Matter of Stats

View Original

2024 : Simulating the Final Ladder After Round 21

This year’s post Round 21 simulations suggest that the teams can be roughly grouped as below:

(For anyone who is curious about the precise methodology I’ve employed, see this blog post over on the General Probability and Statistics journal)

  1. Sydney and Brisbane Lions: virtually certain of being finalists; 90-95% chances of Top 4; 40-50% chances of being Minor Premier

  2. Port Adelaide, Geelong, and Western Bulldogs: 95% chances of being finalists; 40-60% chances of Top 4; 1-5% chances of being Minor Premier

  3. Fremantle, GWS, and Carlton: 75-80% chances of being finalists; 15-25% chances of Top 4; tiny-1% chances of being Minor Premier

  4. Hawthorn: 50% chances of being finalists; tiny chances of Top 4; no chance of being Minor Premier

  5. Essendon: 20% chances of being finalists; tiny chances of Top 4; no chance of being Minor Premier

  6. Collingwood: 5% chances of being finalists; no chance of Top 4 or of being Minor Premier

  7. Melbourne: tiny chance of being finalists, no chance of Top 4 or Minor Premier

  8. Gold Coast, St Kilda, Adelaide, West Coast, North Melbourne, and Richmond: no chance of being finalists, Top 4, or Minor Premier

LADDER FINISHES

The ladder projections using the Heretical outputs appear below.

There were a few large movements in teams’ Final chances this week, the biggest being:

  • Hawthorn -24% points

  • Carlton -13% points

  • Fremantle -12% points

  • Port Adelaide +27% points

  • GWS +13% points

  • Essendon +11% points

There are now 9 teams with about a 50% or better chance of playing Finals, and six teams with a 25% or higher chance of finishing Top 4.

TEAM AND POSITION CONCENTRATION

The HHI figures for the latest simulation replicates appear below. These provide us with estimates of:

  • How many ladder positions a team is currently effectively fighting for

  • How many teams are currently effectively fighting for a given ladder position

Overall uncertainty fell sharply this week, with the 8 least certain teams now effectively fighting for between about 5 and 8 ladder positions, and the 8 least certain ladder positions having about 6 to 7.5 teams effectively fighting for them.

On the teams side, those with the most certainty are Sydney, St Kilda, North Melbourne, Richmond, Adelaide, and West Coast, each of whom is effectively fighting for no more than 3 spots. On the ladder positions side, those with the most certainty are 1st, and 14th to 18th, each of which has effectively only about 2.5 or fewer suitors.

The ladder positions with the most uncertainty remain those in the middle of the table, from about 2nd to 11th.

We also see that the average team is still effectively competing for about 4.6 positions, and the average ladder position has about 4.6 teams competing for it.

WINS AND LADDER POSITION

Let’s look at the number of wins MoSHBODS thinks will be sufficient for given ladder positions.

Those probabilities vary by team, of course. Let’s have another look at them.

The main conclusions here is that no team other than the Western Bulldogs is a better than 50% chance of playing Finals with only 13 wins.

One key takeaway from this chart is that 15 wins make most teams (that can get to 15 wins) better than 50% chances of finishing Top 4, the exceptions being GWS and Geelong.

Here’s what it looks like on a team-by-team basis for all 18 ladder positions.

LIKELY IMPORTANCE OF PERCENTAGE

Next, we’ll use the simulations to investigate the estimated likelihood that any given pair of ladder positions will be determined after Round 24 based on percentage.

So, our current estimates are that:

  • There’s now about a 2-in-5 chance that 8th will be decided on percentages (it’s down by about 6% points on last week)

  • There’s now just over a 50% chance that 4th will be decided on percentages (up by about 5% points on last week)

  • There’s now about a 3-in-20 chance that 1st will be decided on percentages (up by about 4% points on last week)

TOP 2s, 4s, and GRAND FINALS

This week we’ll have a first look at the most-likely Top 2 and Top 4 teams, in order, at the end of the home and away season, and at the most-likely Grand Final pairings.

Top 2s

A Sydney/Brisbane pairing, in either order, accounts for over 50% of all simulation replicates, with a Sydney/Port Adelaide pairing in that order accounting for another 10%.

Interestingly, the Port Adelaide/Sydney pairing occurred only one-twentieth as often.

Port Adelaide’s only appearance in 1st amongst the most common Top 2s is when it is paired with Brisbane Lions, a pairing that occurred in only 1 replicate in 40, which is about half as often as a Brisbane Lions/Port Adelaide pairing.

The Top 10 pairings account for just under 90% of all replicates and, altogether, 44 different ordered paird occurred at least once across the 10,000 simulation replicates.

Top 4s

The key thing to note in the Top 4 table is that even the most-common quartet appeared in only 1 replicate in 25, so there remains a lot of uncertainty about this outcome.

That most- common set was Sydney/Brisbane Lions/Fremantle/Western Bulldogs, which appeared marginally more often than Sydney/Brisbane Lions/Western Bulldogs/Port Adelaide.

The Top 10 quartets account for only about 30% of replicates and, altogether, 581 different ordered sets occurred at least once across the 10,000 simulation replicates.

(For the curious, the most-common Top 8, in order, was Brisbane Lions/Sydney/Port Adelaide/Western Bulldogs/Carlton/Geelong/GWS/Fremantle, but it occurred in only 0.4% of all replicates making it an almost meaningless result. Altogether, across the 10,000 replicates there were 5,477 different ordered sets, 19 of which occurred in at least 20 replicates.)

Grand Finals

Despite the Top 2 places most often being occupied by Brisbane Lions and Sydney, the most common Grand Finals see Brisbane Lions take on Western Bulldogs or Port Adelaide. That accounts for almost 30% of all simulation replicates.

A Brisbane Lions/Sydney clash comes only third and accounts for another roughly 11% of replicates.

The Top 10 pairings, ignoring order, account for almost 75% of replicates and, altogether, 52 different pairs occurred at least once across the 10,000 simulation replicates, 33 of them in at least 29 replicates.

FINALS CHANCES

Finally, here’s what the simulations provide by way of finals-related probabilities.

A quick look at the AFL Futures markets would suggest that there might be a bit of value

IN THE MISS FINALS MARKET ON

Dockers

IN THE FLAG MARKET ON

Power and Dogs

Your regular reminder that I do not bet on these Futures markets.