How Hockey ChangedBodies
Connor McDavid Got Faster. The League, Not So Much.
Five seasons of NHL tracking show the fastest skaters hitting top speed more often while the typical skater barely moved. And the fastest single skate ever tracked belongs to a winger with seven goals.
By Edgehalla ResearchPublished 11 min read
Connor McDavid hit 24.61 miles an hour against Calgary on Oct. 8, 2025, eight minutes into the second period in Edmonton. The NHL’s tracking had never clocked him faster.
It was one skate of many. McDavid hit 22 mph 151 times that season. The previous high since the league began tracking every skater was Nathan MacKinnon’s 117, in 2023-24. Calgary’s entire roster managed 28.
“You just think, ‘How much faster can it get?’” Nick Foligno, then the Columbus Blue Jackets’ captain, said in 2019, before any of this was measured. “And every year it does.” Five seasons of tracking say he was half right.
On this page
The fastest skaters got faster
The part he got right is at the top. Since 2021-22 the NHL has tracked every skater in every game with cameras and sensors, a system it calls NHL EDGE. Each time a skater hits 20 mph, the system counts a burst. A 22 mph burst also counts as a 20 mph one.
McDavid’s 151 bursts at 22 mph were about 5.5% of every 22 mph burst in the league that season. On his own he out-burst 30 of the NHL’s 32 teams. Only two teams had more: Colorado, and his own Oilers, whose total includes his 151. The next-busiest skater, Philadelphia’s Owen Tippett, had 61. McDavid turned 29 that season, and the tracking had never seen him, or anyone, this fast this often.
McDavid is the extreme case of a real change at the top of the league. To compare players and seasons fairly, we count bursts per 60 minutes on the ice, a full game of ice time, so a player who skates more minutes doesn’t look faster just for that. The charts and tables label it “per 60”.
By that count, skaters hit 22 mph about 13% more often in 2025-26 than in the first tracked season, 2021-22. The number of players whose top speed reached 23 mph went from 89 to 124.
There is one catch, and it is why this story gives two numbers for every rise. Sort every skate into three speed bands: 18 to 20 mph, 20 and up, and 22 and up. In 2022-23, the second tracked season, all three bands dipped at once, and players of every age hit 20 mph less often. The 22 and 20 mph bands have since climbed back past where they started; the 18 to 20 mph band has not. Nobody can say whether the cameras changed that year or the game did. If you count from that dip year instead of the first season, the 22 mph rise is about 24%.
Every speed band dipped in 2022-23, and only the 22 mph band climbed well clear of where it started
Bursts per 60 minutes on the ice, all skaters; each band’s 2021-22 rate is set to 100, so 113 means 13% more often
index
Show data
| Season | 22+ mph | 20+ mph | 18–20 mph |
|---|---|---|---|
| 2021-22 | 100 index | 100 index | 100 index |
| 2022-23 | 92 index | 95 index | 95 index |
| 2023-24 | 102 index | 98 index | 98 index |
| 2024-25 | 106 index | 100 index | 99 index |
| 2025-26 | 114 index | 102 index | 98 index |
- 2022-23: League-wide dip
- Reference: 2021-22 (100 index)
| Season | 22+ mph bursts | 22+ per 60 | 20+ per 60 | 18–20 per 60 | Players with 23+ mph top speed |
|---|---|---|---|---|---|
| 2021-22 | 2,419 | 0.186 | 4.24 | 19.64 | 89 |
| 2022-23 | 2,219 | 0.171 | 4.01 | 18.73 | 89 |
| 2023-24 | 2,473 | 0.191 | 4.17 | 19.17 | 94 |
| 2024-25 | 2,562 | 0.197 | 4.25 | 19.42 | 119 |
| 2025-26 | 2,753 | 0.211 | 4.33 | 19.28 | 124 |
The typical skater stayed about where he was
Below the top end, almost nothing moved. Skaters hit 20 mph only about 2% more often in 2025-26 than in 2021-22, or 8% more counting from the dip year. Skates between 18 and 20 mph, the band just below, barely moved either: about 2% fewer than in the first season, about 3% more than in the dip year.
