2026 Victorian State Election

Violin Icon
Violin Plot
Shows seat distribution.
Fatter = more likely.
Taller = more seats.
Simulations Explained
Each point inside the violins represents one of 100 sampled simulations of the election outcome.
White text shows Average Seats won per party across all 1000 Simulations.
Hover or Touch any point to see seat numbers.
For more details, see the Methodology below.
See individual electorate projections below.

Next Government Probabilities

COAL majority COAL
--%
COAL + ON minority COAL+ON
--%
LAB majority LAB
--%
LAB + GRN minority LAB+GRN
--%
No clear minority Unclear
--%
88 Electorates · Seat-by-Seat

Probabilities by Electoral District

Hover any seat to inspect win probabilities, or click to jump to its full breakdown below. ← Swipe across the chamber to explore all 88 seats →

Note on Probabilities
This model is probabilistic: a party winning in 80% of simulations does not mean it is 'predicted' to win.
It simply means that outcome is the likelier result given current polling, historical patterns, and candidate-level uncertainty.
Take 5 seats where the favourite is given about an 80% chance of winning:
Richmond (GRN), Thomastown (LAB), Kew (LIB), Gippsland East (NAT), Bendigo West (ON).
On average, one of these favourites is expected to lose.

Loading...

Projected First Preference Votes

Violin Icon
Violin Plot
Shows vote distribution.
Fatter = more likely.
Taller = higher vote share.
TCP:
Understanding the Projections
The violins show the first preference vote distribution across 1000 simulations. Fatter areas indicate the most likely outcomes.
Hover to see exact percentages, or click Simulate to view sample results as white, clickable dots on the violins.
See the workings of the model here.

Chances of Winning

Methodology

Guide to the Model

How the 2026 Victorian forecast converts 2022 district results, candidate lists, and statewide polling into First Preference vote distirbutions and seat-by-seat win probabilities.

1. Incumbency Advantage

Sitting MPs tend to outperform their party’s underlying vote, buoyed by name recognition and an established local profile. We can measure this personal vote by comparing a party’s Lower House (Legislative Assembly) result against its Upper House (Legislative Council) vote in the same district, adjusted for differences in who contested each ballot.

In Victoria, this advantage averages around 1.3% for a first-term MP and rises to roughly 4.4% by a fourth term, with notably stronger effects in rural seats. In 2026, 20 incumbents are retiring or not recontesting—12 of them Labor MPs. Removing those personal votes contributes to putting strong Labor seats like Bendigo East and Macedon at risk of falling to One Nation and the Liberals.

Example MP
District
2026 Status
Est. IA
Jacinta Allan
Bendigo East
Retiring (6 terms)
-3.7
Danny Pearson
Essendon
Retiring (2 terms)
-3.0
Bill Tilley
Benambra
Retiring (4 terms, rural)
-7.3
Jess Wilson
Kew
1st-term gain
+1.3
Of statistical interest
The Lower-House vs Upper-House comparison works by projecting Upper House votes onto each seat’s Lower House Three-Candidate Preferred (3CP) ballot—reallocating absent minor parties via fitted transition matrices—and measuring how much incumbent MPs outperform those projections. For incumbents who held their seat in 2022, the model applies a 40/60 shrinkage estimator: 40% comes from the incumbent's observed 2022 3CP residual, and 60% from a linear regression conditioned on log-tenure length and demographic classification (Inner Metropolitan, Outer Metropolitan, Provincial, and Rural). For newly elected incumbents since 2022—where no prior personal-vote residual exists—the model relies purely on the linear prediction.

2. Changing Parties: Constructing a 2022 Baseline

The 2022 election result is the natural starting point for modelling 2026. While Labor, the Coalition, and the Greens are recontesting every district, the mix of minor parties changes substantially between elections, making raw 2022 results not directly comparable to 2026. The first step in the process is therefore estimating how voters would have voted in 2022 had the 2026 set of parties been on the ballot.

In 2026, this adjustment is most consequential for ON and VSOC, both of which have expanded to contest all 88 districts. At the same time, parties like AJP and FPV have scaled back the number of seats they are contesting. Re-expressing 2022 under the 2026 candidate list substantially lifts the starting statewide vote share for ON from 0.3% to 3.9%.

