With the benefit of some time for review, I discovered that my simulation code has been, among other things, calculating Venue Performance Values in what can only be described as an unexpected manner. The same issue affected both my Standard and my Alternative approaches and has been corrected for this week’s version.

(For details about the methodology, see this blog post and also refer to this one from last week that talks about an alternative approach to adding time-based variability to team ratings that I’m currently trialling)

Applying the (conventional) “traditional” methodology to 10,000 of the 50,000 home and away season simulation replicates yields the following chart showing teams’ Finals fate overall and as a function of the teams they are defeated by during the Finals or, if they win the Grand Final, the team that they defeated.

Overall, we see that the teams fall fairly naturally into seven groups:

• Geelong: just under 2-in-5 chances for the Flag and just under 3-in-5 chances to make the Grand Final. They will start strong favourites against any opposition in the Grand Final.

• Melbourne and Sydney: about 1-in-5 chances for the Flag and about 1-in-2 (Melbourne) or 2-in-5 (Sydney) chances to make the Grand Final

• Collingwood: about 1-in-10 chances for the Flag and about 1-in-4 chances to make the Grand Final

• Fremantle, Brisbane Lions, Richmond, and Western Bulldogs: about 1-in-70 to 1-in-25 chances for the Flag and 1-in-30 to 1-in-13 chances to make the Grand Final

Geelong is the only team that wins more than 50% of the Grand Finals that they make, although Sydney comes close.

WEEK OF ELIMINATION IN FINALS

In this next chart we look at teams' chances for various Finals finishes, ignoring their home and away ladder positions (ie we focus solely on the heights of the bars in the previous chart). The numbers shown inside a cell are the percentage of simulation replicates (multiplied by 100) in which the specified team went out in the specified week of the Finals.

We see here that, if we define the season in terms of the five events listed above plus "Miss the Finals", the most-likely finishes for each team are estimated to be:

• Win Grand Final: Geelong

• Lose in a Preliminary Final: Melbourne, Sydney, and Collingwood

• Lose in a Semi-Final: Fremantle

• Lose in an Elimination Final: Brisbane Lions, Richmond and Western Bulldogs

GRAND FINAL PAIRINGS 

In this final chart we look at all of the Grand Final pairings that occurred in at least one of the simulation replicates. The numbers shown inside a cell are the percentage of simulation replicates (multiplied by 100) in which the team named in the row defeated the team named in the column in the Grand Final.

We see that the most common Grand Final pairing is Geelong and Melbourne, which occurs in about 1-in-4 replicates with Geelong winning almost 70% of them. Next, we have the Melbourne/Sydney pairing, which occurred in about 1-in-6 replicates, with Sydney winning a fraction over half of them, then Geelong/Collingwood, which occurred in about 1-in-7 replicates, with Geelong winning almost 70% of them as well. Then comes the Geelong/Sydney pairing, which occurred in about 2-in-15 replicates, with Geelong winning a tick under two-thirds of them.

CHARTS FOR THE ALTERNATIVE METHOD

Applying the alternative methodology (see this post for details) to both the home and away season simulations and then to 10,000 of those for the purposes of simulating the Finals yields the following, alternative outputs.

As we’ve seen each week in the comparison, this methodology tends to improve the results for strong teams, because it introduces less variability into the ratings used for games in the more distant future. Here that benefits Geelong.

Overall, the team-by-team main differences are:

• Geelong’s Flag chances increase from about 39% to 54%

• Melbourne’s Flag chances decrease from about 22% to 18%

• Sydney’s Flag chances decrease from about 19% to 17.5%

• Collingwood’s Flag chances decrease from about 10% to 5%

• Fremantle’s Flag chances decrease from about 2.5% to 0.5%

• Brisbane Lions’ Flag chances decrease from about 1.5% to under 0.5%

• Richmond’s Flag chances decrease from about 4% to 2.5%

• Western Bulldogs’ Flag chances decrease from about 2% to 1%

For completeness’ sake, here are the Grand Final matchup numbers under this alternative methodology:

We now have:

• Geelong v Melbourne Grand Final: over 1-in-3 replicates, Geelong wins almost 80% of them (compared with 1-in-4 and 70%)

• Melbourne v Sydney Grand Final: about 1-in-6 replicates, Sydney wins just over 55% (compared with 1-in-6 and about 51%)

• Geelong v Collingwood Grand Final: a little over 1-in-8 replicates, Geelong wins just over 80% (compared with 1-in-7 and about 70%)

• Geelong v Sydney Grand Final: about 3-in-9 replicates, Geelong wins just under 75% (compared with 2-in-15 and about 66%)