How Hockey ChangedThe Vanishing Game
Did Garbage Goals Take Over the NHL?
No. Rebounds, scrambles and tips have been about three NHL goals in ten every season since 2017-18. The goal that’s fading starts at the blue line, and the flood of goals from the crease came mostly from how rink scorers wrote down where goals happened.
By Edgehalla ResearchPublished 11 min read
On March 8, 2018, Tyler Johnson scored on a rebound for Tampa Bay against the Rangers. The scorer at the Lightning’s rink put it 23 feet from the net. That season, Tampa Bay’s scorers logged just five goals of 258 within 10 feet.
Three weeks later at Madison Square Garden, J.T. Miller scored for Tampa Bay off another rebound, at the end of a scramble. The Garden’s scorer put it 9 feet out. Over the season, the Garden’s scorers logged about three goals in ten that close.
Those scoresheets feed a theory called kitchen-sink hockey. Goalies are too good to beat clean now, it says, so goals come from garbage: rebounds, scrambles and tips jammed in from the crease.
On this page
Rebounds, scrambles and tips have stayed about three goals in ten
No. Garbage goals did not take over.
Start with the goals from those Tampa Bay-Rangers games. Johnson’s came less than three seconds after an earlier Lightning shot, with no whistle between. We count that as a rebound. Miller’s at the Garden was a rebound too. It was also the last of three or more Tampa Bay shots inside six seconds, so it counts as a scramble. Earlier on March 8, Anthony Cirelli scored a goal the NHL marked as a tip-in. All three are garbage goals in our count.
We sorted every goal scored with a goalie in net that way, using the play-by-play: the league’s official log of every shot, miss, block and goal, typed in by the scorers at each rink. That came to more than 63,000 goals in about 11,000 regular-season games from 2017-18 to 2025-26. Empty nets, penalty shots and shootouts are out. Garbage goals were between 28% and 31% of goals in every one of the nine seasons. Our goal-by-goal records start in 2017-18, so this story can’t say whether garbage goals grew in the 2000s or 2010s.
Garbage goals held at about three in ten while point-shot goals shrank
Share of NHL goals scored with a goalie in net, all strengths, regular season
Show data
| Season | Rebound, scramble or tip | Rebounds | Tips and deflections | Point shots |
|---|---|---|---|---|
| 2017-18 | 29.9% | 17.9% | 13.4% | 7.0% |
| 2018-19 | 28.8% | 16.3% | 13.4% | 7.0% |
| 2019-20 | 31.2% | 19.3% | 13.3% | 6.6% |
| 2020-21 | 30.8% | 19.2% | 13.1% | 6.7% |
| 2021-22 | 30.7% | 18.9% | 13.2% | 6.0% |
| 2022-23 | 30.5% | 17.9% | 13.7% | 5.3% |
| 2023-24 | 30.1% | 17.2% | 13.5% | 5.3% |
| 2024-25 | 28.4% | 16.1% | 13.0% | 5.7% |
| 2025-26 | 27.9% | 16.0% | 12.3% | 5.6% |
The top line is every garbage goal. It barely moves. It went from about 30% in 2017-18 to about 28% in 2025-26, a slip too small to call a fall. There was no rise. The first 184 games of 2026-27 came in lower still, at just over a quarter of goals.
The pieces sit still too. Rebound goals ran between 16.0% and 19.3% of goals, about one in six. They peaked in the two shortened seasons, 2019-20 and 2020-21, and were lowest in the last two. That small dip is unproven. Tips ran between 12.3% and 13.7%. Goals labelled scrambles did rise, from 2.6% of goals to 3.6%. That is about 80 more a season, out of about 7,500 goals. But a scramble only gets counted if the scorer wrote down the misses and blocks before it, and that record-keeping changed. So we don’t trust this rise. It comes back later in this story.
Leave out the power plays and the answer is the same. At 5-on-5, with no penalty on either team, garbage goals went from 29.0% of goals to 27.4%.