Seats Contested
Baseline Vote
Party
2022
2026
2022%
2026%
One Nation
4
88
0.28%
3.88%
Vic Socialists
22
88
1.34%
2.74%
Freedom Party
58
1
1.75%
0.01%
Animal Justice
88
40
2.51%
1.55%
Of statistical interest

Generating this counterfactual requires removing parties no longer contesting and adding new entrants to each seat. Both steps rest on a single principle: to estimate preference flows from a change in the ballot, it is sufficient to know a district’s vote shares with a party included in the candidate set, and with it excluded.

  • Removing parties: Since 2022, the Victorian Electoral Commission (VEC) has published Distribution of Preferences (DOP) tables showing how each eliminated candidate’s preferences flowed to the remaining parties. For the weakest-polling candidates, the table gives the exact preference flow to the remaining field. For stronger exiting candidates, the DOP tables are reverse-engineered to project how their votes would have flowed with the weaker candidates still included.
  • Adding parties: New entrants are inserted using the Upper House (UH) minor party flow model. This uses each district’s 2022 UH result as a base, and estimates how votes would flow if only a given set of parties contested (e.g., Labor, Coalition, Greens, and a selection of minor parties grouped into Left, Centre, and Right ideological blocs). Non-contesting minor party votes are routed via a logit model based on bloc alignment, each bloc’s UH vote share, and the number of contesting parties in each bloc, before being allocated sub-proportionally among candidates within the same bloc. For newly registered parties that did not contest the 2022 Upper House, an initial share is first carved out using a prior based on past electoral performance where available, with the model applied on top.

3. Modelling the Independent Vote

Independents are notoriously difficult to model probabilistically because most have no party brand and no historical performance baseline to work off. Rather than treating every new independent the same—as on track to not even achieve 2% of the vote—Victoria 2026 introduces a dedicated Independent Vote Model built on observable signals of Independent campaign seriousness: how early they announced, whether they have past experience as a local councillor, and whether they belong to an established political grouping (such as Climate200-backed "Teal" independents).

Independents who launch long before nominations close, or who already have a council profile, historically tend to win larger vote shares. Announcing months in advance doesn't guarantee success—plenty of long-shot campaigns have discovered that—but serious, well-resourced contenders almost always organise well ahead of nominations (with the fascinating exception of Suzanna Sheed in 2014). The plot below shows how Independent vote shares in the 2014–2022 Victorian elections correlated with campaign length and candidate characteristics.

Of statistical interest
Generic independents are modelled in two stages:
(1) A Gamma GLM predicts the joint Independent vote share in a district using grouped candidate covariates—campaign length, presence of an elected councillor, the district’s Independent vote at the previous election, and the parked vote shares of any recontesting Independents.
(2) A secondary split model divides the gained vote among individual Independents using their candidate-level covariates.
Independents affiliated with established movements (Teals since 2018 and The West Party in 2026) are modelled separately via historical Lower-vs-Upper House party defection rates (see here for a federal analogue). Priors for The West Party are also informed by comparable regionalist movements (e.g. NQ First at the 2020 Queensland election and Voice for the West at the 2014 Victorian election).

4. Choosing the Right Polling Swing

Applying statewide polling shifts to 88 distinct electorates requires the right tool:

  • Uniform Swing applies the flat statewide shift in polling from the 2022 baseline to every district equally, ignoring local baselines. A massive ON surge from 3.9% to 22.3% statewide does not land identically in an inner-Melbourne seat where their baseline is under 1% and a regional seat where they start near double digits. Further, applying a flat negative swing can project a party below 0% in low-polling seats.
  • Proportional / ALR Swing scales each party’s shift relative to its local baseline while mathematically locking every seat’s total to exactly 100%. As the table on the right shows for One Nation, a seat’s projected swing scales naturally with its underlying local strength—the same dynamic that accurately captured the dispersal of One Nation’s surge at the 2026 South Australian election.
  • Splitting the "Other" vote: Combined Other support has fallen sharply in 2026 polling, likely driven largely by voters for minor right-wing parties switching to One Nation. To avoid unfairly penalising independents for that drop, the model adjusts the internal split of the Other bloc using DemosAU’s Legislative Council polling benchmarked against the 2026 South Australian results.
ON Swing
District
Base
Unif.
ALR
Kew
1.0%
+18.4
+6.5
Bentleigh
2.9%
+18.4
+15.5
Kalkallo
3.2%
+18.4
+18.7
Morwell
12.9%
+18.4
+39.5
Of statistical interest
For a point estimate, a renormalised proportional swing is mathematically equivalent to an Additive Log-Ratio (ALR) shift. Working in ALR space lets the model simulate both statewide polling error and seat-by-seat deviations from the statewide swing using a Multivariate Normal covariance structure, while guaranteeing every party stays above 0% and every seat sums to 100%. Statewide polling variability is derived from historical polling errors from various Australian elections and decreases as election day approaches, while seat-level variance is modelled off residuals from the election-to-election shift in the 4 Victorian elections since 2006. These are backtested so that simulated first-preference distributions match the calibration of past Victorian results. Finally, because unusually large statewide swings create wider errors in local dispersion, the ALR error covariance has been scaled up for 2026—informed by the increase in seat-level variability observed at the 2026 South Australian election relative to previous elections.

