores lowest.
  • The number doesn’t care who won. Both teams in a game get the same number.
  • A shootout adds nothing. It has no plays on our line, so the game’s number stops at the end of overtime.

An average game this season moves about 143 swing points.

If you followed our Seismograph posts, the magnitude was built from this same count, with 50 swing points added for shootouts, on a log scale. The stress index drops both. Montreal–Utah would have been about a 7.8 there. Here it’s 294 swing points.

Minutes in doubt

The second number is time. Minutes in doubt counts how long a game spent with neither team clearly ahead, which we define as a win chance between 25% and 75%. Montreal–Utah spent 41 of its 60.5 minutes there. Edmonton–Calgary spent 8 of 60.

Swing points tell you how much the game moved. Minutes in doubt tell you how long it stayed close. A game can be close all night with few big moves, and a blowout can have one wild stretch.

Every team, through November 10

Stress index by team, 2026-27 through November 10
TeamGamesStress per game (swing points)Minutes in doubt per game
Montreal Canadiens1816334.5
Buffalo Sabres1716238.3
Toronto Maple Leafs2016231.0
Utah Mammoth1816031.4
Anaheim Ducks1815727.1
Nashville Predators1815737.2
Pittsburgh Penguins1815631.2
Washington Capitals1615232.4
San Jose Sharks2015232.9
Ottawa Senators1915035.6
Detroit Red Wings1914935.3
Winnipeg Jets1914928.9
New York Islanders1714632.1
Tampa Bay Lightning1714522.6
Florida Panthers1914530.8
Seattle Kraken1614529.3
Boston Bruins1814433.9
Los Angeles Kings1614434.6
Philadelphia Flyers2014130.7
Edmonton Oilers1914128.4
Vancouver Canucks2013921.6
Minnesota Wild2013823.3
New Jersey Devils1913633.8
Chicago Blackhawks1813129.6
Colorado Avalanche1913029.9
Carolina Hurricanes1912930.4
Vegas Golden Knights1912833.6
St. Louis Blues1812728.7
Dallas Stars1712730.3
Calgary Flames1812629.4
Columbus Blue Jackets1612230.8
New York Rangers1911725.2

Sorted by stress per game. Each team name opens its game log, most stressful games first.

Source: HGB win-probability model, NHL play-by-play, 2026-27 regular season through Nov 10, 2026 (292 games).

Montreal (163 swing points a game), Buffalo (162) and Toronto (162) play the most stressful games. Buffalo leads in minutes in doubt, at 38.3 a game. Rangers games are the calmest, at 117 swing points, and Vancouver’s spend the least time in doubt, 21.6 minutes a game.

Stress measures how much a team’s games make you sweat, not how good the team is. A team that wins a lot of close games and a team that loses a lot of them can land in the same place.

Where to find it

  • /stats has a Stress column for every team. Open a team’s game log to sort its games by stress.
  • Every game page shows that game’s stress next to the win-probability chart.
  • The exact rule is in How we measure.