With the first two games of the season in the bag, and a third to follow on Friday night, I thought it might be a good time to check in on Munster’s Net Efficiency trends as McMillan’s structure takes hold.

As a quick refresher, this is how I’ve defined my Net Efficiency metric.

Net Efficiency & Rolling Average — Plain-English Explainer

What is Net Efficiency?

Net Efficiency tells you how much a team gains (or loses) on the scoreboard per-22 entry, after accounting for both attack and defence.

Formula (per game):

Attack efficiency (pts/att entry): how many points you score each time you reach the opposition’s 22.
Defensive efficiency (pts conceded/def entry): how many points you allow each time the opponent reaches your 22.
Net Efficiency: the per-entry scoreboard edge.
- Positive = you’re winning the per-entry battle.
- Negative = they’re more efficient than you per visit.

Quick Examples

Munster vs Scarlets (2025/26):
- Attack: 3.0 pts/entry on 11 entries
- Defence: 1.7 pts conceded/entry on 12 entries
- Net Efficiency = 3.0 − 1.7 = +1.3 (Munster win the per-entry battle comfortably)
Munster v Cardiff (2025/26):
- Attack: 2.8 pts/entry on 8 entries
- Defence: 2.5 pts conceded/entry on 8 entries
- Net Efficiency = 2.8 − 2.5 = +0.3 (tight edge per visit)

Tip: If both teams have equal entries, Net Efficiency almost predicts the margin by multiplying by entries.
E.g., +0.3 net × 8 entries ≈ +2.4 pts of “efficiency edge.”

Why Net Efficiency is useful

Separates quality from quantity. It doesn’t reward empty visits or long possession without red-zone pay-off. I think it clarifies possession stats into something tangible, and I’ll be using it to cross-reference my new LBR metrics on the Work Rate series.
Comparable across opponents. Normalising by entries makes a 5-entry game and a 15-entry game comparable.
Actionable levers. You can win by (a) improving strike rate, (b) tightening red-zone defence, or (c) changing entry volume. The metric shows which lever moved.

What is a Rolling Average?

A rolling average smooths game-to-game noise so you can see the trend rather than one-off spikes.

2-match rolling: average of this game and the previous one.
3-match rolling: average of this game and the previous two.

I’ll be using a 3-match rolling average this season, as I think it gives you finer details to work with.

Why use it?

Reduces volatility. One weird game (red card, storms, rotated team) won’t whip the line all over the place.
Reveals form. A steady climb above zero = improving per-entry outcomes; a drift below zero = trouble in red-zone attack/defence.

How to read the charts

Per-game dots/line: exact Net Efficiency for each match (spiky).
Rolling line: smoothed trajectory
- Above 0: across recent games, we’re gaining points per entry.
- Below 0: we’re losing points per entry.
Season boundary line: helps you compare last season’s trend vs the new season’s start.

Practical thresholds (rules of thumb)

Attack ≥ 3.0 pts/entry and Defence ≤ 2.0 pts/entry → you’ll win most games, even with equal entries.
Net Efficiency ±0.8 per entry is a big swing. With ~10 total entries, that’s ~8 points of “per-entry” edge.
Entry parity matters. If entries are equal, Net Efficiency decides it. If entries aren’t equal, volume can overwhelm small efficiency gaps.

Common pitfalls

Tiny sample of entries. With 3–5 entries, one bounce of a ball skews the number — that’s why I use the rolling line to judge form, rather than looking at something like a one-off try concession that loses a game.
Ignoring entry differential. A small negative net can be offset if you create many more entries (and vice versa).
Opponent context. Against tier-2 teams, you’ll often post inflated attack efficiency; use rolling averages to keep perspective.

Turning insights into something useful

If Net Efficiency is negative because…

Attack is low (≤ 2.0 pts/entry): improve first-phase strike, maul threat, and set-piece starters; reduce wasteful phases before the 22.
Defence is high (≥ 3.0 conceded/entry): fix goal-line spacing/folds, slow opposition ball in the 22, prioritise exits that avoid repeat entries. That last point was a killer for Munster last season.
Entry counts are low: kick/territory plan to gain more starts in the opposition half; target +2 entries as a benchmark.