5. Preference Flows & Simulating the Winner

Under preferential voting, finishing first on first preference votes does not guarantee victory—especially when the remaining ballots sit with ideologically distant parties. For ON to defeat a COAL candidate in a seat like Bass, they must typically outpoll the Coalition by roughly 10% on first preferences, because eliminated ALP and GRN voters break heavily for the Coalition over One Nation.

In each simulation, the model eliminates minor parties down to the final two contenders in each seat and applies Two-Candidate Preferred (TCP) preference flows to determine the winner. TCP flows per seat are estimated using the 2025 federal election results. Because 2026 features TCP matchups rarely seen in past Victorian elections—particularly One Nation finishing in the top two or three—these baselines are augmented with fresh preference data from the 2026 South Australian election, alongside the Secret Harbour and Farrer by-elections.

Eliminated
TCP Matchup
Avg. Flow
Labor
COAL
v
ON
74 - 26
Greens
80 - 20
CoalitionSH
ALP
v
ON
31 - 69
GreensSH
93 - 7
One Nation
ALP
v
COAL
25 - 75
Greens
88 - 12
Coalition
ALP
v
GRN
73 - 27
One Nation
71 - 29
Augmented with: SH Secret Harbour
Of statistical interest
Rather than simulating a full distribution of preferences across all n! possible elimination orders of n candidates, the Upper House Minor Flow Model returns to collapse minor-party votes into the final 4CP, and then 3CP, candidate set. Once the final TCP pair is determined, district-level preference flows to the TCP parties are imputed from the overlapping federal divisions at the 2025 federal election. For One Nation matchups, the model uses a Federal Senate–anchored projection of the two 2025 ON–COAL and ON–ALP contests (Maranoa and Hunter) onto Victorian federal electorates, adjusted toward flows observed or estimated in the 2026 South Australian election, Secret Harbour, and Farrer by-elections (see analyses by Antony Green and Ben Raue). Stochastic preference uncertainty is injected at both the statewide (σjoint = 3%) and individual candidate (σind = 2%) levels—values optimised via leave-one-out backtesting across the 2010–2022 Victorian elections to tightly calibrate predicted win probabilities across all candidates against the observed win rates.

Farrer by-election

Projected First Preference Votes

Violin Icon
Violin Plot
Shows vote distribution.
Fatter = more likely.
Taller = higher vote share.
TCP:
Understanding the Projections
The violins show the first preference vote distribution across 10,000 simulations of the Farrer by-election. Fatter areas indicate the most likely outcomes.
Hover to see exact percentages, or click Simulate to view sample results as white, clickable dots on the violins.
See the workings of the model here.

Chances of Winning

One Nation is the favorite, winning 50% of the 10,000 simulations. The Independent Michelle Milthorpe is the second favourite, winning 36%, while the Liberals (14%) and Nationals (0.6%) are the only other parties with statistically plausible paths to victory.
Paths to Victory
One Nation holds a clear advantage in the most likely Two-Candidate Preferred (TCP) matchup against Michelle Milthorpe. However, Milthorpe is strongly favoured against either the Liberals or Nationals if One Nation falls to third. The Liberals' clearest path relies on leapfrogging the Independent to face One Nation in the final count.
View Details of TCP Matchups
TCP Matchup
FrequencyFREQ.
Average TCP
Proportion WonPROP. WON
ON
v
IND
57.5%
53.8 - 46.2
73%
IND
v
LIB
21.9%
55.1 - 44.9
90%
ON
v
LIB
18.4%
48.6 - 51.4
62%
IND
v
NAT
1.3%
56.0 - 44.0
93%
ON
v
NAT
0.9%
50.0 - 50.0
52%
Note on Average TCP: The 'Average TCP' column above calculates the conditional Two-Candidate Preferred vote: the average TCP in the simulations where this specific matchup occurs. This excludes cases where either candidate was eliminated early.
Show Polling Place Projection Map
Note on Projections: These booth-level figures are modeled average first preference vote estimates solely based on the 2025 vote in each booth. They do not account for hyper-local candidate factors, and carry substantial uncertainty, so they should only be treated as a baseline for what the model's average by-election result would look like.
View Win Probabilities Trend