The goal that’s fading starts at the blue line
The goal that did change comes from the other end of the zone. Ten days after that March 8 game, on March 18, 2018, defenceman Victor Hedman scored for Tampa Bay against Edmonton on a slap shot from the point, logged 56 feet from the net. We count any shot from the first 20 feet inside the blue line as a point shot. In 2017-18 point shots were 7% of goals. From 2022-23 on, they have been between 5% and 6%.
Across a full season, that is about 85 fewer point-shot goals in the league than in 2017-18. Point-shot goals went from about 0.39 a game to 0.32. Long goals in general fell the same way: goals from 40 feet or more went from 13% of goals to 10%.
Teams took fewer point shots, and each one went in as often as ever
Point shots as a share of unblocked attempts (every shot that wasn’t blocked: goals, saves and misses) and of goals, slap shots as a share of goals, and how often point shots went in; NHL regular season, goalie in net
Show data
| Season | Point shots, share of attempts | Slap shots, share of goals | Point shots, share of goals | Point-shot shooting % |
|---|---|---|---|---|
| 2017-18 | 21.7% | 11.5% | 7.0% | 2.0% |
| 2018-19 | 22.0% | 10.2% | 7.0% | 2.1% |
| 2019-20 | 23.3% | 10.7% | 6.6% | 1.9% |
| 2020-21 | 23.4% | 10.7% | 6.7% | 1.9% |
| 2021-22 | 20.7% | 10.0% | 6.0% | 2.0% |
| 2022-23 | 18.8% | 9.1% | 5.3% | 1.9% |
| 2023-24 | 19.1% | 8.8% | 5.3% | 1.9% |
| 2024-25 | 20.0% | 8.3% | 5.7% | 1.9% |
| 2025-26 | 19.0% | 8.5% | 5.6% | 2.0% |
Read the bottom line of the chart first. It is flat. About 2% of point shots go in, the same as in 2017-18. Teams just take fewer of them. In 2017-18 they got about 19 point shots a game past the shot-blockers. In 2025-26 it was about 16.
The drop shows up at 25 of 31 rinks, so it isn’t one scorer’s habit. Slap-shot goals give a second check that skips distance altogether. A scorer can log a slapper a few feet off, but it still goes down as a slap shot. Those goals fell by about a quarter, from 11.5% of goals to 8.5%.
That check still leans on the scorers, though. They also type in what kind of shot each one was, and over these years those labels shifted. Snap-shot goals went from about 15% of goals to about 29%, mostly at the expense of wrist shots (about 50% to 39%). If wrist shots can turn into snap shots on paper, the slap-shot count can drift too. Keep that in mind at the crease, next, where the label that shifted is distance.
Maybe defencemen still shoot as often, and more of their shots hit a shin pad on the way. Recorded blocked shots did jump, from about 29 a game in 2022-23 to 34 in 2023-24. Then they eased off. By 2025-26 a game had only about 2 more blocks than back in 2017-18. The timing is off too. Most of the drop in point shots that got through came in 2021-22 and 2022-23. In those two seasons a game had no more blocks than in 2017-18. So blocks can’t explain all of it.
On the scoresheet, goals crept toward the crease
That leaves the other half of the kitchen-sink theory: the crease. Goals logged within 10 feet of the net rose in most seasons, from about 19% of goals in 2017-18 to about 28% in 2025-26. They rose in 28 of 31 rinks.
Now look one step farther out, at goals logged 10 to 20 feet from the net. They fell from about 36% of goals to about 26%. Farther out, goals from 20 to 40 feet grew from about 32% to 36%. Goals from 40 feet and out, where every point shot lands, fell from about 13% to 10%.
