2019 : Simulating the Final Ladder After Round 17
/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.)
(Note: this is the second time I have created this post. Squarespace, in a rare moment of madness, chose to overwrite this week’s with last week’s blogpost.)
RESULTS
We’re now in a situation where 8 teams have a 65% or better shot at the Finals, and a ninth team, Port Adelaide, has about a 1 in 3 shot. As well, four teams now have a 60% or better shot at a Top 4 finish, and two more better than 1 in 4 chances. Geelong still has the proverbial stranglehold on the Minor Premiership, but of the three teams trailing them by two wins, Brisbane Lions is estimated as having by far the greatest chance of overhauling them.
The biggest losers in terms of Expected Wins this week were Port Adelaide (down 0.9 wins) and Sydney (down 0.8 wins), and the biggest gainers were the teams that defeated them, Brisbane Lions (up 0.9 wins) and Carlton (up 0.8 wins).
In terms of estimated probability of making the Finals, the largest moves were:
BIGGEST LOSERS
Port Adelaide (down 30% points)
North Melbourne (down 12% points)
BIGGEST GAINERS
Essendon (up 19% points)
Adelaide (up 12% points)
Richmond (up 11% points, making it 38% points in three weeks)
GWS this week saw a 21% point decrease in its Top 4 prospects (making it 47% in two weeks), West Coast a 21% point decrease, and Port Adelaide an 11% point decrease. On the positive side, Brisbane Lions saw another 25% point increase (making it a 61% point increase in three weeks), Collingwood a 14% increase, and Richmond saw a 9% point increase (making it 21% points in two weeks).
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 few ladder finishes, with most teams having at least five positions for which they are 9% chances or better. Also, no team has a greater than 20% chance of finishing in any of the positions 5th through 14th.
WINS REQUIRED FOR TOP 8 AND TOP 4
GWS still have the best prospects of a Top 8 berth should they finish 11 and 13 for the season, but would only be about a 40% chance should they do so. Only Adelaide, amongst the remaining teams, is assessed as having better than a 1 in 4 chance of playing Finals with just 11 wins.
With 12 wins, the Giants are virtual certainties of playing Finals, and Adelaide have about a 90% chance. The estimated chances for most other teams lie in the 65 to 80% range, the exceptions being the Dogs and the Saints, who are worse than even-money prospects even if they finish the season 6 and 0.
We talked last week about the probability of the team in 8th spot finishing with just 10 wins. That remained roughly constant this week at about 0.1%, or 1 in 1,000.
GWS and Geelong are still the teams with the highest chances of finishing Top 4 with only 13 wins, but they’re only about 15 to 20% chances of doing so with that many wins. They’re still 90% or better chances though with 14 wins. Essendon, Richmond, Western Bulldogs and West Coast have the bleakest prospects of a Top 4 finish with only 14 wins. Their estimated conditional probabilities are only around 45 to 55% 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-Brisbane Lions 1-2 finish as, fairly comfortably, the most likely, it occurring in about 3 simulations in 10. Geelong-Collingwood and Geelong-West Coast finishes are the next most-likely, each occurring in about 3 or 4 simulations in 20.
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, Brisbane Lions, Collingwood, West Coast, Richmond, GWS, and Adelaide. Combined, the 10 combinations represent 92% of all of the simulations.
Looking next at Top 4s, we see that even the most-likely combination appears in less than 5% of simulations, that being for a Geelong-Brisbane Lions-Collingwood-West Coast finish.
Seven 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, Brisbane Lions, Collingwood, West Coast, Richmond, and GWS. Combined, the 10 combinations represent only 28% of all of the simulations.
In terms of the Top 8, there were 25,259 different orderings this week across the 50,000 simulations, none of them occurring more than 71 times (0.14%). In other words, even the current most-likely Top 8 ordering is about a 700/1 shot.
If we ignore order, we find 318 unique combinations of teams in the Top 8s across the simulations, the most commonly occurring set of being the eight teams with the highest number of Expected Wins in the simulations, namely Adelaide, Brisbane Lions, Collingwood, Essendon, Geelong, GWS, Richmond, and West Coast, which turned up in about 1 simulation in 3.
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 47% of the simulations (down 1% point), and 8th and 9th are separated by percentage in about 46% of the simulations (down 6% points). As well, 8th and 10th are separated by percentage in about 18% of the simulations, and 8th from 11th in about 6%.
In general, the likelihood that percentage will determine key positions has decreased a little 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 quite significantly, with the Gini coefficients rising to around 0.68.
At the team level, we saw reductions in uncertainty for 15 teams, the biggest being for Brisbane Lions, Richmond, North Melbourne, Essendon and Port Adelaide. The biggest increase in uncertainty came for West Coast who are now estimated as having 15% or better chances for each of positions 2nd through 5th.
Port Adelaide, Western Bulldogs, Essendon, Fremantle, Adelaide, North Melbourne, Hawthorn, and Richmond (who fill ladder positions 7th through 13th) 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.
If we adopt a ladder position viewpoint instead, 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 13th have the widest range of possible tenants.
Every ladder position except 17th 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 5th through 12th.
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.)
Essendon are involved in the three games with the highest estimated impact on the Final 8, and also in the fifth highest. Port Adelaide are involved in two of the five highest, and half of the ten highest. Two of the games in the upcoming Round 18 are in the ten games with highest estimated impact.
To finish we’ll again 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).
Here, we find Richmond are involved in three of the five highest impact games, West Coast in the two highest, and Collingwood in four of the seven highest.
