Matter of Stats

View Original

2019 : Simulating the Final Ladder After Round 16

The results of the latest 50,000 simulations appear below.

(If you’re curious about the methodology used to create them, you can start here.)

RESULTS

We’re now in a situation where 8 teams have a better than 60% shot at the Finals, and a ninth team, Essendon, still has an almost 50% shot. Five teams now also have about a 1 in 2 or better shot at a Top 4 finish. It’s beginning to look a lot more like musical chairs …

The biggest losers in terms of Expected Wins this week were GWS (down 0.9 wins), and the biggest gainers were the team that beat them, Brisbane Lions (up 0.9 wins). The Western Bulldogs also gained 0.8 Expected Wins taking them to 10.6.

In terms of estimated probability of making the Finals, the largest moves were:

BIGGEST LOSERS

  • Fremantle (down 21% points, making it a 56% point reduction in three weeks)

  • Adelaide (down 16% points)

BIGGEST GAINERS

  • Port Adelaide (up 18% points)

  • Richmond (up 13% points, making it 27% points in two weeks)

GWS this week saw a 26% point decrease in its Top 4 prospects, Adelaide a 17% point decrease, and Collingwood a 16% point decrease (making it a 31% point decline in two weeks). On the positive side, Brisbane Lions saw a 25% point increase (making it a 36% point increase in two weeks), West Coast a 17% increase (making it a 34% point increase in two weeks), and Richmond saw a 12% point increase.

Geelong are now just over 75 in 100 chances for the minor premiership, and West Coast about 1 in 8 chances.

DETAILED LADDER FINISH ESTIMATES

The detailed view of each team’s estimated probability of finishing in each of the 18 possible ladder positions appears below. Blank cells represent ladder finishes that did not occur even once in the 50,000 simulations, while cells showing a value of 0 represent estimated probabilities below 0.05%.

There remains a lot of uncertainty about all but the top and bottom ladder finishes, with most teams having at least five positions for which they are 10% chances or better. Also, no team has a greater than 20% chance of finishing in any of the positions 3rd through 14th.

WINS REQUIRED FOR TOP 8 AND TOP 4

After this weeks’ results, GWS are a little less likely to land a Top 8 finish with only 11 wins, but they’re still about 1 in 3 chances of playing into September should they scrape out just 2 more wins in their last 7 games. No other team has a better than 1 in 4 shot of doing the same.

With 12 wins, GWS remain near certainties for a Finals berth, Port Adelaide are close to 9 in 10 chances, and most other teams are around 3 in 4 chances. The major exceptions are St Kilda, who have only 1 in 2 chances (even with the 6 and 1 finish that 12 wins would require), and Melbourne, Western Bulldogs and West Coast, who have only about 3 in 5 chances.

There was some talk this week about the likelihood of a team making the Finals with only 10 wins. That remains a possibility, but a very remote one, as you can see from the chart below, which shows the relationship between win counts and ladder position (where the probabilities add across the win counts for a given ladder position).

From this we can see, for example, that an 8th-place finish is most often associated with 12 wins (which is also true of a 7th-place finish). Note that the cell for 10 wins and 8th-place is white, which tells us that the associated conditional probability is less than 1%. It is, in fact estimated to be 0.1%, meaning that there’s only about a 1 in 1,000 chance that the team that finishes 8th at the end of the home-and-away season will have exactly 10 wins.

GWS and Geelong are the teams with the highest chances of finishing Top 4 with only 13 wins, but they’re only about 40% chances of doing so with that many wins. They’re 90% or better chances though with 14 wins. Essendon, Richmond, Fremantle and West Coast have the bleakest prospects of a Top 4 finish with only 14 wins. Their estimated conditional probabilities are only around 2 in 3 should they finish 14 and 8 for the season.

TOP 2s AND TOP 4s

This week, again, we’ll look at what the simulations are suggesting are the most-likely combinations of teams finishing in key positions.

For Top 2s, we have the situation as shown at right, which now sees a Geelong-West Coast 1-2 finish as, fairly comfortably, the most likely, it now occurring in about 1 simulation in 3. Geelong-Collingwood and Geelong-Brisbane Lions finishes are the next most likely, each occurring in about 1 simulation in 8.

No other pairing has an estimated probability above 10%.

Across all 10 of the combinations shown here, seven different teams appear at least once: Geelong, West Coast, Collingwood, Brisbane Lions, GWS, Richmond, and Adelaide. Combined, the 10 combinations represent 89% of all of the simulations.

Looking next at Top 4s, we find, again, a much greater diversity of potential, with even the most-likely combination appearing in less than 3% of simulations, that being for a Geelong-West Coast-Collingwood-Brisbane Lions finish.