On paper, goals slid from 10 to 20 feet out to inside 10 feet
Share of NHL goals scored with a goalie in net, by recorded distance from the net, regular season
Inside 10 ft28.1%
10 to 20 ft25.8%
20 to 40 ft36.0%
40 ft and out10.1%
Show data
| Season | Inside 10 ft | 10 to 20 ft | 20 to 40 ft | 40 ft and out |
|---|---|---|---|---|
| 2017-18 | 19.3% | 35.8% | 31.9% | 13.0% |
| 2018-19 | 19.1% | 35.8% | 32.8% | 12.3% |
| 2019-20 | 22.4% | 33.8% | 32.1% | 11.8% |
| 2020-21 | 23.2% | 32.6% | 32.9% | 11.3% |
| 2021-22 | 23.0% | 31.6% | 34.6% | 10.9% |
| 2022-23 | 25.6% | 29.7% | 34.9% | 9.9% |
| 2023-24 | 26.9% | 29.0% | 34.0% | 10.0% |
| 2024-25 | 26.1% | 27.5% | 36.1% | 10.4% |
| 2025-26 | 28.1% | 25.8% | 36.0% | 10.1% |
Compare the first two panels. They mirror each other. If players really were crashing the net more, there would be more goals in total near the net: the inside-10 panel would climb while the 10-to-20 panel held. That didn’t happen. Goals inside 10 feet gained about 9 in every 100. Goals 10 to 20 feet out lost about 10 in every 100, a touch more. Put the two together and goals within 20 feet stayed between about 54 and 56 in every 100, every season. That is the pattern you would get if the goals stayed put and the scorers started writing them down a few feet closer, moving them from the second panel into the first. To see where that happened, go rink by rink.
Tampa Bay and the Rangers used to see different games
Go back to those two games. Tampa Bay beat the Rangers 5-3 at home on March 8. The Lightning’s scorer logged none of the eight goals inside 10 feet. The closest was 16 feet out, and the Rangers’ Ryan Spooner had a rebound of his own logged at 26 feet. On March 30 at the Garden, Tampa Bay won 7-3. Nine goals went past a goalie that night, and the Garden’s scorer logged four of them inside 10 feet. Same two teams, same month.
Over the whole 2017-18 season, Tampa Bay’s scorers put under 2% of goals within 10 feet, the lowest of any rink. The Rangers’ scorers put about 31% there, the highest. The Lightning and the Rangers played in both buildings that season, and so did every other team. When two rinks host the same teams and log goals that differently, most of the difference comes from the scorers (see arena scorer bias for the same lean in hits).
Some of any rink’s number is just which games it hosted. One rink gets a month of goal-mouth scrambles, another a run of goals off the rush. In 2017-18, about 19 goals in every 100 across the league were logged inside 10 feet. Now picture every rink’s scorer using the same tape measure. Rinks would still differ, just from the games they happened to host, and a typical rink would land about 3 goals in 100 above or below the league. That is luck alone. The real rinks sat about 6 in 100 away, more than twice as far. Tampa Bay sat about 17 in 100 below.
From 2022-23, rinks stopped disagreeing about close goals, until the gap was no bigger than luck (dashed line)
How far a typical home rink sat from the league’s share of goals logged within 10 feet, in goals out of 100. Dashed line: the gap luck alone would leave if every rink’s scorer measured the same way and rinks differed only in the games they hosted
in 100
Show data
| Season | Gap between rinks | Luck alone |
|---|---|---|
| 2017-18 | 6.4 in 100 | 2.6 in 100 |
| 2018-19 | 6.0 in 100 | 2.6 in 100 |
| 2019-20 | 5.7 in 100 | 3.0 in 100 |
| 2020-21 | 6.5 in 100 | 3.4 in 100 |
| 2021-22 | 6.1 in 100 | 2.7 in 100 |
| 2022-23 | 4.0 in 100 | 2.8 in 100 |
| 2023-24 | 3.3 in 100 | 2.9 in 100 |
| 2024-25 | 2.8 in 100 | 2.9 in 100 |
| 2025-26 | 2.6 in 100 | 2.9 in 100 |
- 2022-23: Rinks start agreeing
The solid line is how far the real rinks sat from the league. The dashed line is how far luck alone would put them. The solid line drops in 2022-23 and keeps dropping until it meets the dashed one. The typical gap fell from about 6 goals in 100 to 4 that season. By 2025-26 it was no bigger than luck alone would leave. The lowest rink that season logged about 24% of goals within 10 feet. No rink was anywhere near 2%.
