How Hockey ChangedStandings and results
Regular-Season Records Predict NHL Playoff Upsets Almost Exactly
The team with the better record wins about five NHL playoff series in eight. That is almost exactly what its record says it should, once you knock off the luck every record carries. Across 934 series since 1918, a forecast built only on regular-season records expected 345 upsets. There were 342. What changed since the 1970s is the gap between teams, and with it the best team’s odds of winning four rounds in a row.
By Edgehalla ResearchPublished 11 min read
At 8:35 of overtime on April 30, 2023, Carter Verhaeghe’s wrist shot from the right circle slipped under the crossbar in Boston, and the Florida Panthers were through. The Bruins had won 65 games that season, an NHL record. Florida had won 42.
Sixteen days earlier, the analyst Neil Paine had asked why the league’s best regular-season teams so rarely win the Stanley Cup. His answer needed no curse: “The hockey playoffs are just random.”
He was right about the randomness. A century of playoff series says most of it was sitting in the regular-season standings before the first game of spring.
On this page
The better team wins five series in eight, as often as its record says it should
So when a 65-win team goes out in the first round, what failed? The playoffs, or the standings? Start with the record. Since the spring of 1918, NHL teams have met in 934 playoff series in which one of them had the better regular-season record. The better team won 592, about five in eight. Hockey’s favourites lose more than a third of the time.
Paine had checked the Cups won by Presidents’ Trophy winners (more on that below). We wanted to test every series in every round, back to the first one, and see whether the answer has changed. So we built the plainest forecast we could. It reads one thing: each team’s regular-season points percentage, the share of the standings points on offer that a team took. The 2022-23 Bruins took .823 of theirs. Florida took .561. A .600 team took 60%.
Then the forecast takes some of the luck out. Every record has bounces in it. A team that wins 10 of its first 12 games is probably good, just not quite that good, and a whole season works the same way, less strongly. So the forecast pulls each record part of the way back toward average. It pulls harder on a short season: Detroit had the league’s best record in 1994-95, but over 48 games, so the forecast trusts it less than it would trust 82.
From the two adjusted records, it works out each team’s chance of winning one game. From that, it works out the chance of winning a best-of-seven (or best-of-five, or best-of-three). That series chance, set before the first puck drops, is what we mean by “the forecast gave them” from here on. It never sees a playoff result.
It expected the better team to win about 63% of those 934 series. It expected 345 upsets. There were 342.
In every era, the better team won about as often as its record predicted
Share of NHL-vs-NHL playoff series won by the team with the higher regular-season points percentage, against the forecast from the two records alone
The better team won 75.3% of series in 1967-79 and 56.1% since 2013-14; the forecast, from records alone, expected 70.4% and 59.0%.
- Better record won
- Forecast
Read across the pairs of bars. No era misses by more than about five points, and the misses go both ways. Original Six favourites beat the forecast by a little. Cap-era favourites fell a little short. For a century, the playoffs have produced about as many upsets as the regular season said they would. How often the favourite won is another matter.
In the 1970s, the favourite won three times in four
In some decades the better team was close to a sure thing. In the six-team 1950s it won 25 of 30 series. In the 1970s, with the league stretched to as many as 21 teams, it won 68 of 92, about three in four.
That is what made the spring of 1971 a shock. The Boston Bruins, NHL.com recalls, had “rampaged to the NHL’s best regular-season record (57-14-7) and scored a League-record 399 goals.” Montreal had 24 fewer points and a goalie, Ken Dryden, “who had entered the postseason with just six career NHL appearances.” The Canadiens won in seven, and went on to win the Cup.
The favourite’s edge fell after the 1970s, and the forecast saw it coming
Share of playoff series won by the team with the better regular-season record, and the forecast from the records alone
Show data
| Year | Better record won | Forecast |
|---|---|---|
| 1920 | 54% | 60% |
| 1930 | 56% | 58% |
| 1940 | 74% | 67% |
| 1950 | 83% | 70% |
| 1960 | 66% | 64% |
| 1970 | 74% | 72% |
| 1980 | 69% | 66% |
| 1990 | 63% | 64% |
| 2000 | 62% | 60% |
| 2010 | 54% | 58% |
| 2020 | 59% | 61% |
The solid line is what happened. The dashed line is the forecast. Both fall after the 1970s. In the 2010s the better team won 82 of 153 series, 54%, not far above a coin flip. It has recovered a little this decade (59% of 86). The forecast never sees a playoff game, and it drew the same slide: 72% in the 1970s, 58% in the 2010s. Whatever made the favourites weaker was already in the standings.