Now take the regulars, the players with 40 or more games in a season. Every season, half of all regular forwards hit 20 mph somewhere between about 3 and 8 times per full game of ice time. Line the forwards up from fewest bursts to most, and the one halfway down the list hit 20 mph about 5.2 times per 60 minutes in 2021-22. In 2025-26 the forward halfway down hit it about 5.2 times. In the dip year he was at 4.8, so even counted from there, the rise is about 8%. The defenceman halfway down his list rose a little, from about 1.5 to 1.7, and most of that came in the last season.
The typical forward’s 20 mph bursts barely moved in five seasons
Bursts per 60 for the player halfway down the list; the shading leaves out the top and bottom quarters of players; 40+ games
bursts per 60
Show data
| Season | Forwards (halfway down) | Forwards (halfway down) (range) | Defence (halfway down) | Defence (halfway down) (range) |
|---|---|---|---|---|
| 2021-22 | 5.2 bursts per 60 | 3.2 bursts per 60 to 7.8 bursts per 60 | 1.5 bursts per 60 | 0.9 bursts per 60 to 2.3 bursts per 60 |
| 2022-23 | 4.8 bursts per 60 | 3.0 bursts per 60 to 7.4 bursts per 60 | 1.5 bursts per 60 | 0.9 bursts per 60 to 2.2 bursts per 60 |
| 2023-24 | 4.9 bursts per 60 | 3.3 bursts per 60 to 7.5 bursts per 60 | 1.5 bursts per 60 | 1.0 bursts per 60 to 2.4 bursts per 60 |
| 2024-25 | 5.1 bursts per 60 | 3.4 bursts per 60 to 7.7 bursts per 60 | 1.5 bursts per 60 | 0.9 bursts per 60 to 2.3 bursts per 60 |
| 2025-26 | 5.2 bursts per 60 | 3.3 bursts per 60 to 7.7 bursts per 60 | 1.7 bursts per 60 | 1.1 bursts per 60 to 2.6 bursts per 60 |
Read the chart left to right and the forwards’ line is nearly flat. The shading, where the middle half of forwards sit, barely moves either.
Top speed tells the same story. Every regular has one fastest skate a season. For the average regular forward, that skate was about 22.4 mph in 2021-22 and 22.5 mph in 2025-26. For the average defenceman it went from about 21.9 to 22.1. Five seasons, and less than a quarter of a mile an hour: 0.09 mph for the forwards, 0.16 for the defencemen.
The defencemen’s small rise has a face. Matthew Schaefer, the first pick in the 2025 draft, spent 2025-26 on the New York Islanders’ blue line at 18, playing 24:41 a game. He hit 22 mph 33 times, the most by a defenceman in our five seasons of data; the previous high was Cale Makar’s 27 in 2024-25. He hit 20 mph 269 times, the most by any defenceman in a single season since tracking began, NHL.com reported. He became the youngest player to win the Calder Trophy. The Islanders as a team went from 30 bursts at 22 mph in 2021-22 to 134, the third-most in the league.
Schaefer was one of many. In 2021-22, 46% of regular defencemen hit 22 mph at least once. In 2025-26, 56% did.
Hitting 22 mph spread past McDavid, too. Take his bursts out and the league’s 22 mph total still went from 2,330 to 2,602, about 12% more. Now set aside the 25 skaters who hit 22 mph most often each season. In 2021-22 they had 798 of those bursts, and in 2025-26 they had 888, about 11% more. Everyone else, together, went from 1,621 to 1,865, about 15% more. So the skaters outside the top 25 picked up 22 mph bursts faster than the stars did.
In 2022-23, the whole league slowed down at once
Which starting point to trust depends on what happened in 2022-23. That season every speed band fell. The 22 mph band was back above its first-season level a season later, the 20 mph band two seasons later; the 18 to 20 mph band never quite got back. McDavid’s own line shows it. He scored 153 points that season, 30 more than the year before, and the tracking had him hitting 20 mph 19.5 times per 60 minutes in 2021-22 and 17.4 in 2022-23, about 11% fewer.
He had company. Follow the same 500 regulars from 2021-22 into 2022-23 and, as a group, they lost about 8% of their 20 mph bursts. The drop hit players under 26 (−5.9%), players 26 to 29 (−11.2%) and players 30 and over (−8.0%), forwards and defencemen alike. They covered a little less ice per 60 minutes, and their fastest skates slowed by 0.06 mph, but those two slips were about the size of an ordinary change from one season to the next. After that, from one season to the next, the same players’ bursts moved by less than 1% (+0.6%, −0.7%, −0.4%). McDavid was one of those 500 regulars whose number fell that year.