Tampa Bay’s scorers moved the most, and the Rangers’ barely moved
The rinks that used to log goals far from the net did the moving. Tampa Bay made the biggest jump in the league, from about 5% of goals inside 10 feet over the first three seasons (2017-18 to 2019-20) to about 25% over the last three (2023-24 to 2025-26). The Rangers’ rink barely moved: about 31% in 2017-18, about 30% in 2025-26. Those are single seasons. Pooled three seasons at a time, as in the table below, the Garden sits a little lower, at about 27% early and 28% late. Either way it hardly moved. By 2025-26 the whole league logged about 28 goals in 100 inside 10 feet, right where the Garden had been all along. Carolina went the other way. It is one of only three rinks where the share fell, from about 28% to about 24%. The table below has seven of the rinks.
| Home rink | 2017-18 to 2019-20 | 2023-24 to 2025-26 | Goals counted (first / last three seasons) |
|---|---|---|---|
| Tampa Bay | 4.8% | 24.8% | 726 / 730 |
| Boston | 8.7% | 28.3% | 652 / 681 |
| St. Louis | 12.3% | 28.6% | 640 / 681 |
| Florida | 12.2% | 28.0% | 707 / 657 |
| New Jersey | 14.7% | 25.5% | 641 / 703 |
| NY Rangers | 27.2% | 27.8% | 666 / 673 |
| Carolina | 27.9% | 23.9% | 623 / 723 |
Read down the 2017-18 to 2019-20 column and the rinks run from about 5 close goals in 100 to more than one in four. Read down the 2023-24 to 2025-26 column and they all sit between about 24% and 29%. Boston’s scorers logged about 9% early and about 28% late.
Carolina’s rink logged the highest share of close goals of all 31 rinks in the first three seasons. In the last three it logged the lowest. Line up every rink that has both stretches and the same thing shows: where a rink sat early gives no hint of where it sat late.
The rinks started coming together a season before the jump in recorded misses we covered in Shots on Goal Fell. Shooting Didn’t.. We don’t know why they started agreeing, and the NHL hasn’t said what changed. Aaron Knodell of Puck Over the Glass spotted the same thing in December 2024. The home-rink lean in shot locations was shrinking, he found, most of all in 2023-24, with early signs in 2022-23. He believes the league’s puck tracking now feeds the locations. He hasn’t confirmed it. Either way, shot distances from before 2022-23 can’t be compared with those after.
Twice the rebound chances, the same rebound goals
The scoresheet also explains the second number that seems to prove kitchen-sink hockey: rebound chances. Johnson’s goal on March 8 only counts as a rebound because the scorer wrote down the Lightning shot before it. Recorded rebound attempts roughly doubled, from about 4.7 a game in 2017-18 to about 9 in 2025-26. Rebound goals stayed at about one a game.
So the share of rebound attempts that went in fell from about one in five to about one in ten.
Recorded rebound attempts doubled while rebound goals stayed near one a game
Both teams, shots with a goalie in net, NHL regular season
a game
Show data
| Season | Rebound attempts | Rebound goals |
|---|---|---|
| 2017-18 | 4.7 a game | 1.0 a game |
| 2018-19 | 4.8 a game | 0.9 a game |
| 2019-20 | 6.2 a game | 1.1 a game |
| 2020-21 | 5.6 a game | 1.1 a game |
| 2021-22 | 6.2 a game | 1.1 a game |
| 2022-23 | 7.8 a game | 1.1 a game |
| 2023-24 | 8.9 a game | 1.0 a game |
| 2024-25 | 8.7 a game | 0.9 a game |
| 2025-26 | 9.0 a game | 0.9 a game |
Scorers changed more than locations between 2022-23 and 2024-25. Where shots were taken, what type they were, and which misses and blocks got logged all shifted. Every label in this story is built from those records. A jam play that misses the net after a save, or a second shot that hits a shin pad, only becomes a rebound attempt if the scorer writes it down. Recorded misses rose from about 24 a game in 2017-18 to about 30 in 2025-26. Recorded blocked shots jumped from about 29 a game in 2022-23 to about 34 in 2023-24.