Win Probabilities over Time

Note on Trend: The official candidate field was declared on April 14. To provide a continuous timeline from the beginning of 2026, simulations prior to this date operate in hindsight—retroactively applying the final 12-candidate field to calculate what the race would have looked like had the true slate of parties contesting been known from the start.
Looking Ahead: Model Evaluation & State elections A full evaluation of the First Preference model's performance will be published here once the official results are finalised in June. Following the by-election, analysis and modelling will turn to the 2026 Victorian state election.

Deconstructing the Model

Tracking how incumbency, party changes, and polling swings shape each party's final projected vote.
Party
2025
Result
Incumbency
Adjust.
Parties
Contesting
National
Poll Swing
Labor
Realloc.
Model
Average
Seat Poll
Avg
Final
Average
One Nation
6.6%
+1.1
+0.1
+20.6
+2.2
30.5%
31.6%
31.3%
Ind. Milthorpe
20.0%
+1.4
-0.7
-0.4
+7.5
27.8%
30.2%
29.5%
Liberal
43.4%
-4.8
-8.6
-9.4
+0.5
21.1%
19.1%
19.7%
National
0.0%
-
8.5%
-2.6
+0.1
6.0%
7.0%
6.4%
Greens
4.9%
+0.3
-0.7
-1.1
+1.6
5.1%
4.5%
4.6%
Legalise Cannabis
0.0%
-
3.3%
-1.4
+1.1
3.0%
2.0%
2.1%
Shooters F&F
3.5%
+0.3
-0.7
-1.3
+0.1
1.9%
1.9%
2.0%
Family First
2.1%
+0.3
-0.3
-0.9
-
1.2%
1.3%
1.2%
People's First
2.0%
+0.1
-0.2
-0.9
-
1.1%
1.2%
1.2%
Ind. Woodward
0.0%
-
0.9%
-
+0.1
1.0%
0.6%
0.9%
Ind. Pappin
0.0%
-
0.9%
-
+0.1
1.0%
0.6%
0.9%
Sustainable Australia
0.0%
-
0.4%
-0.2
+0.1
0.4%
0.1%
0.2%
Labor
15.1%
+1.0
-0.3
-2.4
-13.4
0.0%
0.0%
0.0%
Trumpet of Patriots
2.4%
+0.2
-2.6
-
-
0.0%
0.0%
0.0%

1. Liberals' Lost Incumbency Advantage

With the departure of the outgoing Liberal leader Sussan Ley—an eight-term incumbent in a rural electorate—the party loses an estimated 4.8% personal vote advantage. This penalty is calculated by comparing the party’s primary vote in the House of Representatives against their Senate vote (as Senate voting is heavily party-based, it provides a clean baseline free from local candidate effects). While the 2025 data was clouded by a strong independent candidate (Michelle Milthorpe), the 2022 election provides a perfect real-world check: without a major independent on the ballot, Sussan Ley's personal vote gap was a similar 4.3%.

2025 Result
Incumbency Change
Liberal
43.4%
-4.8

2. Entering & Exiting the Field

Twelve candidates are contesting the Farrer by-election, up from nine in 2025. To account for this larger field, the model uses 2025 Senate results to reallocate votes away from non-contesting parties TOP and establish baselines for new minor party entrants LCPSAP . For the major new additions to the ballot, the NAT vote is carved directly out of the broader Coalition baseline, while the two new independents are injected as 'generic' independents via a separate model.

