Ask most coaches how you score and you will hear some version of “get into the 22 and build pressure”. The 22-metre entry has become the metric everyone trusts, the assumption being that field position is most of the battle. The data, and a closer look at how tries are actually scored, says it is a long way from the whole story.
Get into the 22, and 43% of the time nothing happens
In the Framesports data, around 57% of 22 entries end in a try. Put the other way round, you fight your way into the opposition 22 and come away with nothing 43% of the time.
That is the number that should give the territory-first crowd pause. Getting there is necessary, you cannot score from your own half without eventually crossing their line, but it is nowhere near sufficient. Nearly half of all that field position is wasted, which means what you do on entry matters far more than the entry itself.
Not all entries are equal
Here is the deeper problem with the metric. A “22 entry” counts a grinding twelve-phase slog up the middle exactly the same as a 60-metre line break that happens to pass through the 22 on its way to the posts. One arrives with a set, narrow defence already waiting. The other arrives against scramble. They are not the same event, and treating them as one tick on a stat sheet flatters the slow, territorial route.
This is why serious analysis has moved towards expected-points thinking and grouping phases into momentum blocks rather than counting entries in isolation. The value is in how and where you arrive, not the arrival.
Our own data backs this up in a way that should make every “build the phases” coach pause: across matches, the more rucks a team grinds through, the fewer tries it tends to score. Phase count is not pressure. Beyond a few quick phases it is usually a defence getting its line set while you run out of ideas. The tries come before the grind starts, or not at all.
Tries come from breaking the line, and from chaos
Look at where tries genuinely originate and the picture flips towards attack, not territory.
- The line break is the most try-productive moment in rugby. Analysis of elite rugby has found a clean line break produces a try within two phases roughly 39% of the time, and most of the breaks that do not score still keep you on the front foot. Beating a defender is worth more than another ten metres of territory.
- Transition is lethal. The most dangerous ball in the game is turnover ball, because you are attacking a defence that is unset, narrow and scrambling rather than a set line. It is no accident that the best counter-attacking sides score a stack of their tries straight off turnovers.
- Possession and territory barely predict winning. Test sides have beaten good opposition with under 40% of the ball. If hoarding territory was the answer, that could not happen. It keeps happening because tries come from cutting teams open, and you can do that from anywhere.
What this means for your team
The takeaway is not “ignore the 22”, it is “stop treating field position as the goal”. If 43% of your entries die, the work is in your strike plays and your finishing, not in kicking for three more metres. And if line breaks and transition are where tries actually live, the brave plan is usually the right one: attack space when you see it, play heads-up off turnovers, and back yourself to score from your own half rather than kicking it away and starting the territory grind again. That is more attacking rugby and, the evidence says, smarter rugby.
How we worked this out
The 57% entry-to-try figure comes from the Framesports data, millions of events across thousands of matches at every level, built on the most accurate data collection system in rugby, so the number is reliable. It is measured on the analysed team’s own 22 entries and tries; we do not infer opposition data or match results. The line-break, transition and possession points draw on published analysis of elite rugby, noted as such rather than claimed as our own. For the wider series, start with rugby statistics, and on building this into a plan, smarter gameplans with data.