Only in 2022-23 did every age group dip at once
Change in 20+ mph bursts per 60 for the same players, one season to the next; players with 40+ games in both seasons, by age on February 1 of the first season
From 2021-22 to 2022-23 returning players lost 8% of their 20+ mph bursts, in every age group. Later season pairs moved by less than 1% overall.
- Under 26
- 26 to 29
- 30 and over
Getting a year older doesn’t move a whole league like that. In the other seasons, some age groups sped up a little and others slowed down, so the average barely moved. A change in how the system measures would move everyone at once. So would a league-wide change in how the game was played that year. The system itself, 14 infrared cameras above the ice and sensors in the puck and in every player’s jersey, was installed in all 32 arenas during the 2021-22 season, the NHL says. We found no NHL statement that it was recalibrated after that first season, and the data cannot tell a tracking change from a real one. So every rise in this story comes with both starting points: the first tracked season, and the second.
One later change in the data is visible, and it turns out not to matter. From 2025-26, 91% of stored top speeds carry four decimal places, against none before, so something changed in how the speeds are stored. If it had changed how they are measured, the same players’ top speeds would have jumped that season. They moved 0.03 mph, slower, from 2024-25. The 2025-26 rise at the top end is not a side effect of the new format.
Newcomers lifted the everyday burst, and veterans lifted the top end
After the dip, two groups of players pushed the numbers back up. New players pushed up the everyday 20 mph burst. Returning players pushed up the 22 mph top end.
Start with the 20 mph burst. After the dip, the players who stayed in the league barely changed how often they hit 20 mph. So when the league’s number crept up, part of the reason had to be the players coming in and going out. Each summer 128 to 189 skaters leave the tracked league and 137 to 150 arrive. In all four changeovers from one season to the next, the arrivals hit 20 mph more often than the players they replaced.
The arrivals were faster than the players they replaced, every season
20+ mph bursts per 60 for who left, who stayed and who arrived, all skaters
Arriving players hit 20+ mph 5.38 times per 60 in 2025-26; the players who left after 2024-25 had hit 3.90.
- Left
- Stayed (first season)
- Stayed (second season)
- Arrived
The last pair of seasons shows how it works. The 150 skaters who arrived in 2025-26, Schaefer among them, hit 20 mph about 5.4 times per 60 minutes. The 128 who left after 2024-25 had hit it about 3.9 times. The 790 who stayed hit 20 mph about 4.3 times in both seasons.
Young players usually hit 20 mph more often than the veterans they replace. So newcomers pushed the league’s number up a little every season. In the dip year that push was too small to stop the fall. After the dip, the league hit 20 mph more often in each of the next three seasons, and who supplied the rise changed each time. In 2023-24, returning players and newcomers each supplied about half. In 2024-25, returning players supplied most of it. In 2025-26, returning players held level, so the whole rise came from newcomers.
Add those three seasons together, and newcomers replacing the players who left account for a little over half of the rise at 20 mph, about 54%. The rest came from returning players hitting 20 mph a little more often themselves, in 2023-24 and 2024-25.
At 22 mph, it was mostly the returning players. They hit 22 mph more often in each of the three seasons after the dip, including 2025-26, when they held level at 20 mph. About 77% of the extra 22 mph bursts since 2022-23 came from those same players getting faster, and newcomers account for the rest. McDavid is the extreme example. In the season he turned 29, he went from 83 bursts at 22 mph to 151.
The fastest skate ever tracked belongs to a seven-goal winger
If the top end got faster, you might expect the fastest skate on record to be McDavid’s. It isn’t. On March 12, 2026, with 7:55 left in the third period of a home game against Washington, Beck Malenstyn of the Buffalo Sabres was clocked at 24.94 mph. The Sabres’ player bio calls it “the fastest sustained skating speed (24.94 mph) of any NHL skater in the NHL EDGE Era.” Russian Machine Never Breaks (RMNB), a Capitals blog, put it more simply: “Malenstyn set the NHL’s new all-time skating speed record.” It passed Miles Wood’s 24.82, set for Colorado in April 2025. Malenstyn finished that season with seven goals. He hit 22 mph 14 times; 38 skaters did it more often.