Party Changes
Trumpet of Pat.
-2.5
Leg. Cannabis
+3.3
Sust. Australia
+0.4
Nationals
+8.5
Ind. Pappin
+0.9
Ind. Woodward
+0.9

3. Choosing the Right Polling Swing

Applying national polling shifts to a single electorate requires the right tool:

  • Uniform Swing applies the flat shift in national polls since the last election (e.g., ON shifting from 6.4% to 23.7%), but it fails to account for a party's local strength. In Farrer, the COAL baseline (38.4%) is substantially higher than their 31.8% national vote in 2025, while GRNandLAB are much lower.
  • Proportional Swing respects the local baseline through scaling it by the national swing, but it breaks the fundamental reality that total swings must sum to exactly zero (meaning the final vote total won't be 100%).
  • The Additive Log-Ratio ( ALR ) transformation solves both problems. It scales shifts relative to Farrer's local baseline while mathematically locking the final total to exactly 100%. This methodology was accurate in capturing how One Nation's surge during the recent South Australian election translated to their votes in individual seats.
*Note on the Greens: The GRN polling swing shown in the overall matrix includes a historical by-election penalty, which for Farrer is estimated to be 1.1%.
Swing
Farrer
Vote
Unif.
Prop.
ALR
COAL
38.4%
-9.6
-11.6
-12.3
LAB
15.8%
-4.8
-2.2
-2.6
GRN
4.5%
+0.4
+0.1
+0.0
ON
7.8%
+17.3
+21.0
+20.3
IND
22.6%
0.0
0.0
-0.6
OTH
10.9%
-3.3
-4.6
-4.8
TOT.
100%
0.0
+2.7
0.0

4. Labor Voters' Strategic Reallocation

Labor is not fielding a candidate in the Farrer by-election, so their adjusted baseline vote is shattered and redistributed into the remaining parties. Similar cases from historical by-elections suggest the primary beneficiary is the leading independent, absorbing an estimated 56% of the orphaned vote. On average, 21% cascades to ideologically aligned minor parties GRNLCPSAP. These estimates rely on only 6 by-elections since 1990 without a Labor candidate, but with a prominent Independent, meaning Labor's reallocation is incorporated with significant uncertainty.

Poll-adjusted Vote
ALP Realloc.
Labor
13.5%
-13.5
Ind. Milthorpe
20.3%
+7.5
Greens
3.4%
+2.9
Leg. Cannabis
1.8%
+0.1
Sust. Australia
0.2%
+0.0

5. The Final Projection: Blending National and Seat Polls

The final projection blends the information from the national polling trend with that of recent seat polls of Farrer using dynamic Bayesian weighting:

  • Historical Calibration: How much should we trust a local poll over the national polling trend? By testing the model against a decade of past elections and by-elections to minimize historical prediction errors, the model is tuned to an optimal blend. This blend is used to inject polling error uncertainty using multivariate normal distributions in ALR space, specifically accounting for historical by-election polling errors being ~14% greater than those in general elections.
  • The Farrer Blend: Currently, Farrer has two published seat polls, and the latest national poll was taken 7 days prior to the by-election. The historical calibration gives the local seat polls a dominant average weight of 55% in the final calculation, leaving 45% for the National-polling-based model. The weights differ per party, which is why some parties' final average is closer to that of the seat polls than others.
  • Minor Party Allocation: Pollsters don't prompt respondents for every minor party. To fill these gaps, the model uses Dirichlet distributions to split the 'Other' vote into its constituent parties, as well as the 'Coalition' vote between LIB and NAT. Here, the model applies a probabilistic mix: 52% of the simulations draw their split from the seat polls, while the other 48% draw from the historical baseline.
Nat-Poll Model Avg
Seat Poll Avg
Final Blend
ON
28.9
23.8
25.9
IND
20.3
25.0
23.0
LIB
21.1
19.1
19.9
NAT
6.0
7.0
6.6

6. Simulating the Winner

In our preferential voting system, the winner is determined by sequentially eliminating the lowest-polling candidate and reallocating their voters to the remaining parties. To avoid modelling each possible order of elimination - with 12 candidates there are over 479 million of them! - the model assumes the two parties with the highest first preference vote will form the final Two-Candidate Preferred (TCP) matchup.

There is one important exception: if neither the Liberals nor the Nationals place in the top two. Because their voters typically heavily preference one another, the model checks if reallocating 70% of the smallest Coalition party's vote to the other would mathematically push them into the top two. If so, they replace the initial second-placed party.

