Bomber's Theory of Grand Finalists

[Bomber]

1.0 - The Theory
Teams competiting in the NRL Grand Final of any given season would have satisfied at least one of two requirements:

• They would have won at least one game of the previous season's finals series, and/or
• They would have been ranked as one of the top five defenses of the previous season in terms of tries conceded

As will be demonstrated below, in six out of the previous seven seasons, the two grand finalists would have satisfied at least one of these requirements.



2.0 - The Evidence

2.1 - RULE ONE - Teams that have won one game (minimum) in finals series of that year.

Team/s that qualified for the grand final in the season following the one listed are displayed in red bold

​

• 1998 - Brisbane, Canterbury, Parramatta, Roosters, Melbourne, Canberra
• 1999 - Canterbury, Parramatta, St George Illawarra, Cronulla, Melbourne
• 2000 - Canberra, Newcastle, Parramatta, Brisbane, Sydney Roosters
• 2001 - Cronulla, St George Illawarra, Newcastle, Parramatta, Brisbane
• 2002 - Roosters, St George Illawarra, Brisbane, Warriors, Cronulla
• 2003 - Melbourne, Warriors, Roosters, Penrith, Canterbury
• 2004 - Penrith, North Queensland, Melbourne, Roosters, Canterbury
• 2005 - Wests Tigers, Melbourne, St George Illawarra, Parramatta, North Queensland

2.2 - RULE TWO - Teams listed in Top 5 Defenses in terms of tries conceded

Team/s that qualified for the grand final in the season following the one listed are displayed in red bold. Only the top five teams are shown regardless of ties

​

• 1998 - Brisbane (50), Parramatta (59), Melbourne (61), Newcastle (65), Roosters (65)
• 1999 - Parramatta (49), Cronulla (59), Brisbane (60), Roosters (64), Melbourne (66)
• 2000 - Brisbane (68), Parramatta (79), Cronulla (79), Canberra (82), Newcastle (91)
• 2001 - Parramatta (67), Cronulla (89), Brisbane (89), Canterbury (94), St George Illawarra (94)
• 2002 - Roosters (69), Bulldogs (72), Parramatta (74), Brisbane (76), Warriors (87)
• 2003 - Canterbury (73), Roosters (74), Canberra (80), Brisbane (80), Melbourne (85)
• 2004 - Roosters (64), St George Illawarra (72), Canterbury (85), North Queensland (91), Wests Tigers (91)
• 2005 - Melbourne (80), Parramatta (80), Roosters (83), Brisbane (85), St George Illawarra (91)

3.0 - Discussion of Results
The system accurately 'chose' a grand finalist five times out of seven under Rule One (winning a final in the previous season), and six times out of seven under Rule Two (top five defense in previous season). Collectively, the system accurately chose at least one grand finalist six times out of seven, and both grand finalists four times out of seven.

However, the system failed to predict either of the 2002 Grand Finalists based on data from the 2001 season. That season, both the Warriors and Roosters exited the finals after failing to win a game. In addition, the Roosters' defense conceded 101 tries, making them the sixth-ranked outfit in the premiership.

The system merely reduces the number of possible grand finalists from fifteen to approximately eight, and is therefore potentially no more meaningful than a random selection of eight teams.

The system failed to account for the rise of the Roosters in 2002 or the Panthers in 2003. Both teams won the premiership that season but failed to score in the previous season's data.

Five premiership winning teams featured in the Top 5 defenses of the previous season, as opposed to just three premiership winning teams having a win in the previous season's finals series.

The system only takes into account data from when the NRL began and therefore can only be based on limited data.

These teams satisfied both of the criterion and made the grand final in the following season - Melbourne (1999), Newcastle (2001), Parramatta (2001), Roosters (2003), Canterbury (2004) and North Queensland (2005). Of these teams, only Parramatta and the Cowboys lost their grand final. In Parramatta's case, they were playing another team that satisfied both of the criterion. The 2001 Grand Final is the only instance where this has occured.

4.0 - Conclusion
Not without its flaws, but it will be interesting to see how it works out in 2006. If the system is to be believed, at least one of these seven teams will certainly make the grand final (with 86% accuracy!):

Melbourne, Parramatta, Roosters, Brisbane, St George Illawarra, Wests Tigers, North Queensland

 

[Southernsaint]

Ah, Bomber, the quintesential Uni student!!

 

[Bomber]

lol, I haven't lost the knack, despite graduation!

 

[Jono078]

Good read. I'm hoping the Knights can prove the theory wrong though .

 

[innsaneink]

Very well done, if I can follow something of that detail right thru to the end @ 12.30 am on a Sunday morning, youre doing a good job of communicating your point. Too often posts like this lose me, and I couldnt be bothered going back over it.

 

[brendothejet]

14% Isnt Much Of A Margain For Newcastle

 

[Balmain_Boy]

The system failed to account for the rise of the Roosters in 2002, the Panthers in 2003 or the Wests Tigers in 2005. All three teams won the premiership that season but failed to score in the previous season's data.

I don't get it/ You have us in the top 5 defenses in 2004 yet say our rise wasn't predicted??

 

[Bomber]

Typing error, my mistake

 

[eloquentEEL]

If the system is to be believed, at least one of these seven teams will certainly make the grand final (with 86% accuracy!):

Melbourne, Parramatta, Roosters, Brisbane, St George Illawarra, Wests Tigers, North Queensland

If you consider all teams being identically weighted with a 1 out of 15 chance of making the GF, then the statistical chance of at least one from any list of seven teams making the next GF is 73% (see working below) so at 86% chance (based on historical evidence), your rules may not be as strong/impressive as first thought... but they are valid nonetheless. Perhaps it could be a challenge to other forummers to see if they can find a stronger predictor.

Working:

Both GF teams OR 1st team only OR 2nd team only
= (7/15 * 6/14) + (7/15 * 8/14) + (8/15 * 7/14)

= 0.2 + 0.267 + 0.267

= 0.73

 

[White Poiner]

That is quite interesting.
Just like John Tutor. :roll:

 

[TiLLs]

"Well check out the brain on Brad!"