The fastest skate of each tracked season has gone to Denis Gurianov (2021-22), Ryan Poehling (2022-23), Owen Tippett (2023-24), Miles Wood (2024-25) and Beck Malenstyn (2025-26), and none of them led the league in 22 mph bursts that season. Ranked by how often they hit 22 mph, those five finished 6th, 8th, 4th, 17th and 39th in their seasons. McDavid led or shared the league lead in 22 mph bursts in four of five seasons and never had the single fastest skate. His 24.61 ranked third in 2025-26, behind Malenstyn and Nathan MacKinnon, who hit 24.78 the same season. MacKinnon’s 22 mph bursts that year fell to 45, from 117 two seasons earlier.
| Season | Skater | mph | Date | Game | His 22+ bursts (rank) |
|---|---|---|---|---|---|
| 2021-22 | Denis Gurianov | 24.60 | Oct. 16, 2021 | DAL at BOS | 40 (6th) |
Continues below
| Season | Skater | mph | Date | Game | His 22+ bursts (rank) |
|---|---|---|---|---|---|
| 2021-22 | Alex Formenton | 24.21 | Oct. 14, 2021 | TOR at OTT | 31 (9th) |
| 2021-22 | Brendan Lemieux | 24.17 | Nov. 27, 2021 | OTT at LAK | 16 (30th) |
| 2022-23 | Ryan Poehling | 24.32 | Oct. 15, 2022 | TBL at PIT | 29 (8th) |
| 2022-23 | Brayden Point | 24.26 | Feb. 7, 2023 | SJS at TBL | 44 (3rd) |
| 2022-23 | Martin Fehérváry | 24.21 | Jan. 31, 2023 | WSH at CBJ | 8 (59th) |
| 2023-24 | Owen Tippett | 24.21 | Jan. 10, 2024 | MTL at PHI | 46 (4th) |
| 2023-24 | Luke Hughes | 24.19 | Feb. 27, 2024 | NJD at SJS | 25 (10th) |
| 2023-24 | Connor McDavid | 24.19 | April 5, 2024 | COL at EDM | 66 (2nd) |
| 2024-25 | Miles Wood | 24.82 | April 10, 2025 | VAN at COL | 23 (17th) |
| 2024-25 | Quinn Hughes | 24.56 | March 22, 2025 | VAN at NYR | 7 (86th) |
| 2024-25 | Martin Necas | 24.49 | Dec. 5, 2024 | COL at CAR | 47 (3rd) |
| 2025-26 | Beck Malenstyn | 24.94 | March 12, 2026 | WSH at BUF | 14 (39th) |
Continues below
| Season | Skater | mph | Date | Game | His 22+ bursts (rank) |
|---|---|---|---|---|---|
| 2025-26 | Nathan MacKinnon | 24.78 | April 5, 2026 | STL at COL | 45 (4th) |
| 2025-26 | Connor McDavid | 24.61 | Oct. 8, 2025 | CGY at EDM | 151 (1st) |
Teams show the same split. A team’s single fastest skate one season tells you almost nothing about its fastest skate the next, because one chase or one breakaway sets it. A team that hits 20 mph often one season usually does it often the next. So a team’s top speed comes and goes with a single chase, while its burst count comes back year after year.
The two fastest teams finished first and 14th
The top end got faster while the typical skater stayed put. So which kind of speed shows up in the standings: McDavid’s 22 mph bursts, or the 20 mph kind a whole roster brings? Edmonton and Colorado give two different answers. Colorado has led the league in 20 mph bursts a game in all five tracked seasons, with Edmonton second every time. In 2025-26 the Avalanche hit 20 mph 32.2 times a game and had the league’s best record, 121 points.
Edmonton had by far the most 22 mph bursts, 258 to Colorado’s 175, and 59% of them were McDavid’s. His 681 bursts at 20 mph were the most in the league too, but they were only about 28% of the Oilers’ total. The rest of the roster supplied the other 72%. The Oilers finished 14th in the standings with 93 points.
So take McDavid out. Remove his bursts from the Oilers’ totals, do the same to every team’s busiest skater, and rank the rest. The rest of the Oilers were second in the league in 20 mph bursts a game in each of the first three seasons, third in 2024-25 and eighth in 2025-26. At 22 mph they were still second in 2025-26, with 107 bursts. Even without McDavid, the Oilers ranked in the league’s top eight every season. In 2025-26, though, most of their 22 mph bursts were his.