Misses don’t line up with every jump, though. Rebound attempts jumped in 2019-20, 2022-23 and 2023-24. In 2019-20, recorded misses fell. Some of those jumps have no cause we can find. In a busy goalmouth, a scorer can easily miss a jam play that went wide. A goal stops the game, and every one goes on the sheet. That is why this story counts goals, and also why we don’t trust the small rise in scramble goals. A scramble needs three recorded shots in six seconds.
The puck tracking counts the same rebounds
Everything so far comes from the play-by-play, which the scorers type in. One record of these goals isn’t written by a scorer at all. NHL EDGE, the league’s tracking system, follows the puck and every player many times a second. Since 2023-24 we have read its clip of nearly every goal (see Anatomy of a Goal). The tracking marks a rebound when a puck moving at 20 miles an hour or more reached the goalie in the 3 seconds before the scorer shot. It puts rebounds at about 18%, 18% and 17% of goals in its three seasons, with no climb.
That flat line only helps if tracking and the scorers mean the same thing by a rebound. Goal by goal, they give the same answer, rebound or not, on about 86 goals in 100. That flatters them. In 2024-25 our count called about 16 goals in 100 rebounds. The other 84 weren’t, tracking mostly agreed, and each of those counts as a match. The harder test is the 16. Tracking agreed on about six in ten of them (63.5%). So the two records pick out somewhat different goals. They still find about as many: 18 rebounds in 100 goals by tracking that season, 16 by our count.
Tracking can test the crease too. It measures the release, the spot where the puck left the shooter’s stick, and it puts about a third of goals inside 10 feet, more than even today’s scorers do. That fits scorers in the older seasons placing close goals too far out, as Tampa Bay’s did in 2017-18. Tracking starts in 2023-24, though, so it can’t check those seasons. In the seasons it covers, tracking and the scorers put a goal on the same side of 10 feet about 85% of the time.
Tips are the third kind of garbage goal, and there tracking falls short. Its tip flag caught about half as many goals as the NHL’s tip-in label in 2024-25 (6.4% of goals against 12.9%). So for tips, this story uses the NHL’s label throughout.
Puck tracking finds the same share of rebounds every season, and more close goals than the scorers log
Share of goals scored with a goalie in net, 2023-24 to 2025-26, puck tracking against the scorers’ play-by-play
Tracking puts rebounds at 17.9%, 18.0% and 17.3% of goals in 2023-24, 2024-25 and 2025-26 and releases inside 10 feet at 33.0% to 34.4%, with no trend in three seasons.
- 2023-24
- 2024-25
- 2025-26
Look at the first two sets of bars, the rebounds. Tracking found about 18 in 100 goals in 2023-24 and about 17 in 2025-26, and our count sits a little below it each year. The last two sets are close goals, where tracking runs higher than the scorers every season.
Seven goals in ten still aren’t garbage goals
The theory said goalies got so good that goals now get jammed in from the crease. Nine seasons of goals don’t show it. Rebounds, scrambles and tips held their share. The goal that thinned out came from the other end of the zone, off a shot from the point.
- In 2025-26, about seven goals in ten had no rebound, no scramble and no tip. The other three in ten are the garbage goals, and their share hasn’t grown.
- From the point, teams got about 16 shots a game through to the net in 2025-26, down from 19. About 2 in 100 still went in, so the league scored about 85 fewer of those goals a season than in 2017-18.
- At 5-on-5, with no penalty on either team, point-shot goals went from 6.7% of goals to 5.6%, and garbage goals showed no rise, the same as at all strengths.
