Hits don't measure defense in hockey
The NHL broadcast standard for praising a defenseman includes the word "physical" and a hit count from the box score. A blue-liner with five hits in a game has played a strong physical game. A blue-liner with twelve has played a brutal one. Hits get cited as evidence of defensive intensity, of commitment, of the way the team is going to wear opponents down by playoff time. The hit total is a real number. It means something. What it means is mostly the opposite of what the broadcast convention says it means. You can only hit a player who has the puck. Players who have the puck a lot belong to teams that have the puck a lot. The team racking up the hits is, on average, the team that is being outplayed.
The possession-deficit signature
Across the last fifteen NHL seasons, the team-level correlation between hits per 60 minutes and shot attempts against per 60 minutes is positive and significant. The teams with the most hits are also the teams getting the most shots directed at them. This is structural. The body checks that register as hits in the official scoring are almost all defensive-zone or neutral-zone events where the puck-carrier was an opponent. To be in position to deliver them, you must have first ceded the puck. Every hit is, in some sense, a record of a possession your team did not have.
The best teams in the league tend to be middle-of-the-pack or below in hit totals. Colorado in their Cup-winning seasons, Vegas in 2023, Florida in 2024 — all sat roughly 20th to 25th in hits per 60. The Cup finals consistently feature a team in the bottom half of the league in hits. The teams in the top five for hits are usually the teams chasing the game and chasing forecheck contact because they cannot generate sustained zone time.
The scorer-bias problem
Hits are also one of the most subjectively scored events in the NHL. There is no automated tracking of body checks the way there is for shots, and the home-rink scorers vary widely in what they count. The same contact that gets logged as a hit in Anaheim might not register in Tampa. Year after year, the league-leading teams for hits per game are disproportionately concentrated in a handful of buildings whose scorers have a low bar for what counts.
The gap between the most generous and the least generous rink scorers can be on the order of 30% in hit-counting rates. Public hockey analysts have been documenting this for over a decade. Front offices have access to internal tracking that bypasses the public hit column entirely. The broadcast graphic is a public artifact of a counting process whose precision the league itself does not vouch for.
Hits as effort, not skill
There is a charitable read of the hit total that survives once you concede the possession story. A player who is chasing the puck, racing back on defense, and finishing checks is at minimum trying hard. Hits can be a proxy for effort, and effort is not nothing. A coach watching a third-line winger rack up six hits at the next bench will notice. The player will get more shifts.
Effort is not the same as effectiveness. The fourth-line winger who delivers six hits in eight shifts and finishes minus-one with no zone exits has worked hard for a losing outcome. The top-pair defenseman who has zero hits because he gapped up and forced two early turnovers had a much better defensive game. The hit column rewards the first player and is blind to the second. The coach watching the full sequences will weight them differently. The viewer relying on the broadcast graphic will see one number for the worker and no number for the gap-up.
What predicts run prevention
At the team level, the actual predictors of goals against per 60 are shot attempts against, scoring chances against, high-danger chances against, and shot quality against — almost all of which are zone-time and exit efficiency stories. Teams that prevent goals do so by not spending time in their own end. The way you do not spend time in your own end is to exit cleanly with possession, which is not a hitting skill. It is a passing and skating skill that the hit column actively misrepresents.
At the player level, micro-stats like controlled zone entries allowed, controlled exits made, and defensive-zone puck recoveries correlate strongly with on-ice goals against rates. Hit totals correlate weakly to negatively with the same outcomes once you control for ice time and usage. Coaches who deploy a defenseman for high-leverage defensive shifts are increasingly looking at the entry and exit numbers and not at the hit count.
The playoff myth
The most durable version of the hits-equal-defense story is the playoff variant: that physical teams who hit more wear down skilled teams over a seven-game series. This is the explanation that gets offered every spring when a bottom-half-of-the-league hits team upsets a Presidents' Trophy contender. The data does not support it. The playoff hits leaderboard is largely the same as the regular-season hits leaderboard, with the same teams racking up the same totals for the same possession-deficit reasons.
Playoff series outcomes correlate with shot share, expected goal share, and goaltending. They do not correlate with hit differential once you control for the other variables. The teams that win in May and June are usually not the teams that lead their series in hits. The "wore them down" story survives because it is satisfying. It survives despite the scoreboard mostly disagreeing with it.
The honest read
A team's hit total is best read as a signal about possession share, not about defensive intensity. A high-hits team is usually a team that did not have the puck, and the modern playbook for winning hockey is built around having the puck. A player's hit total is best read as a signal about role and effort, weighted heavily by the home rink's counting bias. Neither version of the number tells you much about whether the defense is good. The defensive stats that do live in the entry, exit, and chance columns. The broadcast rarely shows them. They are the ones the coaches actually watch.