Two more teams show how loosely speed and winning travel together. Buffalo was the slowest team in the league in 2023-24, at 15.6 bursts over 20 mph a game. By 2025-26 the Sabres were third-fastest (27.0) and fourth in the standings with 109 points. On April 4, 2026, they clinched a playoff spot, ending what the Associated Press called “an NHL-record 14 seasons of futility.” “The guys have worked so hard to get to this point,” said Lindy Ruff, back for a second stint as their coach. Nobody in that story credited speed.
Calgary went the other way. In 2021-22 the Flames had 111 points and sat 15th in bursts over 20 mph a game. By 2025-26 they were the slowest team in the league (14.4 a game) and finished with 77 points. Their leader in 22 mph bursts, Morgan Frost, had six. McDavid had 151.
Across the league, a typical speed edge is worth about 3 points
Four teams can’t settle it. Buffalo got faster and better, Calgary slower and worse, but plenty else changed on both rosters. To know whether speed wins, you need every team.
Put all 32 teams of 2025-26 on one chart and the faster teams do sit higher. Colorado was fastest and first. Calgary was slowest and near the bottom. But the dots spread wide. Edmonton and Buffalo were less than three bursts a game apart and 16 points apart in the standings. Dallas hit 20 mph only 16.7 times a game, the fourth-fewest in the league, and still had the third-best points percentage, the share of available standings points a team takes.
Faster teams finished higher in 2025-26, but Dallas and Edmonton show how wide the spread is
Bursts over 20 mph per game against points percentage, all 32 teams
Points %
Bursts over 20 mph per game
Show data
| Bursts over 20 mph per game | Points % | |
|---|---|---|
| NJD | 19.3 bursts | 53.0% |
| NYI | 26.2 bursts | 55.5% |
| NYR | 19.4 bursts | 47.0% |
| PHI | 21.2 bursts | 59.8% |
| PIT | 19.2 bursts | 59.8% |
| BOS | 16.4 bursts | 61.0% |
| BUF | 27.0 bursts | 66.5% |
| MTL | 26.1 bursts | 64.6% |
| OTT | 23.8 bursts | 60.4% |
| TOR | 20.0 bursts | 47.6% |
| CAR | 21.8 bursts | 68.9% |
| FLA | 20.7 bursts | 51.2% |
| TBL | 21.6 bursts | 64.6% |
| WSH | 24.3 bursts | 57.9% |
| CHI | 19.9 bursts | 43.9% |
| DET | 20.7 bursts | 56.1% |
| NSH | 16.6 bursts | 52.4% |
| STL | 19.8 bursts | 52.4% |
| CGY | 14.4 bursts | 47.0% |
| COL | 32.2 bursts | 73.8% |
| EDM | 29.7 bursts | 56.7% |
| VAN | 18.8 bursts | 35.4% |
| ANA | 24.2 bursts | 56.1% |
| DAL | 16.7 bursts | 68.3% |
| LAK | 20.5 bursts | 54.9% |
| SJS | 19.4 bursts | 52.4% |
| CBJ | 23.4 bursts | 56.1% |
| MIN | 20.3 bursts | 63.4% |
| WPG | 17.5 bursts | 50.0% |
| VGK | 20.8 bursts | 57.9% |
| SEA | 19.1 bursts | 48.2% |
| UTA | 26.9 bursts | 56.1% |
- Best-fit line: from (14.0 bursts, 49.1%) to (33.0 bursts, 67.1%)
The table below scores how closely speed and the standings moved together, from 0 for no link to 1 for a perfect match. In 2025-26 the score was 0.47, the highest of the five seasons. That score is not a share. Turned into one, speed lines up with only about a fifth of the gaps between teams in the standings that season. Everything else a team does covers the rest. Luck alone would rarely produce a season that tight. With only 32 teams, though, the true score could be as low as about 0.2 or as high as about 0.7. In the other four seasons the dots spread wider. Put all five seasons together, 160 team-seasons with each team’s speed measured against that season’s average, and the score is 0.29.
