An Analysis of Strength of Schedule for the Men's 2027 AFL Season
/The men’s AFL fixture for 2027 was released this week and we will, as tradition insists, analyse it to see which teams we think have done better or worse than others.
In particular, we’ll look for answers to two questions:
Which teams fared best and which worst in terms of the overall difficultly of the teams they face in their schedule (the Strength of Schedule analysis) before and after adjusting for venue effects? Our measure here will be the MoSHBODS Combined Ratings of a team’s opponents, which we’ll adjust for venue effects for the supplementary analysis
Which teams fared best and which worst in terms of the matchups they missed out on given that only 23 games out of a possible 34 all-plays-all fixture are played (the Impact of Missing Schedule analysis)? Our measure here will be how much more or less likely would each team be to finish in the Top 10 ladder positions were the missing parts of the fixture actually played in some form.
THE FIXTURE
The 2027 AFL Fixture has, as usual, all 18 teams playing 23 of a possible 34 games, with each missing 6 of the home and 5 of the away, or 5 of the home and 6 of the away clashes that an all-plays-all full schedule would entail, with the exceptions of the team pairs shown in the table below who, because of the Gather Round, end up playing each other twice with the same team as the notional home team.
There is a single bye week for every team, these having been accommodated in 2027 by playing only seven games in Rounds 11, 12, 14 and 15, and eight games in Round 13. Long-time readers will know that I am not a fan of mid-season byes, so having only one instead of two per team this year makes me only half as unhappy.
THE RULE OF THIRDS
In determining the 99 games to be excluded from the schedule, the League has once again in the interests of what it calls "on-field equity" applied a 'weighted rule', which is a mechanism for reducing the average disparity in ability between opponents across the 207 home-and-away games, using the ladder positions from 2026 after the Finals series as the measure of that ability.
Those ladder positions are used to split the teams into three groups of six, with teams playing a larger proportion of the available matchups within their own third than with teams from the two other thirds.
Last year’s fixture included 71 of a possible 90 contests between teams from the same third, but this season that number has dropped to 68. The difference has been made up by fixturing one more contest involving a Top 6 and a Bottom 6 team, one more involving a Top 6 and a Middle 6 team, and one more involving a Middle 6 and a Bottom 6 team.
HOW TO measurE THE strength of aN OPPONENT
Measuring the strength of the schedule that a team faces across the home and away season requires, at the very least, a measure of the relative ability (or ‘strength’) of each of their opponents. Some analyses that you’ll see will use a team’s competition points from the previous home and away season as the measure of their ability, while others will instead use the team’s percentage (points for divided by points against).
To simplify the analysis and recognising the imbalanced nature of the draw, some will only consider these measures for the teams faced twice in the season, implicitly assuming that the combined strength of playing each opponent once is the same for all teams. That clearly isn’t true for two reasons:
An opponent playing at home is usually harder to defeat (ie stronger) than when playing away
By definition a team doesn’t play itself and so the collective ability of all opponents is different for every team
As a way of dealing with that first point, some analyses look to adjust for where games are played, for example increasing the estimated strength associated with those games where a team is forced to play interstate or away.
Finally, some analyses will adjust for the number of days rest that a team has had after the previous game, despite the evidence that this has little, if any, effect for the majority of teams.
We will, instead and as usual and as mentioned, measure each team’s relative ability based on its MoSHBODS Combined Ratings and, for the supplementary analysis, adjust this figure for venuing using the Venue Performance Values (VPVs) for each team at each venue, which is a measure of how much above or below expectation a team tends to perform at a given venue once you adjust only for the ratings of the two teams. They are generally positive for a team’s home grounds and negative for away grounds with this year some important exceptions.
CALCULATING STRENGTH OF SCHEDULE
The analyses this year will use the same methodology as we used last year, details of which appear in this post from 2015 and this one from the year before. The Venue Performance Values will be those calculated as at the start of the 2027 season, and the Team Ratings will be those that will apply for the first game of the 2027 season (which is about 63% of what they were at the end of the 2026 season).