Tyler Johnson’s rebound on March 8, 2018 went down at 23 feet. That season, rebounds, scrambles and tips made up about three NHL goals in ten. In 2025-26 they still did.
Tampa Bay’s scorers logged five goals inside 10 feet that season, out of 258. In 2025-26 they logged more than one in four. The NHL hasn’t said what changed.
Share this article
Scoring tools
Live Edgehalla tools on the same subject.
Show the work
- Tests written down
- Not in advance (exploratory)
- Results computed
Every goal into a guarded net (shootouts, penalty shots and empty nets excluded) in 11,052 regular-season games from 2017-18 to 2025-26 (63,140 goals), from our play-by-play database, with 2026-27 to date (1,085 goals) as a holdout. Labels: rebound (same-team attempt within 3 s of a recorded attempt, no whistle), scramble (3+ same-team attempts within 6 s), deflection (NHL tip-in or deflected), point shot (released within 20 ft inside the blue line). Checks: odd- against even-numbered games, 31 home rinks pooled 2017-20 against 2023-26 (Arizona’s late window is 2023-24 only; Seattle and Utah have no early window), rink spread against chance by season, a location-free slap-shot count, and NHL tracking for 2023-24 to 2025-26. Labels in full. A rebound is an attempt by the same team within 3 seconds of an earlier recorded attempt (a save, a miss or a block) with no whistle between. A scramble is a goal at the end of three or more attempts by one team inside 6 seconds. A deflection (a “tip” in the body) is a goal the NHL records as a tip-in or deflection. A point shot is released between the blue line and 20 feet inside it, from the NHL’s recorded location. The labels are ours, built from the NHL’s play-by-play. Garbage goals (kitchen-sink goals) carry any of the three labels; they overlap, and most scramble goals are also rebounds.
Continues below
Distance is from the NHL’s recorded event location to the centre of the goal line. Gap between rinks. For each season we took every home rink’s share of goals (both teams) logged within 10 feet, and measured how far a typical rink sat from the league average (the standard deviation, in percentage points). Chance is the same measure expected if every rink had the league rate, given its own goal count (binomial). By season, gap against chance: 2017-18 6.38 vs 2.62; 2018-19 5.96 vs 2.60; 2019-20 5.74 vs 2.99; 2020-21 6.52 vs 3.42; 2021-22 6.11 vs 2.73; 2022-23 3.99 vs 2.81; 2023-24 3.27 vs 2.88; 2024-25 2.75 vs 2.90; 2025-26 2.60 vs 2.93. Pooled 2017-20, rinks ran 4.8% to 27.9% (spread 5.45, chance 1.57); pooled 2023-26, 23.9% to 31.1% (spread 1.83, chance 1.72). The body’s “where a rink sat early gives no hint of where it sat late” is a correlation across 31 rinks of r = -0.04. Rewrite figures (2026-10-27, new files story-figures.csv and example-goals.csv; every earlier output is unchanged). Tampa Bay 2017-18: 5 of 258 goals at its home rink logged under 10 ft. Fewer point-shot goals: point-shot goals a game in 2017-18 minus 2025-26, times a 1,312-game season = 84.9. Not garbage goals: 72.1% of 2025-26 goals had no rebound, scramble or deflection label; screens and traffic are not recorded, so this does not mean those goals had none.
Continues below
Example goals: every goal into a guarded net (empty nets excluded) in the three 2017-18 Tampa Bay-Rangers regular-season games, with scorer, our labels and recorded distance. Trend checks. Slopes, percentage points a season (all strengths / 5v5): kitchen sink -0.20 / -0.15; point -0.23 / -0.17; rebound -0.23 / -0.19; within 10 ft 1.13 / 1.09. Odd- and even-numbered games, pooled 2017-20 to 2023-26: kitchen sink odd 30.6% to 28.7%, even 29.3% to 28.8%; within 10 ft odd 20.8% to 27.1%, even 19.5% to 26.9%. The 2026-27 holdout rebound share so far is 14.8%, a small, early sample that isn’t cited in the body. Tracking labels (NHL EDGE, 2023-24 on, from our goal-anatomy reading of each goal’s tracking clip). A tracking rebound is a goal where, in the 3 seconds before the scorer’s shot, the puck came within 5 feet of the defending goalie while moving at 20 mph or more. Release distance is the puck’s distance from the net at release. Tracking and play-by-play are matched goal by goal on game and event id. Rink windows in the body (“the first three seasons” and “the last three”) are 2017-18 to 2019-20 and 2023-24 to 2025-26, pooled.