Five other orderings appeared in 2% or more of the simulations.

Across all 10 of the combinations shown here, six different teams appear at least once: Geelong, West Coast, Collingwood, Brisbane Lions, GWS, and Richmond. Combined, the 10 combinations represent only 22% of all of the simulations.

In terms of the Top 8, there were 34,356 different orderings this week across the 50,000 simulations, none of them occurring more than 26 times (0.05%). In other words, even the current most-likely Top 8 ordering is a 2,000/1 shot.

LIKELIHOOD OF PERCENTAGE DETERMINING POSITIONS

Next we’ll analyse how likely it is that key positions on the final home-and-away ladder will be determined by percentage because the teams in those positions finish tied on competition points.

This week, 4th and 5th are separated by percentage in about 48% of the simulations (up 2% points), and 8th and 9th are separated by percentage in about 52% of the simulations (up 1% point). As well, 8th and 10th are separated by percentage in about 22% of the simulations, and 8th from 11th in about 8%.

In general, the likelihood that percentage will determine key positions has increased a little again this week.

TEAM AND POSITION CONCENTRATION

One way of measuring how much uncertainty there is in the competition is to use the Gini measure of concentration commonly used in economics for measuring income inequality to quantify the spread of each team's estimated final ladder positions across the 50,000 simulation replicates, and to quantify the concentration in the probabilities across all the teams vying for any given ladder position.

This week saw the uncertainty about the final ordering of the teams decrease a little more, with the Gini coefficients rising to around 0.65.

At the team level, we saw reductions in uncertainty for 14 teams, the biggest being for Fremantle and St Kilda. The biggest increases came for Collingwood and GWS.

Essendon, Port Adelaide, Western Bulldogs, North Melbourne, Adelaide, Richmond and Fremantle (who fill ladder positions 6th through 12th) now have the most uncertainty, while Gold Coast, Geelong, and Carlton (who fill positions 18th, 1st, and 17th), still have the most certainty about their eventual ladder finishes.

Next, if we adopt a ladder position viewpoint, we see that 1st and 18th remain the positions with the narrowest range of likely occupants, with 90% to 95% chances that either of two teams will occupy them at season’s end, whilst positions 6th through 12th have the widest range of possible tenants.

Every ladder position except 1st, 2nd and 3rd became at least a little more certain this week about who will occupy it come the end of the home-and-away season, with most of the biggest increases coming for positions 11th through 17th.

GAME IMPORTANCE

Here are the updated assessments of the 30 most-important games between now and the end of the home-and-away season. (See this blog for details about how these are calculated.)

This week, 21 of the Top 30 are common to last week’s list.

We see some significant changes for some teams (especially Fremantle and Adelaide) this week in the level of involvement in games assessed as being in the Top 30 in terms of importance:

  • 7 games involving Essendon (down 1), Richmond (down 1), or Port Adelaide (down 1)

  • 6 games involving Adelaide (up 4), or North Melbourne (up 2)

  • 4 games involving Brisbane Lions (up 1), or Western Bulldogs (up 2)

  • 3 games involving Collingwood (up 1), GWS (up 1), or West Coast (no change)

  • 2 games involving Carlton (up 1), or Fremantle (down 6)

  • 1 game involving Geelong (no change), Gold Coast (down 1), Hawthorn (no change), Melbourne (no change), St Kilda (no change), or Sydney (down 2)

We can again see the commonsense of this list when we compare it with the estimated probabilities for teams finishing in 8th or 9th, which are:

  • 28%: Essendon

  • 27%: Port Adelaide

  • 26%: Adelaide

  • 19%: Richmond and North Melbourne

  • 18%: Western Bulldogs

No other team has a higher than 13% estimated probability of finishing in 8th or 9th combined.

To finish, this week we’ll also look at the importance of the remaining games in terms of their estimated likely impact on the Top 4 (using an equivalent methodology to that we used for assessing the likely impact on the Top 8).

We see, as we might expect, that the games assessed as being amongst the seven most important all involve pairs of teams with 12% or higher chances of finishing Top 4.

Summarising across all of the games in the Top 30, we have:

  • 7 games involving Collingwood, GWS, Richmond, or West Coast

  • 6 games involving Brisbane Lions

  • 4 games involving Adelaide

  • 3 games involving Carlton, Hawthorn, Melbourne, Port Adelaide, or Western Bulldogs

  • 2 games involving Gold Coast or North Melbourne

  • 1 game involving Essendon, Geelong, or Sydney

  • 0 games involving Fremantle or St Kilda