Who will win the 2019 NRL Premiership?
At this time of year, is there anything else you want to know more than the answer to this question?
For our crystal ball, we turn to Monte Carlo simulations. These simulations work on the principle that if we know the inputs to a complex system and how they relate to each other, then we can test the outcomes of that system using random numbers to simulate different situations.
At its most basic, just imagine if you simulated the outcome of football matches by rolling dice. Numbers one and two might represent a win for the Gold Coast and numbers three through six might be a win for Wests. If you repeat that a couple of thousand times, not only will you be extremely bored but the Gold Coast will “win” about 33% of the time and Wests 66%.
Now take the same approach for the nine finals games, with the winner advancing per the NRL’s system, but instead of using dice, you generate a random number between zero and one and calculate the win probability using Archimedes (form) Elo ratings. Then repeat it 5,000 times over. The number of times that the Storm or Roosters or Broncos or Eels “win” the premiership across your simulations should give you some insight into the probability of that happening in real life. I call this the Finals Stocky and I present its findings.
I won’t lie. It’s probably what you’re expecting.
It’s strange to be in a situation where there’s a dual golden age of two clubs that are dominant for such an extended period. Manly and Melbourne of the late 2000s/early 2010s might be the most recent example but we all know what the go was there. Before that? Maybe Canterbury and Parramatta in the mid-1980s or Easts and Manly in the mid-1970s.
If being able to bear witness to these two historical titans do battle is not appealing, then the prospect of another Roosters versus Storm grand final may not fill you with much excitement. Fortunately, we can break down the outputs of the Stocky to see what the odds are of given grand final match up is. Please consult the following table.
Or, if you prefer, in terms of previous grand finals being replayed:
- 2018 is one in 4
- 2009 is one in 13
- 2013 is one in 33
- 2016 and 2006 is one in 50
- 2007 and 2008 is one in 125
- 2000 is one in 1000
That means there’s a 75% chance the 2019 grand final won’t be a replay of 2018 but there’s a very high chance the finale will feature the Storm or the Roosters.
Perhaps more interesting is who would win a given matchup. Please then consult this table.
Go along the top row to find who you think will be (or want to be) the premiership winner and then go down the columns to find out the odds of them winning a given grand final match-up.
Of course, neither the Sharks and Sea Eagles nor the Eels and Broncos will feature in a grand final together, as they play the first round of elimination games and two of those four will not see week two.
The Storm are pretty heavily fancied against any you care to name, except the Sharks, which may be a fair reflection of reality given that Cronulla have beaten Melbourne four times in three years. See also, the 2016 grand final.
First week’s tips
As was written in Moneyball:
“My job is to get us to the play-offs. What happens after that is fucking luck.”
I personally don’t like drawing conclusions about what might happen in the post-season based on the small sample size provided. Fortunately, we can at least rely on regular season data for the first week of fixtures. Doing tips for the qualifiers also gives me an opportunity to use the brand new Poseidon simulator (done along the same lines as the Finals Stocky but using Poseidon ratings to simulate individual matches) to help set a line. Poseidon had a relatively good run in not particular great year for the Jury (2013 to 2018 average 68%, this year 65%).
Roosters v Rabbitohs
- Archimedes (form Elo) likes Roosters (67% win probability)
- Poseidon (offence/defence ratings) likes Roosters (85% by 14.6 points)
Souths have beaten the Roosters twice this year, both at ANZ and at the SCG. In fact, Wayne Bennett hasn’t lost to Trent Robinson since the Broncos lost their first week final against the Roosters in 2017 (overall record is 9-7 in Bennett’s favour). While the numbers are unequivocal that the Roosters are ascendant and I personally would tip the Roosters, Souths are well in with a shout.
Storm v Raiders
- Archimedes likes Storm (68% win probability)
- Poseidon likes Raiders (52% by 0.1 points)
It is very difficult to separate these two at a Poseidon level but there is a slight favour to the Raiders. While Melbourne have scored 55 tries at home this season, second best in the league (league average is 43), Canberra has only given up 31 tries while playing away. Let’s also not forget that the Raiders went to Melbourne on August 17 and won, a feat that happens slightly more often than you think with three of 2017’s and 2019’s four Storm losses in Melbourne and four of six in 2018. A Raiders win is in play, if leaning unlikely. I’d go Melbourne.
Sea Eagles v Sharks
- Archimedes likes Sharks (52% win probability)
- Poseidon likes Sharks (53% by -0.4 points)
Poseidon has thrown up a strange one. Manly have a higher expected score by virtue of their superior conversion kicking (74% at home to the Sharks’ 66% away) but a lower winning probability than the Sharks because the latter is more likely to score tries. The Archimedes result is also a little strange: Cronulla are the third best in the league right now. Whether that’s because they’re legit contenders or because they belted a Tigers team that had built up a rating off a soft draw, I’ll leave to you to decide. Manly’s injuries are going to make it tough, so the Sharks might be the go here.
Eels v Broncos
- Archimedes likes Eels (67% win probability)
- Poseidon likes Eels (87% by 16.9 points)
Poseidon estimates a Parramatta win by 13 points or more as better than even money. The Broncos have played atrociously all year, but particularly poorly in Sydney. This was on display for all to see against the Bulldogs. If the Eels can maintain some sort of organisation, Brisbane will be easy picking.