Since then, the gap between playoff teams has shrunk by more than half
The standings showed a league closing ranks. In the 1970s, the two teams in a playoff series were on average about 14 points apart in points percentage: think of a .600 team against a .462 team. In the 2010s the average gap was about 6 points, a .600 team against a .540 team.
Some of any gap in the standings is luck, so we took the luck out. Each season, we measured how spread out the records were and removed the spread that luck alone would produce over that many games. What is left is roughly the standings you would get if the season never ended. In 1975-76, in an 18-team league, a typical team’s real strength sat about 14 points from the league average. By 2010-11 it was about 6 points. Last season it was about 8. The good teams and the bad teams are much closer together than they were.
NHL teams got closer together, and so did playoff opponents
How far a typical team’s real strength sat from the league average once luck is removed (five-season average), and the average gap in points percentage between playoff opponents, every season since 1926-27
points
Show data
| Season | Typical team’s distance from average, luck removed | Average gap between playoff opponents |
|---|---|---|
| 1926-27 | 12 points | 12 points |
| 1927-28 | 12 points | 5 points |
| 1928-29 | 13 points | 4 points |
| 1929-30 | 12 points | 16 points |
| 1930-31 | 11 points | 4 points |
| 1931-32 | 11 points | 4 points |
| 1932-33 | 9 points | 8 points |
| 1933-34 | 6 points | 3 points |
| 1934-35 | 6 points | 6 points |
| 1935-36 | 8 points | 7 points |
| 1936-37 | 9 points | 7 points |
| 1937-38 | 10 points | 14 points |
| 1938-39 | 11 points | 11 points |
| 1939-40 | 11 points | 8 points |
| 1940-41 | 11 points | 8 points |
| 1941-42 | 12 points | 9 points |
| 1942-43 | 13 points | 6 points |
| 1943-44 | 12 points | 25 points |
| 1944-45 | 13 points | 25 points |
| 1945-46 | 12 points | 6 points |
| 1946-47 | 10 points | 11 points |
| 1947-48 | 7 points | 11 points |
| 1948-49 | 9 points | 10 points |
| 1949-50 | 10 points | 11 points |
| 1950-51 | 10 points | 24 points |
| 1951-52 | 11 points | 14 points |
| 1952-53 | 12 points | 8 points |
| 1953-54 | 11 points | 6 points |
| 1954-55 | 10 points | 13 points |
| 1955-56 | 11 points | 15 points |
| 1956-57 | 9 points | 6 points |
| 1957-58 | 8 points | 15 points |
| 1958-59 | 9 points | 13 points |
| 1959-60 | 10 points | 11 points |
| 1960-61 | 10 points | 12 points |
| 1961-62 | 10 points | 13 points |
| 1962-63 | 10 points | 3 points |
| 1963-64 | 10 points | 6 points |
| 1964-65 | 9 points | 6 points |
| 1965-66 | 9 points | 8 points |
| 1966-67 | 9 points | 6 points |
| 1967-68 | 10 points | 6 points |
| 1968-69 | 10 points | 11 points |
| 1969-70 | 11 points | 9 points |
| 1970-71 | 13 points | 13 points |
| 1971-72 | 13 points | 15 points |
| 1972-73 | 14 points | 13 points |
| 1973-74 | 14 points | 11 points |
| 1974-75 | 14 points | 13 points |
| 1975-76 | 14 points | 12 points |
| 1976-77 | 14 points | 16 points |
| 1977-78 | 13 points | 16 points |
| 1978-79 | 12 points | 12 points |
| 1979-80 | 12 points | 15 points |
| 1980-81 | 11 points | 15 points |
| 1981-82 | 11 points | 13 points |
| 1982-83 | 11 points | 11 points |
| 1983-84 | 11 points | 12 points |
| 1984-85 | 10 points | 11 points |
| 1985-86 | 9 points | 12 points |
| 1986-87 | 9 points | 9 points |
| 1987-88 | 8 points | 9 points |
| 1988-89 | 7 points | 12 points |
| 1989-90 | 8 points | 6 points |
| 1990-91 | 9 points | 12 points |
| 1991-92 | 9 points | 8 points |
| 1992-93 | 10 points | 9 points |
| 1993-94 | 10 points | 8 points |
| 1994-95 | 10 points | 14 points |
| 1995-96 | 8 points | 11 points |
| 1996-97 | 8 points | 9 points |
| 1997-98 | 8 points | 8 points |
| 1998-99 | 8 points | 8 points |
| 1999-00 | 8 points | 7 points |
| 2000-01 | 8 points | 9 points |
| 2001-02 | 8 points | 6 points |
| 2002-03 | 8 points | 8 points |
| 2003-04 | 8 points | 5 points |
| 2004-05 | – | – |
| 2005-06 | 7 points | 6 points |
| 2006-07 | 7 points | 5 points |
| 2007-08 | 7 points | 6 points |
| 2008-09 | 6 points | 8 points |
| 2009-10 | 5 points | 7 points |
| 2010-11 | 6 points | 6 points |
| 2011-12 | 6 points | 5 points |