| Test | Result | Verdict |
|---|---|---|
| Rebound, scramble and deflection goals are a rising share | 29.9% (2017-20) to 28.8% (2023-26); 26.5% in 2026-27 to date | Does not hold |
More tests below
| Test | Result | Verdict |
|---|---|---|
| Point-shot goals are a falling share | 6.9% to 5.5%; odd games 6.6% to 5.2%, even 7.1% to 5.8%; down in 25 of 31 rinks | Holds, on scorer-recorded data |
| More goals come from within 10 feet | 20.2% to 27.0%, but within 20 ft flat and rink spread fell from 6.4 to 2.6 points | Recording change |
| Rebound goals are falling | 17.8% to 16.4%; rink changes no wider than chance (2.03 vs 2.11); tracking flat 2023-26 | Small, unproven |
Frequently asked questions
What is a garbage goal in hockey?
A goal from the mess around the net: a rebound, a scramble or a tip. J.T. Miller’s goal for Tampa Bay at Madison Square Garden on March 30, 2018 was a rebound at the end of a scramble, logged 9 feet out. By our play-by-play labels, garbage goals were 27.9% to 31.2% of NHL goals scored with a goalie in net each season from 2017-18 to 2025-26, about 3 in 10.
What share of NHL goals are rebounds?
About one in six. By our play-by-play label (a second attempt by the same team within 3 seconds, no whistle between), rebounds were 16.0% to 19.3% of goals scored with a goalie in net each season from 2017-18 to 2025-26. NHL tracking puts them at 17.3% to 18.0% from 2023-24 to 2025-26.
Are more NHL goals scored from in front of the net now?
Not clearly. Goals logged within 10 feet rose from 19.3% to 28.1%, but goals within 20 feet held at 53.6% to 56.2%, and the rise came as home-rink scorers stopped disagreeing about locations. It looks like a recording change.
Are point shots scoring less in the NHL?
Point-shot goals fell from 7.0% of goals in 2017-18 to 5.6% in 2025-26. Teams take fewer unblocked point shots (19.1 a game to 16.1); the ones they take still score about 2% of the time.
Why do NHL shot locations differ by arena?
Each home rink’s scorers record where shots and goals happen, and they have long leaned in different directions. In 2017-18 one rink logged 1.9% of goals within 10 feet and another 31.0%. From 2022-23 that spread shrank, and by 2025-26 rinks differed by no more than chance. The NHL hasn’t said what changed.
More in The Vanishing Game
- Verdict: Data story
For 19 Seasons, an NHL Hit Depended on the Rink
- Verdict: Data story
Hainsworth’s 22 Was Mostly Arithmetic
- Verdict: Data story
The Slapshot Went From One NHL Goal in Four to One in Twelve
- Verdict: Data story
The Last Goalie Who Played Every Night
- Verdict: Data story
Is the Rush Dying in the NHL?
- Verdict: Data story
NHL Fights Fell to 0.18 a Game in 2018-19, Then Climbed Past the 0.20 of Gordie Howe’s Era
Sources and method
- NHL stats (NHL.com), every season since 1917-18
- NHL API reference (community documentation of the stats endpoints)
- Puck Over the Glass, Aaron Knodell: Something strange is going on with the NHL’s play-by-play (Dec. 19, 2024)
Research and analysis, not betting advice. Past results do not guarantee future returns; bet only what you can afford to lose.