A plainer split says the same thing. Each season, take the eight teams that hit 20 mph most often and the eight that hit it least. The fast eight took a bigger share of the available points in all five seasons. Their points percentage was about 3 to 9 higher, which over 82 games is roughly 5 to 14 standings points.
| Season | Fastest 8 | Slowest 8 | Gap | Link between speed and points (0 none, 1 perfect) |
|---|---|---|---|---|
| 2021-22 | 57.2% | 48.5% | +8.7 | 0.35 |
| 2022-23 | 57.4% | 54.4% | +3.0 | 0.21 |
| 2023-24 | 54.4% | 50.5% | +3.9 | 0.20 |
| 2024-25 | 58.2% | 53.7% | +4.5 | 0.25 |
| 2025-26 | 60.9% | 52.8% | +8.1 | 0.47 |
Some of that is just good teams having good skaters. So we ran a harder test, and in effect it works like this. Take two teams that finished one season with the same points. One hit 20 mph about 4 more times a game than the other, a common gap between NHL teams. Did the faster one do better the next season? Across 126 season-to-season comparisons, it did, a little: about 3 more points in the standings. The true figure could be as small as about 1 point or as big as about 5. It is probably real, and it is small. It holds with Colorado left out. In the test’s own units, each extra burst a game was worth about 0.52 points of points percentage the next season, give or take 0.19; without Colorado, 0.56, give or take 0.23. Speed also adds little to a forecast. Last season’s points alone account for 32% of the differences in next season’s points, and adding speed lifts that only to 36%. The same teams show up in several of the 126 comparisons, too, so the test is less sure than it looks.
For scale, luck alone typically moves a team’s season total up or down by about 8 or 9 points. The test used a team’s 20 mph bursts a game, the everyday kind a whole roster brings every night, and the kind that rose only 2% to 8% across the league. Edmonton shows how loose the link is: second in the league in those bursts every season, and 14th in the standings in 2025-26, with by far the most 22 mph bursts.
What five seasons of tracking cannot say
Everyone else is right there underneath
Foligno said something else in that 2019 interview. “It’s amazing that the elite players in this league seem to have that ability to pull away from guys,” he said, “and everyone else is right there underneath it.” The cameras back that line better than his first one. For the typical skater, the game got a little faster. For McDavid, it got faster 151 times a season.
And the fastest skate of the tracking era came late in a March game in Buffalo. Malenstyn, in the words of RMNB, “tried to win a foot race to beat out an icing in the offensive zone.” The cameras clocked him at 24.94 mph. He couldn’t finish the play.
Share this article
Player tools
Live Edgehalla tools on the same subject.
Show the work
- Tests written down
- Not in advance (exploratory)
- Results computed
NHL EDGE season totals for every skater and team, regular seasons 2021-22 to 2025-26, normalised per 60 minutes of ice time (league ice time 12,979 to 13,017 skater-hours, so per-game and per-60 rates tell the same story). Regulars are skaters with 40+ games; average top speed is the mean of each regular’s season-best skate (forwards 22.40 to 22.49 mph, defencemen 21.92 to 22.08). Recalibration check: the same players followed season to season (40+ games in both), by age band on February 1 and by position; the 2021-22 to 2022-23 step is −8.0% (n 500, standard error about 1.4 points), with distance per 60 and top speed (−0.06 mph) also lower by amounts near the normal year-to-year wobble. Turnover split: arrivals, leavers and stayers weighted by ice time; the stayers’ share is their own rate change weighted by their share of ice time, summed over the three pairs since 2022-23. Team test: last season’s speed against next season’s points percentage, controlling for last season’s points (a regression, which is what “teams that finished with the same points” stands for in the body), with and without Colorado (0.52 points of points % per burst a game, SE 0.19; without Colorado 0.56, SE 0.23; variance explained 32% to 36%). Named players and fastest skates come from the same EDGE tables (league ranks share ties) and goals and points from the NHL stats API.