The metric that we’ll use to estimate Strength of Schedule is the combined strength of the teams met across the 24 rounds of the home and away season, adjusted for venue effects for both teams in the supplementary analysis.
The estimated difficulty of a single game will be:
Excluding Venue Effects: Opponent’s Combined Team Rating
Including Venue Effects: Opponent’s Combined Team Rating + Opponent’s VPV for the game venue - Own VPV for the game venue
The second value will generally be higher than the first (ie the game will be assessed as more difficult) if the game is played at an opponent’s home ground instead of at a team’s own home ground, especially if the venue is interstate.
Higher values of either estimate imply that it is a more difficult game in the fixture.
The first thing we need for this metric is a measure of each team’s underlying abilities, and these appear in the table at right.
Recall that Ratings are measured in historical team score standard deviations, which at present maps 1 SD to about 25 points.
(These Ratings are obviously subject to error since that don’t, for example, account for player retirements or new draftings, nor are they even guaranteed to be an accurate measure of the team’s relative abilities at the end of the 2026 season.)
So, for example, Fremantle’s Offensive Rating of +0.25 means that they would be expected to score about 6.3 points more than an average team when playing an average team at a neutral venue.
This year there are far fewer teams that are ordered very differently based on their MoSHBODS Ratings compared to their 2026 ladder finish. In fact, no team is ranked more than three places differently by the two approaches.
As well, in the context of the AFL's competition "thirds", only four teams would be placed in a different third were MoSHBODS to be used rather than the final ladder in defining the boundaries:
Adelaide: Top 6 based on Ladder / Middle 6 based on MoSHBODS
Melbourne: Middle 6 based on Ladder / Top 6 based on MoSHBODS
GWS: Middle 6 based on Ladder / Bottom 6 based on MoSHBODS
Gold Coast: Bottom 6 based on Ladder / Middle 6 based on MoSHBODS
The average and range of the Combined MoSHBODS Ratings of teams from each of the AFL thirds is as follows:
Top 6: Average +0.44 SDs / Range 0.43 SDs
Middle 6: Average +0.08 SDs / Range 0.34 SDs
Bottom 6: Average -0.52 SDs / Range 0.82 SDs
We can see that the Top 6 teams from the final 2026 ladder are, on average, stronger than those from the Middle 6, and that those from the Middle 6 are, on average, stronger than those from the Bottom 6.
Ignoring venue effects, which we'll come to in a moment, the difference between playing an average Top 6 team and an average Bottom 6 team is about 0.96 SDs or about 24 points. Roughly speaking, every 1 point difference translates to a 1% difference in the probability of victory, so 24 points is about 24% points, or about 5.5 wins per season.
Also, with such a relatively larger spread in the ratings across the teams in the Bottom third - about two goals or a 12% point change in win probability - it's quite important which of the teams from this third a team plays. Also, teams from the Bottom 6 will be especially disadvantaged by not being able to play themselves.
VENUE PERFORMANCE VALUES
This year we’ll see that the inclusion of VPVs has a VERY significant effect on the ranking of team schedules.
As noted VPVs measure how much better or worse teams, on average, play at each venue relative to their own and their opponents’ underlying ability. They are a logical extension of the notion of a "home ground advantage", and account for the fact that not all away venues are the same for every team.
VPVs are calculated as follows. For each team–venue combination, the VPV estimate blends the team's rating-adjusted historical performance at that venue with a default that is zero when the venue is in the team's own region, and about −0.43 SDs otherwise. Only the preceding 3.76 years of matches are included in the calculation. The default's weight reaches zero after approximately eight games and the historical residual mean performance is multiplied by the model's regression factor of about 0.23. This serves to prevent VPVs from varying a great deal based on a single blowout.
The Venue Performance Values, calculated as they would be on day 1 of the 2027 season, are summarised in the table below for all of the venues at which a team appears at least once sometime during the 2027 home-and-away season. For details on how these have been calculated, refer to this blog.
