2017 : Simulating the Final Ladder After Round 14

/

The latest set of 50,000 simulations of the remainder of the 2017 AFL season saw 8 teams' Finals chances move by more than 5% points.

(For details on the methodology I've used in the simulations, please see this blog.)

RESULTS

The charts below summarise the distribution of ladder finishes for the 18 teams from the simulations run after Rounds 13 and 14. Note that the ordering of the teams is slightly different in the two charts, reflecting improvement or deterioration in some team's expected ladder finish.

AFTER ROUND 14

AFTER ROUND 13

In the race for the finals, the big winners this week were:

  • Melbourne: +10% points (up by 33% points now in last two weeks)
  • Richmond: +8% points
  • Western Bulldogs: +8% points
  • Port Adelaide: +7% points
  • Sydney: +5% points (up by 14% points now in last two weeks)

The big losers were:

  • West Coast: -13% points
  • Collingwood: -12% points
  • Essendon: -7% points

There remains, though, a lot of variability in the likely finishes for most teams, especially for the teams ranked from about 5th (Melbourne) to 12th (Essendon).

We can see when we look at the same data in the form of a heat map as well.

AFTER ROUND 14

AFTER ROUND 13

If we, again, arbitrarily, said that a roughly 10% or greater chance of finishing in a particular ladder spot was a "reasonable" chance, we now have:

  • 4 teams with reasonable chances for the minor premiership (1 more than last week)
  • 4 to 6 teams with reasonable chances for every position from 2nd through 17th (where last week it was mostly 4 or 5 teams in contention for these spots)
  • 1 team with a reasonable chance for the wooden spoon (1 fewer than last week)

The race for 7th and 8th is particularly intriguing. There are now 7 teams with chances ranging from about 9% to 12% to finish 7th, and 7 teams with chances ranging from about 8% to 12% to finish 8th.

GAME IMPORTANCE

Next, let's update the game importance table. It shows, you'll recall, how each team's chances of making the finals varies depending on the result of each of the remaining games. The overall Weighted Ave Importance figure tells us the average amount by which the teams' finals chances are expected to change on the basis of the outcome of a particular game. A larger figure means that the game is expected to change teams' finals chances, on average, by a larger amount and hence is a more important game.

(For more details about the methodology and the rationale for it, again please see this blog.)

Round 15 is something of a proverbial mixed bag, with three games from the top quintile in terms of importance topping and tailing the round, three more from the bottom quintile, and the remaining three all from the 3rd or 4th quintiles.

The Dogs v Eagles game is estimated as being especially influential, potentially one of the most Finals-shaping of the games in the schedule that remains as it alters the two teams' chances by about 28% points each.

INTER-TEAM DEPENDENCIES

This week, a new analysis that looks at the dependencies between teams in their chances for making the Final 8.

For this analysis we calculate, for each team, the proportion of occasions in the simulations where it made the Final 8 when:

(a) A particular team (the "Nominated Team") did not make the Final 8

(b) That same team did make the Final 8

These are charted below, but only for those teams whose (unconditional) chances of making the Final 8 are between 5% and 95%. 

Each arrow starts at the probability of a team making the Final 8 given that the nominated team (named in the row) did not make the Final 8, and ends at the probability of that same team making the Final 8 given that the nominated team did make the Final 8.

So, for example, the very first arrow shows that Collingwood are about a 10% chance of playing Finals if the Western Bulldogs do, and a 15% chance if they don't.

The length of an arrow is an indication of how dependent two teams' Finals chances are, and is generally longer for teams that are scheduled to meet at least once more in the remainder of the season. The length also depends though on the nature of the nominated team's schedule and the extent to which its wins or losses in those games will, directly or indirectly, open or close the door for the team we're analysing. This week, for example, if the Eagles beat the Dogs, that lifts St Kilda's chances (as you can see in the Game Importance table above).

We can also make an assessment of how generally dependent a team's Finals chances are on the fates of other teams by considering the average length of the arrows in a team's block. It's clear, for example, that the chances of St Kilda, Sydney and West Coast are more dependent on the fate of other teams' than are, say, Melbourne's.

NUMBER OF WINS REQUIRED

We'll finish this week with one final, new analysis, this one looking at how likely it is that a team will play Finals given that it finishes the home-and-away season on exactly 11, 11.5, or 12 wins.

These estimates, because they're based on a sample of simulations, are subject to error (as are all of those in this blog), which here is reflected in the width of the segment.

We see that, for example, GWS are only about a 5% to 10% chance of playing Finals should they win only 1 of their remaining games and finish on 11 wins. The segment for this estimate is quite wide because this outcome occurred in relatively few simulations, so our estimate of the probability is less precise.

There's a surprising amount of variability in teams' estimated chances of making the 8 if they finish on only 11 wins - from near zero for Fremantle to about 40% for Port Adelaide. Port Adelaide's estimate is relative high partly because of its healthy percentage (135 and 2nd best in the competition) but also because its remaining schedule is relatively rich in teams with whom it would most likely be vying for a spot on the edge of the Final 8 (viz Richmond, West Coast, Melbourne, St Kilda, and the Western Bulldogs). Its three wins to get to 11 would, therefore, be more likely to come at the expense of one or more of those teams.

Teams' chances with 12 wins are similarly quite variable, from a low of just over 50% for Fremantle, whose percentage is a paltry 80, to a high of over 90% for Port Adelaide.