Net Efficiency is your per-visit scoreboard edge; the rolling average shows whether that edge is getting better or worse over time — so you can tell if performance is a real trend or just a one-off spike.

With that out of the way, here’s how we’re looking relative to the tail end of last season, with the last three games of 2024/25 rolled in with the start of 2025/26.

On the whole, it’s looking very positive.

Here’s a tight, numbers-first read on Munster’s first two games of 2025/26, set against last season’s baseline and the end-of-season trend you had.

What happened (per-entry focus)

Game	Att entries	Pts/att entry	Def entries	Pts conceded / def entry	Net eff.
@ Scarlets (34–21)	11	3.0	12	1.7	+1.3
v Cardiff (23–20)	8	2.8	8	2.5	+0.3

Model vs Scoreboard:

Scarlets: 11×3.0 − 12×1.7 = +12.6 expected → actual +13.
Cardiff: 8×2.8 − 8×2.5 = +2.4 expected → actual +3.
Our per-entry model is tracking the outcomes almost perfectly.

Context vs last season

Metric	2024/25 avg	Last 5 of 24/25	2025/26 (2 gms)
Attacking entries	8.56	7.2	9.5
Pts / attacking entry	2.88	3.38	2.90
Defensive entries conceded	7.56	8.2	10.0
Pts conceded / defensive entry	2.98	2.66	2.10
Net efficiency	−0.09	+0.72	+0.80

Key Reads

Defence is the story. Concessions per opposition entry have dropped from 2.98 → 2.10. That’s a huge swing and the main reason Net Efficiency is +0.80 despite conceding more entries (10.0 vs 7.56 last year).
Attack is steady, not spiky. 2.9 pts/entry is almost bang on the 24/25 average (2.88) and a little below the late-season hot run (3.38) that got us into the playoffs. We won Scarlets/Cardiff because the goal-line defence improved, not because the attack suddenly caught fire.
Template wins without volume.
- Scarlets: we lost the entry count (11–12) but won big via 1.7 conceded/entry.
- Cardiff: entry parity (8–8) → the small +0.3 per-entry edge decided it.

What’s Working

Red-zone defence: 2-game average 2.10 conceded/entry. That’s the best stretch compared with any 2-game window at the end of last season (last-5 avg 2.66).
Attack conversion is resilient: ~3.0 pts/entry in both games with different profiles (volume win vs Scarlets, parity vs Cardiff).
Per-entry model reliability: “Expected margin” from entries and efficiencies is within ~0.6–1.0 points of the real results across both games — this is excellent when it comes to week-to-week planning.

Danger Points

Entry volume against is high: 10.0 opp entries/game so far vs 7.56 last season. If defensive concessions regress toward 2.8–3.0, margins will evaporate.
Target: bring opposition entries down toward ≤ 8 while keeping concessions ≤ 2.2.
Attack ceiling not yet tested: At 2.9 pts/entry, we’re winning, but the late-season benchmark (3.3–3.5) is what separates comfortable wins from coin-flips when we have entry parity.

Benchmarks for the next 3 games

Maintain: Pts conceded/entry ≤ 2.2
Lift: Pts/att entry ≥ 3.1
Squeeze: Opp. entries ≤ 8 (from 10.0)
Resulting net goal: +0.9 per entry (gives ~+7–9 pts edge in a typical 8–10 entry match)

Two rounds in, our improvement is defence-led: we’re allowing far fewer points per opposition visit to the 22, which is why we won big at Scarlets despite losing the entry count and managed to edge a tight, entry-parity game vs Cardiff.

If we can keep the concessions down and trim opposition entries, the early +0.80 NE trend is sustainable.

Net Effiency #1 – 2025/26