Interestingly, this year the Gold Coast are playing at 13 different venues, an increase of one compared to last season. GWS and North Melbourne are again playing at 11 different venues, while West Coast visit 10. In contrast, Adelaide, Western Bulldogs, Hawthorn and Fremantle play at seven different venues, and Collingwood at only six.
Venue Performance values are, like Ratings, measured in Standard Deviations (SDs), and are added to an opponent's underlying MoSHBODS Combined Rating when used in the Strength of Schedule calculation. So, for example, we can say that Geelong is, on average, a +0.14 SDs better team than their underlying +0.53 SDs Rating when playing at Kardinia rather than some neutral ground.
The supplementary estimate includes both a team’s own VPV for that venue (as an offset) as well as that of its opponent. I include the team’s own VPV because I think, on reflection, this better encapsulates the full venue effect, although prior to 2021 only opponents’ VPVs were included.
To reiterate the rationale for this by way of a concrete example, imagine moving a Brisbane Lions v Richmond game from the Gabba to Carrara assuming that the VPV numbers in the table above apply. Under the methodology where a team’s own VPV was ignored, that change of venue would have virtually no effect on Brisbane’s estimated Strength of Schedule, because Richmond’s VPV at both venues is about the same at around -0.47 to -0.42 SDs, and we would ignore Brisbane’s VPV at those venues. But, Brisbane is estimated to be about a 5-point better side at the Gabba compared to Carrara - a fact which seems worthy of inclusion in the Strength calculation. The fixture with that game at the Gabba is surely an easier one for Brisbane than the one with that same game at Carrara.
This approach tends to increase the estimated schedule strength for teams that have relatively low VPVs at all venues (for example, Essendon and Richmond), and decreases the estimated schedule strength for teams that have relatively high VPVs at home or at all venues. If a team seems to enjoy no home ground advantage anywhere, is it reasonable to therefore assess them as having a more difficult schedule and, conversely, if a team seems to play well at home or at all venues, is it reasonable to therefore assess them as having a less difficult schedule? Ultimately, this is probably a question of personal preference but, again for this year, I’m answering “yes” to both those questions, at least for the purposes of including the venue-adjusted estimates.
Because of the manner in which they are calculated, the Venue Performance Values incorporate the effects, if any, of interstate (technically, ‘out of region’) travel, which you can see for many teams. For example, if you run your eye along the row for the Gabba in the table above you’ll see that most interstate teams are about 0.3 to 0.4 SDs (or about 7.5 to 10 points) worse at this venue.
The common out-of-region default parameter gives sparsely observed away venues similar estimates. because:
only the last roughly 4 years of data is included in VPV calculations
for an out-of-region venue, the weight on the −0.43 SD default only falls to zero once a team has played eight games there within the approximately 3.76-year window. A lot of Victorian teams will not reach that hurdle on interstate grounds. If they do, the VPV then equals the team’s average rating-adjusted performance residual at that venue multiplied by the model’s regression factor of about 0.23, so the estimate remains damped.
Alternative VPV specifications changed some team schedule difficulty rankings, but did not provide compelling predictive evidence for replacing the original method used here. For example, a smoother k=8 alternative reduced average VPV revisions, with essentially unchanged recent margin accuracy. Its opponent-only versus venue-adjusted rank correlation remained negative, at −0.29 compared with −0.43 here. Repeating the analysis using the VPV calculations in the Strength of Schedule analyis of last year (which included about 8 years of matches) also did little to reduce the impact of including a venue component.
2027 STRENGTH OF SCHEDULE
After performing the necessary calculations for all 23 games for every team, taking into account who each team plays as well as where, we arrive at the Strength of Schedule estimates below, within which larger positive values represent more difficult schedules.
(See the IMPACT OF MISSING SCHEDULE section below for each team’s actual and missing schedule.)
Note that the numbers shown in these tables are all aggregates and so will include either 11 or 12 games in the home and the away data, depending on the particular team’s fixture.
