If you’re anything like me, you’ve spent many hours – usually on a Thursday or Friday evening – pouring over information when trying to decide who to play or who to captain for an upcoming round.
There’s many things that usually factor into these decisions. It’s been raining in Auckland, should I move the VC off Shaun Johnson? Should I play everyone I have who is facing the Knights or Titans? Jordan Rapana has the Tigers this week up against their defensively challenged outside backs, should I straight captain him? Cameron Smith is coming off five days rest, just how many unicorn points can he get?
Whilst you shouldn’t be looking at just one variable for team and captain selections, a good place to start is how likely they are to pick up a win.
Looking at Supercoach scoring from 2015-2017, playing in a winning side is worth 7.6 points per player. The average Supercoach player scores 49.1 in wins and just 41.5 in losses. This difference has been increasing as well, from 7.0 in 2015, to 7.8 in 2016 and 8.0 last season.
It becomes even more important if you look at just the top 40 total point scorers over that three-year period. The difference between a win and a loss jumps up to 11.9, and with 24 of the top 40 playing having a 50% record or better during that time frame it becomes an even more important factor.
When analysing the data, seeing a trend that attacking players have a significant difference in wins shouldn’t come as a massive surprise. They rely on attacking stats like tries, line breaks or assists to rack up those high scores, and these are typically scarcer in losses.
However, it’s the size of gap between them for some players that may raise a few eyebrows. Many them have dramatic differences based on results, while others are relatively unaffected, with a number of popular choices averaging fewer than 50 when their side loses.
Below are the top 40 scorers from the past three seasons, ranked by total points and split by averages and PPM in wins and losses. Anyone who is not playing NRL this year (e.g. Semi Radrada) has been removed. Everyone inside the top 40 has played at least 37 games over this period providing a sufficient sample size, with the lone exception being Billy Slater (26 games).
|Player||Win Avg||Loss Avg||Difference||Win PPM||Loss PPM||Difference|
|Jack De Belin||64.32||60.32||4.00||1.03||0.92||0.10|
From above table you can see that Johnathan Thurston averages 82.5 in 35 Cowboys wins over the last three years with a healthy 1.04 points per minute. Yet in in 14 losses that drops to 47.6 with a paltry 0.59PPM. This gap of 0.45PPM is the largest inside the top 40, just ahead of Gareth Widdop with a 0.43PPM variance (0.98PPM in wins, 0.55PPM in losses).
Whilst Thurston has the largest PPM disparity, Widdop has the widest difference in average. The Dragons half knocks out 78.5 in wins, but drops to 43.1 in losses, a gap of 35.5 points. His average in defeat is the lowest of the top 40 players and given the average Supercoach player scores 45.3 points per game, Widdop becomes a below average selection in losses.
For those who like to live on the edge, Shaun Johnson averages 89.6 when the Warriors are victorious but just 56.0 in losses, which is the highest average from a winning side in the top 40. As for his score in losses, given how dire the Warriors have been in defeat over the past few years that’s not as terrible as I would have expected.
New Roosters fullback James Tedesco could be in line for some huge scores this season judging from this data. The former Tiger posted an average of 60.80 in losses over the past three seasons, but 88.9 in wins for a difference of 28.1. The king of tackle busts averages 7.1 in wins but only 6.1 in losses, as well as seeing an increase in errors and 10% fewer run meters. When you consider his record was 22-40 over this period, joining a loaded Roosters side could result in him being a must have at a stacked fullback position.
Anthony Milford shares a similar difference to Tedesco, scoring 27.3 higher in Broncos wins, not only suffering from fewer attacking stats in losses, but also a decrease in run meters by a 33%, fewer kicks and an increase in errors. He is also one of just four inside the top 40 to average fewer than 50 in a loss with an average of just 48.1 points per game.
Jordan Rapana is another from the rocks and diamonds club but aggregating his past three seasons masks some of his inconsistency. The Raiders winger averages 68.4 in wins and 52.4 in losses over this time yet averaged 13.7 more in losses in 2016. That is in stark contrast to his incredible rollercoaster ride in 2017, averaging an eye watering 93.7 in wins but only 53.8 in losses.
As expected as it was that attacking players scores fluctuate the most based on result, it is also equally unsurprising that the scoring from strong base stats players are less likely to change based on results.
Simon Mannering and Jason Taumalolo posted virtually identical PPM regardless of the result. They both actually average more in losses, +1.8 for Mannering and +0.7 for Taumalolo. The composition of their points is different though, with more tackling and fewer hit ups and run meters in addition to fewer attacking stats.
In addition to the above, a few more notable players score more in defeat. Paul Vaughan (+1.7) and Jesse Bromwich (+1.9) all score slightly more in losses, as do a few of the mid-range options, most notably Corey Haraweira-Naira (+6.0), Ethan Lowe (+4.3), Manu Ma’u (+4.3) and Dylan Napa (+5.2).
Billy Slater averages 1.0 more in Storm losses but given the scarcity of that happening– Slater has played in just 4 over the past three years in his 26 games – it’s too small of a sample size to take seriously.
There’s some excellent set & forget candidates amongst the forwards for the risk adverse. Andrew Fifita (+0.6), Trent Merrin (+1.3), Jack De Belin (+4.0), Marin Taupau (+3.8), Tohu Harris (+4.3), Boyd Cordner (+3.2), Jake Friend (+4.9), Jake Trbojevic (+6.2) and Matt Gillett (+2.0) all offer 60+ averages with excellent consistency irrespective of results. Sam Burgess may average 11.1 more in wins, but his 64.8 average in losses ranks only behind Gallen, Fifita and Mannering.
For outside backs, Cameron Munster is a relatively safe bet with only an 11ppg difference between a win and a loss, as is Daly Cherry-Evans with a 10.0 variance.
Whilst it might be stating obvious, the conclusion here is that that winning games certainly matters for players who rely on attacking stats to boost their scores. It matters less so for forwards who rely on base stats, and in some cases, it makes virtually no difference.
As mentioned at the start of this article, no one single variable should be the basis for a Supercoach decision. But If you’re in a situation where you’re looking at a coin flip for a captain decision or filling your last reserve slot, this data suggests going with players from a team that is favoured to win.