The Twitter account Rugby Inside Line has been doing a mighty fine job of both keeping us informed and getting up Stephen Jones’ nose recently.
But perhaps one of the best bits of info it’s disseminated over the past few months is the entire week one statistics from the Gallagher Premiership, which painted a few telling pictures.
And for your overview benefit, the scores:
- Newcastle 21-32 Saracens
- Exeter 40-6 Leicester
- Worcester 20-21 Wasps
- Harlequins 51-23 Sale
- Gloucester 27-16 Northampton
- Bristol 17-10 Bath
So in the light of this all, we thought we’d compare these against the benchmarks we’ve been given to believe are the keys to success in rugby, as well as seeing if we could find a suite of head-to-head stats which marked down a winner.
It’s clearly slightly incomplete information as we’ve no context for the stats to be taken in – for example, Harlequins kicking the most from hand despite racking up a half-century of points rather goes against the notion that teams that pass more often win. But were Harlequins simply kicking more later to close out the game while Sale fought in vain for a way back?
Still, back to benchmarks. One of Graham Henry’s mantras to the All Blacks some time ago was that his team should be the ‘90%’ team, i.e. his teams should get 90% success rates in all their set pieces, tackles and rucks. That doesn’t sound too hard, but consider that at the weekend, not one Premiership team hit 90% of tackles and perhaps a little context does take a hold of this.
Several teams hovered just over 90% on the set piece (and the scrum has become much easier to win since Henry’s days), while not a single team dipped below 90% on own rucks.
So 90% would appear to be the bare minimum – and Neil Back, no less, was quite succinct during the reaction to the tweet about Matt O’Connor’s sacking that Leicester’s acceptable minimum tackle rate back in his day was between 90-95%.
But we’ve been presented with these truisms over the past few months in conversation, and we’re going to test them now:
- The team that tackles best, wins
- The team that dominates territory and possession wins
- Turnovers win matches
- Discipline wins matches
1) The team that tackles best, wins. Evaluated percentage-wise, this is way off the mark. Bristol, Gloucester and Wasps were all winners who failed this in terms of percentage tackles made. But in terms of sheer numbers, only Gloucester missed more tackles than the team they vanquished. In fact, that game sticks out a bit, as it is also the only game in which the winning team had to attempt more tackles than the team it beat as well. Which rather leads us on to the next assumption…
2) The team that dominates territory and possession, wins. Taking the average between the two stats of all teams, the only one that fails is, yep, you guessed it, Gloucester, who were shaded out on territory by the Saints and split possession 50-50. Otherwise, Sarries accounted for a tiny lack of possession with superior territory, while all the other winners were also winners in both categories.
3) Turnovers win matches. This doesn’t seem to ring true either, as only Exeter and Quins won more turnovers than their opponents (and they were the big winners). However, conceding turnovers (doesn’t necessarily correspond to the other team winning one as it includes knock-ons and such) does seem to have more of a negative effect, with only Wasps winning despite conceding more turnovers than their opponents.
4) Discipline wins matches. Doesn’t seem to ring true in total. The highest number of penalties conceded by a winning team was Bristol’s 14, and Bath’s kicking success rate was awful; the game could easily have gone the other way. But Saracens conceded more than Newcastle, Quins conceded more than Sale and Bristol more than Bath – a lot more, 14-6. While knowing more about the timelines of the game could give context to this – did Quins concede a slew of penalties while they were sitting back, for example, as there doesn’t seem to be too much truth to this one in isolation.
All of which means that a deeper dive – and absolutely a bigger sample size! – is needed to discover what the winning formula might be. But in the next game we’ll be looking at what we expect to be a tight one between Australia and South Africa in the Rugby Championship and seeing if we can pick apart the differences between the two teams. And we’ll also take a look at just how many possible statistical events can be measured in a rugby match.
Journey into stats conducted by Lawrence Nolan