| 2012-13 | 7 points | 11 points |
| 2013-14 | 7 points | 7 points |
| 2014-15 | 7 points | 5 points |
| 2015-16 | 7 points | 5 points |
| 2016-17 | 7 points | 5 points |
| 2017-18 | 7 points | 5 points |
| 2018-19 | 8 points | 5 points |
| 2019-20 | 8 points | 6 points |
| 2020-21 | 9 points | 9 points |
| 2021-22 | 9 points | 6 points |
| 2022-23 | 10 points | 8 points |
| 2023-24 | 9 points | 5 points |
| 2024-25 | 8 points | 5 points |
| 2025-26 | 8 points | 7 points |
- 1967-68: Expansion to 12 teams
- 1979-80: 16 of 21 make it
- 2005-06: Salary cap
On the chart, the two lines fall together: the league tightened, and so did the playoff matchups. A favourite with a given edge still wins as often as it always did. Across all 934 series, a team ten points better in the standings won about 66% of the time. A team twenty points better won about 78%. On that score we can’t tell the cap era from the 1970s and 1980s. Favourites lose more often now because they are less often much better.
A record is two parts talent to one part luck
That leaves Paine’s randomness. Most of it lives in the standings. A full regular-season record today is roughly two parts talent to one part luck. Across the 18 full 82-game seasons since 2005-06, about two-thirds (67%) of what separates teams in the standings is real. From season to season that has run from about half (53%) to about three-quarters (77%). In 1975-76, with the teams so far apart, it was nearer nine-tenths (88%). So a team that finishes 20 points above average today is, on the best evidence, about 13 points better.
That is why the standings flatter the teams at the top. Take every record at face value and you would expect the better team to win about two series in three (67%). Take the luck out, as the forecast does, and the expectation falls to about five in eight. That is what happened.
Allowing for luck in the standings gets the upset count right
The better team won 63.4% of 934 series; the forecast that allows for luck expected 63.1% and records at face value 67.2%.
The first tile is what happened. The second is the forecast with the luck taken out. The third takes every record at face value and backs the favourite too hard. The last shows the forecast that allows for luck missing the upset count by about three.
Spring hockey adds no extra luck we can find
Is there something extra in the playoffs, a randomness of tight checking and hot goalies that the regular season doesn’t have? We checked game by game. First we used each team’s first-half record to forecast its second-half regular-season games, all 28,525 of them, shootouts left out. There the forecast was slightly too cautious: the better team won a little more often than it said. Then we ran it on 4,866 playoff games. It was about right.
So a playoff game between teams a given distance apart is no measurably closer to a coin flip than a regular-season game. The playoffs simply hand a real but modest edge to a short series.
The biggest upsets: a 48-point gap, a .875 team and a rookie in goal
Of those 342 upsets, the biggest by the gap in the standings is the 1982 Los Angeles Kings, 63 points and a .394 points percentage, against an Edmonton team with Wayne Gretzky, 111 points and .694. The Kings won the best-of-five 3-2. In Game 3 they came back from 5-0 down. The forecast gave them about a 12% chance, the second-lowest of any series winner.
Second by the gap is the best regular season in NHL history. In 1929-30 the Bruins went 38-5-1, a points percentage “still the best in NHL history,” NHL.com notes, and “did not lose back-to-back games all season until being swept by the Canadiens in the best-of-3 Stanley Cup Final.” Montreal was a .580 team. A best-of-three is short enough that the forecast gave the Canadiens about a 17% chance.
The longest shot of all, by the forecast, came in 1945. Montreal had gone 38-8-4 (.800); Toronto was a .520 team with a rookie in goal, Frank McCool, standing in for Turk Broda, “who was away for military service.” The Leafs won the semifinal in six, a 10% shot, then beat Detroit for the Cup. McCool shut out the Red Wings in the first three games of the Final, and NHL.com says he “remains the only goaltender in NHL history with three consecutive shutouts in the Final.”
