What metres per minute measures
Metres per minute is your total match distance divided by your minutes on the pitch, and that one division is the whole point. Total distance rewards whoever stays out there longest, so a starter who covers 6,000 metres in 80 minutes looks busier than a replacement who covers 2,000 metres in 25, but divide both by minutes played and the replacement was moving at 80 metres per minute against the starter's 75. Set a minimum-minutes rule before you compare anyone, because a ten-minute cameo at full tilt says nothing about an eighty-minute engine. The New Zealand cohort behind the numbers below only counted players who stayed on for at least an hour.
What real matches produce
Four squads from the same New Zealand province wore 10 Hz GPS units for a season each: an amateur club side, two semi-professional provincial sides and a Super Rugby side, which produced just over six hundred match files from players with an hour or more on the pitch. Whole-squad averages ran from 68 to 76 metres per minute, forwards sat between 62 and 71 and backs between 73 and 84 depending on the competition. The spread by position group across those four competitions looks like this.
| Position group | Match average (m/min) | Running at 16 km/h or faster (m/min) |
|---|---|---|
| Front row (props and hooker) | 59 to 66 | 3 to 5 |
| Second row (locks) | 60 to 72 | 6 to 9 |
| Back row (flankers and No. 8) | 63 to 73 | 8 to 11 |
| Half back (9) | 76 to 94 | 14 to 21 |
| Inside backs (10, 12, 13) | 69 to 77 | 12 to 14 |
| Outside backs (wings and fullback) | 73 to 86 | 11 to 16 |
Each range spans the four competitions' averages for that position group. The ordering is the useful part: the half back tops the running charts, the tight five sit at the bottom, and the high-intensity column separates the groups more sharply than the totals do. Fifty-one GPS studies of rugby union pooled together show the same shape, with backs covering more total, relative and high-speed distance while forwards bank more collisions and impacts instead.
The whole-match average hides the real pace
Dividing by all eighty minutes mixes the rugby with the stoppages, and the stoppages are slow. When an international team's GPS data was split by ball-in-play time, every ball-in-play measure came out significantly higher than the whole-match figure, and a club-season study that isolated the single longest ball-in-play passage of each match measured it at 117 metres per minute against a previously reported whole-game average of 68.
| Measure | How it is calculated | What it tells you |
|---|---|---|
| Whole-match m/min | Total distance divided by all minutes played, stoppages included | A fair comparison between players and across a season, as long as the minutes rule and the equipment stay the same |
| Ball-in-play m/min | Distance divided by the time the ball was actually live | The real pace of the game: one international side's forwards ran at roughly 105 m/min and its backs at 110 to 115 in ball-in-play passages of thirty seconds or more |
High-speed metres per minute
Total m/min counts walking the same as sprinting, so most GPS setups also report distance above a speed threshold per minute. There is no single agreed threshold: the New Zealand cohort used 16 km/h and up, other studies draw the line elsewhere, and that inconsistency is exactly why you pick one threshold and keep it for the whole season. There is a positional catch too. When one professional squad's high-speed running was recalculated against each player's own top speed instead of one squad-wide line, the fixed threshold had been undercounting the forwards' high-speed work and overcounting the backs', because genuinely hard running for a slower player starts below the squad-wide line. So compare a prop's high-speed numbers with other props at most, and ideally with his own history.
What m/min says about fatigue
Work rate falls as a match wears on. One professional squad's matches showed distance per minute dropping by about a tenth from the first half to the second for backs and forwards alike, but the two groups did not fade the same way: the forwards lost high-intensity running, sprints and accelerations across the halves while the backs held theirs, and the collision and static load the forwards carry is the likely reason. So a second-half dip in m/min on its own is normal. The sharper monitoring question is whether a player's high-speed output is holding up, and whether this week's number sits inside or outside that player's own usual spread across recent matches.
Training rarely matches the match
A sub-elite Italian squad wore 18 Hz tracking units through a full season of weekly training and matches, and the forwards' and backs' training did not reproduce their match running. Forwards trained at about 53 metres per minute but played at about 62, backs got close on volume while falling short on top speed, high-speed running and hard decelerations, and only the scrum-halves matched their match profile, though there were just four of them in the squad. So if you compare training m/min to match m/min at your club and find a gap, you are in normal company, and the finding cuts one way: games-based training alone left every position short of match running intensity, which is an argument for adding targeted running work rather than assuming the gap closes itself.
What the metric cannot see
- Contact and static work are invisible. Scrums, mauls, tackles and rucks produce huge effort and almost no distance, and forwards rack up far more collisions than backs, so a low m/min from a tight forward is not a low work rate.
- Context moves the number. Opposition, score, tactics, weather and pitch all change how much running a match asks for, so read trends across several matches rather than judging one game.
- The threshold you choose changes every high-speed value in the squad, so record which one you use and never compare numbers made with different thresholds.
- The whole-match average understates the game's real intensity, so when you design drills that should feel like rugby, size them on in-play or peak-period rates rather than the eighty-minute average.