Finally, significant shifts in the political landscape since the 2025 election mean historical TCP preference flows may be inaccurate. The model augments Farrer's baseline TCP flows with recent data from the 2026 South Australian Election (see Ben Raue's analysis).

Eliminated
TCP Matchup
Est. Flow
Liberal
National
COAL
v
IND
85 - 15
One Nation
COAL
v
IND
53 - 47
Independent
COAL
v
ON
75 - 25
Coalition
IND
v
ON
33 - 67

Limitations of the Model

This by-election model is at its core probabilistic: it aims to describe the uncertainty inherent to the by-election outcome rather than to 'predict' an exact result. When tested on historical elections/by-elections, its average error (MAE) is ~3% per party - that's how much you should expect each party's average to be wrong by! Its real power is its ability to interpret which kinds of extreme outcomes are plausible.

Some aspects are captured imperfectly:

  • National polling constraints: National polling poorly captures Independent vote intentions. With the combined minor party and IND vote dropping from 15% to less than 11% since 2025, applying this swing directly would artificially suppress the vote of a prominent local Indendent, Michelle Milthorpe. There are no perfect solutions; the model fixes the national IND baseline at 7.2% (as in 2025), forcing the remaining minor parties to bear the brunt of the swing. If this forces the minor party total below a crude 5% threshold, a small rebalancing adjustment is applied to all parties to stabilise the simulation.
  • Use of Senate results: A fully Bayesian probabilistic model would incorporate uncertainty about each component being estimated. The current model takes some shortcuts in the early stages, with Senate results used for reallocating votes to new parties being accepted as direct fill-ins for votes in the House. While it could be claimed that these newly-contesting parties should be treated with greater uncertainty, the model is none the wiser.
  • Deterministic parameters: The model uses several sub-models to estimate its components, such as incumbency advantage, polling error covariance matrices, or the vote ratio between the Coalition parties. Ideally, the uncertainty in the estimation should be incorporated to make the estimated parameters themselves random. At present, most of these are taken to be deterministic, although the primary election-specific sources of uncertainty in polling error and minor-party splits are rigorously dealt with.
  • ALR transformation bias: Using an ALR transformation to apply polling swings solves many mathematical problems, but carries a subtle statistical cost: one party must be chosen as the 'reference' category - in this case COAL - and made to suffer an artificial slight reduction in its simulated uncertainty. Hence, it is possible that the model is underestimating the variability of the Coalition parties' vote, however my impression is that the effect is on the scale of less than 1% in win probability.

The polling data fed to the Farrer by-election model (adjusted for undecideds) can be downloaded here: Seat Poll (CSV), National Polls (CSV).

See the projections for Farrer at the 2025 election

64.5%

Labor Majority

28.3%

Labor Lead

0.8%

Tie

4.5%

Coalition Lead

1.9%

Coalition Majority

Note on Model Evaluation
The First Preference model was frozen after the final update on Friday, May 2nd. To see how the model will be evaluated on the results of the election, click here. Check back here after the results are in for a full evaluation.
Violin Icon
Violin Plot
Shows the distribution of seats each party might win.
Fatter = more likely. Taller = more seats. Isolated dots = rare outcomes.
Simulations Explained
Each point on the plot represents one of 1000 Simulations of the election outcome.
White text shows Average Seats won per party across the 1000 Simulations.
Black dots represent results of Seats Won in the 2025 election.
Hover or Touch the Violins to see Seat numbers.
For more details, visit the Methodology section.
See individual seat win probabilities below.
Formation of Government
Likely Outcomes:
Labor Majority – Labor Minority
Plausible Outcomes:
Labor Landslide     Hung Parliament
Last model update: 2 May, 2025
National First Preference Vote Shares By Party