Toronto’s 1945 win over Montreal was the longest shot of all
The twelve least likely series wins: the forecast’s chance for the winner before the series, from the two regular-season records alone
The least likely series win was Toronto over Montreal in 1944-45 (10.0%); the 1982 Kings were 12.1% and the 2023 Panthers 12.2%.
Every bar is a series winner the forecast gave less than one chance in four. Six of the twelve came in a first round.
Florida in 2023 was a bigger shock than Columbus in 2019
Back to Boston. Florida’s win in 2023 is the biggest upset of the salary-cap era: .561 against .823, the fourth-largest gap ever and the third-least-likely result by the forecast. Boston, as NBC Sports noted, “had not lost three in a row all season.” Florida’s coach, Paul Maurice, said afterwards: “I don’t think you can find a harder team to play against than the Boston Bruins. They will test you.”
Some famous upsets rank lower than memory suggests. Columbus’s sweep of the 62-win Tampa Bay Lightning in 2019 made the Lightning, as CBS Sports put it, “the first team in NHL history to win the Presidents’ Trophy and get swept in the first round.” By the standings it is only the 17th-biggest upset. Dryden’s Canadiens of 1971 go the other way: 36th by the gap, but 11th by the forecast, because in a league as spread out as the 1970s a record was easier to trust.
The Presidents’ Trophy curse is arithmetic
Upsets like these are why the best team so often goes home early, and why people talk of a curse. Since the Presidents’ Trophy was first awarded, for 1985-86, eight of its 40 winners have gone on to win the Cup, 20%. Since 2005-06 it is two of 21. Between expansion and 1985 the best team won 11 of 18.
Neil Paine found the Cup count close to fair. A running team-strength rating (an Elo rating) said Presidents’ Trophy winners should have won 8.2 Cups since 1986. They had won eight, “a minor shortfall at most.”
The series tell the same story, one round at a time, and the curse goes away. Presidents’ Trophy winners have played 102 series and won 70; the forecast expected about 71. On average they went into each series with about a 69% chance. Win four of those in a row and you have a Cup about 23% of the time, close to the one year in five they manage. The early best teams won more Cups because they had fewer rounds to survive: six teams a year made the playoffs from 1926-27 to 1941-42 and only four from 1942-43 to 1966-67, and the best team won 19 of 41 Cups.
The same arithmetic crowns long shots
The same arithmetic makes room for unlikely champions. Since 1985-86 the Cup winner’s median rank in the regular-season standings has been 4th, and 16 of the 40 were fifth or worse. The 2012 Los Angeles Kings were 13th of 30 and, NHL.com notes, “the first No. 8 seed to win the Cup.” “You dream of winning the Cup, and you know what, I’m glad I was the first King to ever lift it,” said their captain, Dustin Brown.
To rank the long-shot champions, we multiply the forecast’s chances for every series a champion won. A team with an even chance in each of four rounds has one chance in 16 of winning all four. That product is the chance, from records alone, that a champion would get through its exact path. The Kings’ path was a 3.1% shot, the fourth-least-likely of 99 Cup runs since 1926-27. Two were longer still, at 1.4% each: the 1979-80 New York Islanders, 5th in the standings, who won the first of four straight Cups, and the 1994-95 New Jersey Devils, 9th in a 48-game season, who swept a Detroit team with the best record in the league. Toronto in 1945 was next (2.6%).
The 1980 Islanders and 1995 Devils were the longest-shot champions
The chance, from regular-season records alone, that the champion would win every series on its path (the forecast’s series chances multiplied together)
The 1979-80 Islanders and 1994-95 Devils each had a 1.4% path; the 2011-12 Kings 3.1%.
Eight of the ten longest shots on the chart had to win four rounds. The Kings, the Penguins and the Islanders each appear twice.
It keeps happening. Last spring Colorado won the Presidents’ Trophy at .738; in the conference final a Vegas team at .579 swept them.
A playoff series is a fair test for its length
- The regular season still matters, about as much as 82 games of evidence can. A better record is a real edge, and roughly a third of it (33%) is luck.
- The playoffs are no measurably more random than the regular season. Favourites lose more now because the league is more even. Playoff games themselves are no closer to a coin flip than they were.
- A Presidents’ Trophy winner that falls short has usually lost a series it was expected to win about seven times in ten. Call it a bad week.
Home ice is one thing the forecast leaves out, and it is real: the better team won 62% of its playoff games at home and 51% on the road. Why that edge has shrunk is a story of its own.