In the left portion of the table we have the combined strength of the opponents faced by a team at home, split into the contribution from underlying ability and from venue effects. We would generally expect the Aggregate Net Venue Performance figure to be comfortably negative for all teams in this part of the table, since their opponents are likely to have negative VPVs and they themselves are likely to have positive VPVs at the venues involved. That is, indeed, the case here, with Collingwood, Essendon and Richmond the outliers because of their small negative VPVs at the grounds where they play their home games.
Here, the disproportionate impacts of interstate travel on teams’ home schedules are apparent, as all of the non-Victorian teams have Aggregate Opponent Net VPVs of -2.1 or less. Of the Victorian teams, only Geelong and Hawthorn have values this low.
The middle section of the table looks at the combined strength of the teams played away from home, again split into the contribution from underlying ability and venue effects. Here we would expect the Aggregate Net Venue Performance figures to be positive for a team, since their opponents are likely to have positive VPVs at their home grounds. That is, indeed, the case for all teams, though least of all for Hawthorn.
We see that adding venue effects has more impact on the strength estimates of home schedules than away schedules: the correlation between Home Opponent Ability and Home Strength of SOS is +0.47, while the correlation between Away Opponent Ability and Away Strength of SOS is +0.81.
Combining the home and the away pictures to estimate a Total Effective Strength of Schedule (SoS) figure and summarising the results, we can roughly classify the schedules for each team as per the table below.
If the aim of fixturing is to create schedules that are harder for strong teams and easier for weak teams, we can see that this objective is much better satisfied by the Team abilities only view than by the Team abilities plus venue effects view.
On reflection, that’s probably as much as we can ask of the fixturing folk because they have limited ability to adjust for venue effects given the constraints of having to play most home games for most teams at their regular home grounds and the fact that some teams are seemingly just better at their home venue than are other teams at their home venue, and that some teams are just better at away venues than others, too.
On balance, I still favour the team abilities plus venue effects measure because it accounts for both who a team plays and where, while recognising the uncertainty in the venue estimates. The team abilities only measure is more useful for assessing how well the fixture distributes stronger and weaker opponents.
If we include Venue Effects, the difference between the hardest and easiest schedules this year amounts to about 6.3 SDs or about 158 points across the season, which is just under 7 points per game. Excluding Venue Effects the difference is only 1.9 SDs or about 48 points across the season, which is just over 2 points per game
DETAILED GAME-BY-GAME NET VPVs AND STRENGTH OF SCHEDULE
The table below (click on it to access a larger version) provides a full breakdown of the Strength of Schedule calculation for every fixtured game. The top row of figures records the Combined Rating of the relevant team, and the numbers in the body of the table the Net VPVs for each contest.
So, for example, when Brisbane Lions play Fremantle at home at the Gabba the Venue component is -0.51, which is Fremantle’s VPV at the Gabba of -0.37 less the Lions’ VPV of 0.14. If we do the same thing for every Lions home game and add them, we get -5.01, which is the same as the figure in the earlier Actual Schedule table.
The Opponent component of the SoS calculation for that Dockers game is the +0.55 Rating of the Dockers, and if these are added for all Lions home games we get -1.24. The Combined figure is -6.25, which also appears in that earlier table.
IMPACT OF MISSING SCHEDULE
We can also view the schedule on the basis of the opportunities missed by a team as a consequence of playing only 23 of a possible 34 games.
The table below summarises the missing games in the 2027 Fixture, denoting with H's those games missed that would have been home games for the team named at left of the row, and as A's those that would have been away games for the team named at left of the row.
Note that I've ordered the teams on the basis of their final 2026 ladder positions, the same ordering that was used for implementing the AFL's 'weighted rule'.