National First Preference Vote Share By Party

PartyInterval of 95% ProbabilityMean
Coalition30.3% – 38.1%34.1%
Labor25.5% – 37.2%31.4%
Greens9.8% – 15.8%12.8%
One Nation5.4% – 11.3%7.8%
Independent5.1% – 8.5%6.5%
Trumpet of Patriots0.8% – 2.7%1.9%
Legalise Cannabis0.9% – 1.9%1.3%
Libertarian0.8% – 1.6%1.1%
Family First0.7% – 1.4%1.0%
Animal Justice0.2% – 0.5%0.3%
Katter's Australian0.1% – 0.4%0.3%
Shooters Fishers & Farmers0.1% – 0.4%0.2%
Centre Alliance0.1% – 0.4%0.2%
People's First0.1% – 0.3%0.2%
Christians0.1% – 0.3%0.2%
Citizens Party0.1% – 0.3%0.2%
FUSION0.1% – 0.2%0.1%
Victorian Socialists0.0% – 0.2%0.1%
Socialist Alliance0.0% – 0.1%0.1%
Indigenous-Aboriginal0.0% – 0.1%0.1%
HEART0.0% – 0.1%0.0%
Great Australian Party0.0% – 0.0%0.0%
Australian Democrats0.0% – 0.0%0.0%
Note on Probabilities
This model is probabilistic: a party winning in 800 of 1000 simulations does not mean it is 'predicted' to win.
It is simply that that party winning is the likelier outcome given the model's structure.
Take 5 seats where the favourite is given about 80% chance of winning:
Hughes (COAL), Chisholm (LAB), Wentworth (IND), Ryan (GRN), Mayo (OTH).
On average, one of these favourites is expected to lose.

Number of Wins out of 1000 Simulations

ElectorateLaborCoalitionGreensIndependentOne NationOther
Adelaide9875800
Aston2747260000
Ballarat99460000
Banks3376630000
Barker19990000
Barton97326100
Bass4565421010
Bean9362161
Bendigo985130110
Bennelong641357200
Berowra2395152100
Blair870119083
Blaxland981190000
Bonner192807100
Boothby91777600
Bowman1088900020
Braddon759101860
Bradfield19630035100
Brand9990010
Brisbane21911566600
Bruce808190002
Bullwinkel863136100
Burt99540010
Calare12334065310
Calwell968281021
Canberra96313510
Canning415583011
Capricornia708640660
Casey410502127510
Chifley99080020
Chisholm7752223000
Clark9009910
Cook12988000
Cooper843015700
Corangamite942571000
Corio99820000
Cowan939610000
Cowper0749025100
Cunningham9942400
Curtin3376062100
Dawson39740230
Deakin33659916400
Dickson17973808300
Dobell91486000
Dunkley921780100
Durack2247650110
Eden-Monaro934660000
Fadden259680070
Fairfax4942040140
Farrer097902001
Fenner997210
Fisher9878010940
Flinders3592504000
Flynn19078600240
Forde22176400150
Forrest231645810970
Fowler5120048800
Franklin892324810
Fraser9951400
Fremantle8750112310
Gellibrand98711200
Gilmore25669604800
Gippsland0998011
Goldstein0281071900
Gorton93759022
Grayndler99307000
Greenway960400000
Grey13797018460
Griffith765287200
Groom392106790
Hasluck98812000
Hawke93856051
Herbert10985050
Hindmarsh981190000
Hinkler15922053100
Holt95049001
Hotham9918100
Hughes174826000
Hume33897052180
Hunter727269031
Indi0116088400
Isaacs97425100
Jagajaga978157000
Kennedy1390011949
Kingsford Smith9954100
Kingston9982000
Kooyong0322067800
Lalor99270001
La Trobe31968010
Leichhardt361615001014
Lilley97721200
Lindsay1068930010
Lingiari705295000
Longman301694050
Lyne14787019315
Lyons4155735052
Macarthur96530050
Mackellar0307069300
Macnamara6798313000
Macquarie93961000
Makin9937000
Mallee0999010
Maranoa0994060
Maribyrnong996310
Mayo18738140761
McEwen761236120
McMahon98413003
McPherson6805117792
Melbourne660934000
Menzies4395554101
Mitchell19980100
Monash209336045311
Moncrieff295104340
Moore296436026800
Moreton8041346200
Newcastle99208000
New England09940600
Nicholls3990070
O'Connor117878032
Oxley984141100
Page098801101
Parkes098500015
Parramatta8061940000
Paterson6982990021
Pearce9908011
Perth9812170
Petrie199801000
Rankin96632020
Reid932680000
Richmond754122231010
Riverina195702868
Robertson929710000
Ryan3211585300
Scullin9962020
Shortland942560020
Solomon9283004200
Spence99520030
Sturt34756328800
Swan98514100
Sydney98601400
Tangney90199000
Wannon5728026700
Warringah083091700
Watson99280000
Wentworth218108170
Werriwa915800005
Whitlam917801020
Wide Bay595504360
Wills731026900
Wright49450510