Matter of Stats

View Original

2018 : Simulating the Final Ladder After Round 10

Updated final home and away ladder simulations appear below.

(For details on the methodology, see this post from earlier.)

In terms of Expected Wins, the week's big gainers were three of the weekend's winners :

  • Melbourne (+1.0)
  • Essendon (+0.9)
  • North Melbourne (+0.8)

Melbourne and Essendon increased their Expected Wins by a significant amount for the second week running, though Melbourne seems more likely to meaningfully benefit from these increases since Essendon remain only about 7% chances for a Finals berth.

The big losers were the teams the big gainers beat, and included GWS for the second week running:

  • Fremantle (-1.0)
  • GWS (-0.9)
  • Adelaide (-0.9)

Just 2 Expected Wins now separates Sydney in 4th from Adelaide in 9th.

In terms of likelihood of making the Finals, the week's big gainers were:

  • North Melbourne (+20% points)
  • Melbourne (+13% points)
  • Collingwood (+12% points)

The week's big losers were:

  • Adelaide (-19% points)
  • GWS (-17% points)
  • Fremantle (-13% points)

Melbourne made huge gains in the quest for a Top 4 spot (+23% points), mostly at the expense of Adelaide (-18% points).

West Coast have a higher Expected Win count than Richmond, but a slightly lower estimated probability for taking out the Minor Premiership and finishing Top 4 or Top 8.

TEAM AND POSITION CONCENTRATION

This week's Gini concentration measures appear below, and reflect an overall increase in our knowledge about which teams will finish where at the end of the home and away season. (For information about how to interpret these measures, see this blog.)

That said, there were some decreases in the level of certainty about the ladder finishes for a handful of teams, most notably Essendon, who now have a 5% or greater chance of finishing in any of the positions from 9th to 16th.

Melbourne, in contrast, has seen a significant compression in the spread of its potential ladder finishes over each of the last two rounds, and now finds itself with a 10% or higher estimated probability for finishing in any of the first five spots on the ladder. Two weeks ago there were only two ladder positions for which its estimated probability was 10% or higher (6th and 7th).

The average Gini coefficient shown now at the foot of the table shows how, in aggregate, team final positions are becoming more certain.

Looking next at each of the ladder positions we see that all but four ladder positions became at least a little more confident about which team will occupy them come the end of the home and away season, the two big exceptions being 15th and 18th. That said, the overall levels of confidence about who will take these spots remain relatively high compared to other ladder positions.

Certainty increased most of all for positions 3rd and 12th, although a slew of other positions saw increases of similar amounts. We can see the overall increase in certainty averaged across all 18 ladder positions from the figures that appear at the base of the table. These show, as we might expect, a very similar pattern to the figures at the base of the previous table, which adopted a team-based view.

It remains the case that most certainty is associated with the top four and bottom six places on the ladder, and least certainty with positions 7th to 10th. At this point, it looks as though the battle for Finals positions will be for the last couple of them.

WINS REQUIRED

Below are the updated estimates of teams' likelihood of making the Finals depending on the number of wins that they record. (Recall that, because we're working with a sample of simulated final ladders, our estimates have sampling error, so we show them as 95% confidence intervals here with the mean estimate shown as a point. In some cases, our sample is so large that the interval essentially collapses to a point, at least to the resolution shown here.)

Melbourne, in defeating Adelaide and boosting their percentage so much in doing so, significantly increased the likelihood that 12 wins will be enough for them to make the Finals. They're now better than a 75% chance should they record 5 more wins. North Melbourne have about 70% chances and Geelong 65% should they record another 6 wins.

The Western Bulldogs still have the lowest chance of making the Finals with just 12 wins, with an estimated probability of around 40% should they rack up the requisite 8 additional wins. The Gold Coast is the only other team with less than even money odds to make the Finals with only 12 wins, though Fremantle, Essendon and West Coast are quite close to being only even money chances.

Thirteen wins still makes every team near certainties for a spot in the Finals.

Turning next to the analysis for a Top 4 finish we find that Richmond's chances remain high with just 14 wins, but Melbourne's are now higher. They are the only two teams with a greater than 50% chance of a Top 4 finish with 14 wins.

Fifteen wins still virtually guarantees any team a spot in the top 4. Port Adelaide and Adelaide are the teams most likely to miss the Top 4 with 15 wins.

Next, we explore the inter-team dependencies in the composition of the final 8 by estimating the probability that a particular teams makes the 8 conditioned on some other team making or missing the 8.

These are shown in the chart below, as they were last week, as arrows with the base of an arrow marking the estimated probability that a team makes the 8 conditioned on a nominated team missing the 8, and the arrow head marking the estimated probability that a team makes the 8 conditioned on a nominated team making the 8.

Adelaide's Finals chances remain most strongly linked to GWS', while Collingwood's are most heavily influenced by Fremantle's and Essendon's, though neither of these teams are likely Finalists.

GAME IMPORTANCE

Lastly, here's the updated list of the 30 most important games remaining in the fixture on the basis of their expected impact of all 18 teams' finals chances. Matches involving any two of Hawthorn, Adelaide and Geelong sit atop the list, as these are the teams most likely to rely on favourable results in these "8 point games" to secure spots in the Finals, one at the expense of the other.

Games that were also on the Top 30 list last week are marked with an asterisk. Fourteen of the 29 games that were on last week's list and that weren't Round 10 games are still on the list. 

Looking just at the next few rounds, we have the following counts of Top 30 games:

  • Round 11: 3 (of 9 games)
  • Round 12: 2 (of 7 games)
  • Round 13: 1 (of 6 games)