Here are a few interesting points that you can glean from this table (with (+) meaning generally favourable and (-) generally detrimental based solely on the thirds from which the teams come:
Top 6 Teams
(+) Face only two others from Top 6 twice: Sydney and Adelaide
(-) Face only one from Middle 6 twice: Hawthorn
(-) Face only one from Bottom 6 twice: Brisbane Lions, Fremantle and Geelong
Middle 6 Teams
(+) Face only one from Top 6 twice: Carlton
(+) Face only two other teams from Middle 6 twice: Melbourne, Collingwood, St Kilda, and GWS
(-) Face only one from Bottom 6 twice: Western Bulldogs
Bottom 6 Teams
(+) Face only one from Top 6 twice: North Melbourne, Richmond, and Essendon
(+) Face only one from Middle 6 twice: West Coast
(-) Face only two from Bottom 6 twice: Gold Coast and Port Adelaide
To measure the impact of the missing schedule we will estimate the difference in the simulated probability of making the Top 4, Top 6, and the Top 10 between that obtained using the actual 23 games across 24 rounds fixture and that obtained using a 34 games across 35 rounds fixture in which all teams meet all other teams home and away (with the exception of the four pairings tabulated earlier). What we’re measuring here is the difference between an idealised all-plays-all season and a 23 game, 24 round season modelled on the actual fixture.
We assume that all missing games are played at the home team’s most common home ground in the actual season, which this year are as per the table at left.
The simulations compare 23-game seasons with 34-game seasons. Changes in finishing probabilities therefore reflect both the additional matchups and the longer season, which reduces variability and gives teams more opportunities to demonstrate their relative strength, so strong teams’ top-10 chances tend to rise and weak teams’ tend to fall, whatever the matchups. The missing games are appended as 11 weekly rounds, and ratings and venue estimates evolve as games are simulated. The differences should be read as the effect of this completed-season scenario, rather than as a pure measure of fixture unfairness.
For this measure, a larger difference in the 35- vs 24-round probabilities for favourable ladder positions is interpreted as a more disadvantageous schedule.
The table below provides information about how often each team finished in a given ladder position at the end of the home and away season in the 1m all-plays-all simulations (where the 99 missing games are played as 11 new rounds a week apart with a different random ordering of the games in each replicate) versus the 1m simulations based on the actual fixture.
We can summarise this table by looking just at the differences for Top 4, Top 6, and Top 10 finishes.
If we focus solely on making or missing Finals, the teams with the:
Largest increases in Top-10 probability when the missing games are added: Gold Coast, Brisbane Lions, Western Bulldogs, and GWS
Largest decreases in Top-10 probability when the missing games are added: North Melbourne, Adelaide, Richmond, Collingwood, and Essendon
SUMMARY
The table below summarises the team schedule difficulty rankings on a number of the metrics.
It’s interesting to note that the team ordering based on aggregate schedule strength ignoring venue effects is quite similar to the team ordering based on the impact of the missing schedule. We found a similar result last year. The rank correlation this year is +0.77.
CONCLUSION
A final team classification appears in the table below. It primarily uses estimated Schedule Strength including venue effects (left to right) and estimated Schedule Strength excluding venue effects (bottom to top) with annotations in brackets to describe the Missing Schedule impacts where - means the missing games reduce the probability of playing Finals and + means the missing games increase the probability of playing Finals.
My preferred measure of schedule strength includes both opponent strength and the net venue effect, while recognising that the venue estimates are uncertain. Because of that uncertainty, I show both measures rather than treating either ranking as definitive. The completed-season simulations provide a further perspective on how teams' prospects change when the omitted games are added, but also reflect the longer season and the model's evolving ratings. Taken together, these comparisons show where the conclusions are consistent and where they depend on how schedule difficulty is defined and how venue effects are quantified.
CODICIL
It’s been suggested that a better comparison than the 23-game versus 34-game simulations would be randomly selected 23-game seasons versus the actually fixtured 23-game season.
The results of such a comparison appear as the two rightmost columns in the table below. Note that each randomly selected season is structured so that each team plays the other 17 once and six teams twice and has 11 or 12 home games.
The results are very similar to those we obtained using 34-game seasons. The correlation between the percentage point changes is +0.951.