Nobody booked the Kings a plane
In 1982 the Oilers had finished “48 points ahead of the Kings in the Smythe Division standings,” as the Kings’ own website tells it, and “It was expected to be a quick series.” When Edmonton won Game 4 to tie it, the Kings could neither charter a plane nor book a commercial flight to Edmonton for the deciding game, and “ended up sharing an aircraft with the Oilers.”
“That was hilarious. They never dreamed we’d be going back. They never got a plane for us,” Bernie Nicholls said. Daryl Evans, who had scored the overtime winner in Game 3, remembered: “That might have been the quietest flight I’ve ever been on.” The Kings won Game 5, 7-4.
The forecast would have given those Kings about one chance in eight. Forty-one years later it gave the Florida Panthers in Boston almost exactly the same.
Share this article
Betting tools
Live Edgehalla tools on the same subject.
- NHL betting history How favourites, puck lines and totals have paid since 2024.
- NHL Prop Lab Betting ideas tested against real results and market odds.
- Shots on goal props Projected shots for tonight’s skaters against the market line.
- Anytime goal scorer Tonight’s most likely goal scorers, by our model.
Show the work
- Tests written down
- Not in advance (exploratory)
- Results computed
Every playoff game since 1917-18 from the NHL’s game lists, grouped into series by the round and series digits of the game id and checked against the NHL’s records site (977 series both ways; win counts match in 977). Excluded: 11 Stanley Cup series against teams from other leagues (1918-26), the 2020 round robin, and 32 series between teams level on points %, leaving 934. The 1927 Final (Ottawa 2-0-2) ended after four games with two ties. The model: each season, true-talent variance in points % = observed variance minus the variance luck gives over each team’s games (from the league’s shares of wins, half-point results and losses), averaged over five seasons; each record is shrunk toward the league mean by talent ÷ (talent + luck), centred on .500 so overtime-loss points do not count, turned into one game’s win chance by the log5 formula and into a series chance by best-of-N maths (a two-game total-goals series as a normal margin over two games). No home ice, nothing fitted to playoff results. The Presidents’ Trophy tie in 2006-07 and 2020-21 goes to the official winner (Buffalo, Colorado).
| Test | Result | Verdict |
|---|---|---|
| Playoff upsets happen more often than regular-season luck explains | Better record won 63.4% vs 63.1% expected; 342 upsets vs 344.5 (z -0.17), n 934 | No: matches the model |
More tests below
| Test | Result | Verdict |
|---|---|---|
| How much of a full-season record is talent (the share each record is kept, not pulled back to average) | 1975-76 88%; 18 full 82-game seasons since 2005-06: 53% to 77%, median 67%; real-strength spread (one standard deviation, luck removed) 14.1 points in 1975-76, 6.1 in 2010-11, 7.6 in 2025-26; average playoff-opponent gap 13.8 points (1970s) and 5.9 (2010s) | About two parts talent to one part luck today |
| Records taken at face value predict the series | 67.2% expected vs 63.4% actual | Too confident; regression to the mean is needed |
| Holdout: tune the model on even seasons, test on odd | Best scale on even seasons 0.89, which scores worse on odd seasons than the untuned model (log-likelihood -0.83) | Untuned model holds out of sample |
Continues below
| Test | Result | Verdict |
|---|---|---|
| The model fits each era | 1926-42 61.0 vs 59.0 (n 77); 1942-67 70.7 vs 67.5 (n 75); 1967-79 75.3 vs 70.4 (n 97); 1979-93 65.0 vs 66.6 (n 206); 1993-2004 67.5 vs 62.8 (n 160); 2005-13 56.5 vs 58.5 (n 115); 2013-26 56.1 vs 59.0 (n 196) | Every 90% range covers the model |
| Without the 2020 bubble and 2020-21 division seasons | 63.9% vs 63.2%, n 898 | Same |
| Placebo: first-half records forecasting second-half regular-season games | Scale on the model’s odds 1.21 (90% 1.16-1.26), 28,525 games, shootouts left out; playoff games 1.10 (0.96-1.24), 4,866 games (1 = exactly calibrated, below 1 = flatter) | Playoff games no flatter than the regular season |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Modern playoff games are flatter than modern regular-season games | 2005-26: playoff 0.93 (0.64-1.34, 1,780 games) vs regular season 1.21 (1.10-1.32) | Noise: ranges overlap, one era |