Continues below
From 2025-26, 91% of stored top speeds carry four decimals, against none before; the same players’ top speeds did not step at that change (−0.03 mph from 2024-25). In the fastest-skates table the 2025-26 speeds carry four decimals in the source and are shown to two. The top-25 and outside-top-25 counts are season totals of 22+ mph bursts, not rates. 18–20 mph bursts per 60, exact: −1.9% from 2021-22, +2.9% from 2022-23. The two starting points, exact: from 2021-22 the rises are 13.5% at 22 mph and 2.3% at 20 mph; from 2022-23, 23.8% and 8.1%. Turnover, 2024-25 to 2025-26, 20+ mph bursts per 60: arrivals 5.38, leavers 3.90, stayers 4.27 then 4.27. Turnover season by season: stayers’ 20+ per 60 went 4.05 to 4.13 (2022-23 to 2023-24), 4.16 to 4.23 (2023-24 to 2024-25), 4.27 to 4.27 (2024-25 to 2025-26); their own change, weighted by their share of ice time, is about 49%, 86%, −1% of each season’s league rise, and turnover the rest. Turnover added to the league rate in all four pairs, the dip pair included. These stayers are every skater who played in both seasons, with groups weighted by ice time; the returning regulars in the age chart (40+ games in both, each player counted once) moved by less than 1% in the same pairs; the two measures count different players and weight them differently.
Continues below
Edmonton without its top skater (from edge_team_season and edge_skater_season, in the build script’s team-without-top-skater table): each team’s busiest 20 mph skater is taken out of its 20 mph total, and its busiest 22 mph skater out of its 22 mph total, then the rests are ranked; the rest of the Oilers had 21.3 bursts over 20 mph a game in 2025-26. Same-season link: r 0.47 across 32 teams in 2025-26 (90% range about 0.20 to 0.67, Fisher transform), so speed lines up with about 22% of the differences in points percentage (r squared, the “about a fifth” in the body); across all five seasons (160 team-seasons, speed measured against each season’s average) r = 0.29. The scatter’s line is a least-squares fit. Year-to-year test: across 126 team pairs, a team faster by one standard deviation (about 3.9 bursts a game, the “common gap” in the body) earned roughly 2.1 points of points percentage the next season (90% range about 0.8 to 3.3), or about 3.4 standings points over 82 games (range 1.4 to 5.3). Two cautions on that conversion: the effect per burst comes from all five seasons, but the gap it is multiplied by is the 2025-26 spread between teams; each season’s own spread (3.08 to 4.35 bursts a game) gives 2.6 to 3.7 standings points. Over a season, a coin-flip team’s points percentage swings by about 5.5 points, or 9.1 standings points, by chance alone.
Continues below
That coin flip ignores overtime-loss points; with them (a .500 team losing 11% of its games in overtime or a shootout, the 2021-22 to 2025-26 rate), the chance spread is about 5.2 points of points percentage, or 8.5 standings points, which is why the body says 8 or 9. Fastest eight against slowest eight, exact gaps: 3.0 to 8.7 points of points percentage. The same franchises appear again and again across team pairs, so the standard error, our measure of how sure we are, is optimistic.
| Test | Result | Verdict |
|---|---|---|
| Top speed barely moved, but 22+ mph bursts rose | 22+ bursts +13% per 60 from 2021-22, +24% from 2022-23 (2,419 to 2,753); 23+ mph skaters 89 to 124; 20+ bursts +2.3% (+8.1% from 2022-23) | Partly holds |
| The 22+ rise is just McDavid | League 22+ total without him 2,330 to 2,602 (+11.7%, totals); outside the top 25 burst-ers +15% | No |
More tests below
| Test | Result | Verdict |
|---|---|---|
| Tracking was not recalibrated | One all-ages step from 2021-22 to 2022-23 (−8.0% for the same players); later pairs within 1% | Cannot rule out for 2021-22 |
| Arrivals, not returners, drove the rise | Arrivals beat leavers in 4 of 4 pairs; since 2022-23 turnover about 54% of the 20+ rise, returners about 77% of the 22+ rise | Both |
| The fastest skater is the most frequent fast skater | Season fastest skates ranked 6th, 8th, 4th, 17th, 39th in 22+ bursts; team top speed repeats at r 0.17, team bursts at r 0.82 (126 pairs) | No: top speed is a lottery |
| Faster teams win more | r 0.29 same season (160 team-seasons); 2025-26 r 0.47 (90% range 0.20 to 0.67, n 32); +0.52 points % per burst a game next season (SE 0.19) | Small but real |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Fastest 8 against slowest 8 teams by bursts over 20 mph a game, 2021-22 | Points % 57.2% against 48.5% (gap +8.7); same-season r 0.35 | Fastest 8 ahead |
| Fastest 8 against slowest 8 teams by bursts over 20 mph a game, 2022-23 | Points % 57.4% against 54.4% (gap +3.0); same-season r 0.21 | Fastest 8 ahead |
| Fastest 8 against slowest 8 teams by bursts over 20 mph a game, 2023-24 | Points % 54.4% against 50.5% (gap +3.9); same-season r 0.20 | Fastest 8 ahead |
| Fastest 8 against slowest 8 teams by bursts over 20 mph a game, 2024-25 | Points % 58.2% against 53.7% (gap +4.5); same-season r 0.25 | Fastest 8 ahead |
Continues below
| Test | Result | Verdict |
|---|---|---|
| Fastest 8 against slowest 8 teams by bursts over 20 mph a game, 2025-26 | Points % 60.9% against 52.8% (gap +8.1); same-season r 0.47 | Fastest 8 ahead |
Frequently asked questions
Is the NHL getting faster?