| The win-probability curve by standings gap flattened | Logistic slope per point of gap: 1917-42 0.0287 (90% -0.0211 to 0.0782, n 85); 1942-67 0.0632 (90% 0.0339 to 0.1109, n 75); 1967-93 0.0721 (90% 0.056 to 0.0926, n 303); 1993-2005 0.0717 (90% 0.0501 to 0.096, n 160); 2005-26 0.0556 (90% 0.0336 to 0.08, n 311); five slopes vs one, likelihood-ratio 3.39 on 4 df; even/odd seasons 0.0595 / 0.0692 | No detectable change |
| Conference finals are especially upset-prone (since 1979-80) | Round 1 65.0% (n 357, model 62.6); round 2 57.2 (180, 61.6); round 3 53.4 (88, 61.5); Final 68.9 (45, 62.4) | Noise: four rounds, one low |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Shorter series produce more upsets | total goals (2 games) 57.1% (n 35, model 55.2); best of 3 70.0% (n 40, model 61.7); best of 5 63.0% (n 81, model 65.3); best of 7 63.4% (n 778, model 63.3) | No, once gaps are accounted for |
| Presidents’ Trophy winners underperform in series | 70 of 102 won vs 70.8 expected (1985-86 on); cap era 27 of 46 vs 30.4 | No; cap-era shortfall about one SD (noise) |
| Home teams win game 7s | 61.8% in 1967-93 (n 55); 51.7% in 2005-26 (n 87) | Small samples; see the home-ice story |
Frequently asked questions
How often does the team with the better record win an NHL playoff series?
About 63% of the time: 592 of 934 series between NHL teams from 1917-18 to 2025-26. It was about 74% in the 1970s and 54% in the 2010s. Since 1979-80, first-round favourites won 65% (357 series).
How many Presidents’ Trophy winners have won the Stanley Cup?
Eight of 40, from 1985-86 to 2025-26, and two of 21 since 2005-06. They have won 70 of 102 playoff series, close to the 70.8 their records predicted.
What is the biggest upset in NHL playoff history?
By the gap in the standings, the 1982 Los Angeles Kings (.394) over the Edmonton Oilers (.694) in the first round, 48 points apart. By the chance the two records gave the winner, the 1945 Toronto Maple Leafs over the Montreal Canadiens (10.0%), with the 1982 Kings second and the 2023 Florida Panthers over the 65-win Boston Bruins third.
Are the NHL playoffs more random than the regular season?
Not by our test. A forecast built only from regular-season records, allowing for luck, predicted the better team would win 63.1% of series and it won 63.4%. Game by game, the same forecast fits playoff games about as well as regular-season games. Favourites lose more often than in the 1970s because the gap between teams has shrunk.
Do shorter playoff series produce more upsets?
Not once the gap between teams is accounted for. The better team won 70% of best-of-three series (40), 63% of best-of-fives (81) and 63% of best-of-sevens (778), each close to what its matchups predicted.
More in Standings and results
- Verdict: Data story
The Schedule Got Kinder. The Second Night Didn’t.
- Verdict: Data story
Are the NHL’s Stars Pulling Away From Everyone Else?
- Verdict: Data story
Vegas Had the Best First Season of Any NHL Expansion Team
- Verdict: Data story
Does Scoring Depth Win Stanley Cups?
- Verdict: Data story
Same Goals, Half the Blowouts
- Verdict: Data story
Home Ice Lost Almost Half Its Edge in 1992. The Power Plays Didn’t Budge.
Sources and method
- NHL stats (NHL.com), every season since 1917-18
- NHL stats API: playoff and regular-season game lists, and team season summaries (points %)
- NHL Records: playoff series (formats, rounds and series results)
- Neil Paine, “The Boston Bruins Are Cursed! (Maybe.)” (April 14, 2023)
- LA Kings (NHL.com): Leaving on a Jet Plane... with the Edmonton Oilers (April 11, 2019)
- NBC Sports, Keenan Slusher: Panthers oust record-setting Bruins 4-3 in OT in Game 7 (May 1, 2023)
- NHL.com: Stanley Cup champions 1930-1939 (the 1929-30 Bruins)
- NHL.com: Stanley Cup champions 1940-1949 (Frank McCool, 1945)
- NHL.com: Stanley Cup champions 1970-1979 (the 1971 Canadiens)
- CBS Sports, Pete Blackburn: Lightning swept by Blue Jackets in historic first-round upset (April 16, 2019)
- NHL.com: June 11, Kings win Stanley Cup for first time
- NHL API reference (community documentation of the stats endpoints)
Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.