At the very top, yes. In NHL EDGE tracking, 22+ mph bursts rose from 2,419 in 2021-22 to 2,753 in 2025-26 with the same ice time, and players with a 23+ mph top speed from 89 to 124. The typical rate of 20+ mph bursts changed only +2.3%, or +8.1% if you start from 2022-23.
Who is the fastest skater in the NHL?
By a single skate, Beck Malenstyn of the Buffalo Sabres, at 24.94 mph on March 12, 2026 against Washington, the fastest of the tracking era. By how often a player reaches top speed, Connor McDavid: 151 bursts over 22 mph and 681 over 20 mph in 2025-26, both league highs.
What is the most 22 mph bursts in a season?
Connor McDavid’s 151 in 2025-26, 34 more than any player since tracking began in 2021-22, according to NHL.com. The previous high was Nathan MacKinnon’s 117 in 2023-24. Among defencemen, Matthew Schaefer’s 33 in 2025-26 is the most in our five seasons of data.
How fast do NHL players skate?
The average regular forward’s top speed in 2025-26 was 22.5 mph and the average defenceman’s 22.1 mph. The fastest skate tracked that season was 24.9 mph.
Which NHL team is the fastest?
By bursts over 20 mph a game, the Colorado Avalanche in every season from 2021-22 to 2025-26 (32.2 a game in 2025-26). The Calgary Flames were slowest in 2025-26 at 14.4. By bursts over 22 mph, Edmonton led in 2025-26 with 258, 151 of them McDavid’s.
Do faster NHL teams win more?
Somewhat. The fastest eight teams beat the slowest eight on points percentage in all five seasons, by 3.0 to 8.7 points. Speed also adds a little to next season’s forecast, but the betting market mostly prices it.
More in Bodies
- Verdict: Data story
The NHL Is Getting Older Again
- Verdict: Data story
How Much Lighter Did the NHL Really Get?
- Verdict: Data story
Picks Six to Ten Got Safer. After Pick 30, Nothing Did.
- Verdict: Data story
Nobody Drafted Martin St. Louis
- Verdict: Data story
Major Junior Lost Half Its Share of the NHL Draft
- Verdict: Data story
The 1,000-Game Club Had Its Biggest Year. Getting In Is No Easier.
Sources and method
- NHL EDGE: skater and team skating speed, bursts and distance (regular season, 2021-22 on)
- NHL.com: NHL EDGE advanced stats launch (tracking deployed in all 32 arenas during the 2021-22 season; cameras and sensors)
- NHL EDGE: what’s new (EDGE 2.0 site features)
- BlueJackets.com, Jeff Svoboda: The Changing NHL: Scoring, speed the new name of the game (Aug. 5, 2019; Foligno, Kekalainen)
- NHL.com, Pete Jensen: NHL EDGE stats 2026-27 season outlook for Edmonton Oilers (Aug. 10, 2026; McDavid’s 151 and 681 bursts)
- NHL.com: NHL EDGE stats 2026-27 season outlook for New York Islanders (Aug. 18, 2026; Schaefer’s bursts, ice time and Calder Trophy)
- Russian Machine Never Breaks, Ian Oland: Beck Malenstyn breaks NHL’s all-time skating speed record in game against Capitals (March 13, 2026)
- Buffalo Sabres: Beck Malenstyn player bio (fastest sustained skating speed of the EDGE era)
- Associated Press via Fox Sports: Buffalo Sabres clinch a playoff spot to end the longest drought in NHL history (April 4, 2026)
- NHL stats (NHL.com), every season since 1917-18
Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.