Insights

footyhome's own ratings model, the betting market's view, and the long-run trends in the game. For analysis and discussion; footyhome does not give betting advice.

Next games: model against the market

Game	When	Model	Margin	Market	Gap
Sydney v Geelong	R1 · Wed, 17 Mar	56%	+4	–	–
Fremantle v Western Bulldogs	R1 · Thu, 18 Mar	74%	+18	–	–
Hawthorn v Brisbane	R1 · Fri, 19 Mar	52%	+1	–	–
Richmond v Carlton	R1 · Sat, 20 Mar	19%	-26	–	–
Adelaide v GWS	R1 · Sat, 20 Mar	68%	+13	–	–
St Kilda v Collingwood	R1 · Sat, 20 Mar	42%	-6	–	–
North Melbourne v Port Adelaide	R1 · Sun, 21 Mar	63%	+9	–	–
Melbourne v Essendon	R1 · Sun, 21 Mar	82%	+26	–	–
West Coast v Gold Coast	R1 · Sun, 21 Mar	36%	-10	–	–
Brisbane v Collingwood	R2 · Thu, 25 Mar	76%	+20	–	–
North Melbourne v Carlton	R2 · Fri, 26 Mar	35%	-11	–	–
Fremantle v Adelaide	R2 · Fri, 26 Mar	68%	+13	–	–
GWS v West Coast	R2 · Sat, 27 Mar	76%	+20	–	–
Gold Coast v Richmond	R2 · Sat, 27 Mar	81%	+25	–	–
Port Adelaide v Melbourne	R2 · Sat, 27 Mar	42%	-6	–	–
Essendon v Sydney	R2 · Sun, 28 Mar	19%	-26	–	–
Western Bulldogs v St Kilda	R2 · Sun, 28 Mar	61%	+8	–	–
Geelong v Hawthorn	R2 · Mon, 29 Mar	58%	+6	–	–

Market figures appear once bookmakers open their markets for these games. A large gap means the model and the market disagree; it is not a prediction that one of them is right.

Power rankings

footyhome Elo after each club's latest game. 1500 is average.

1Brisbane1688▼4 2Geelong1670▲26 3Fremantle1643▲85 4Sydney1639▲135 5Hawthorn1622▼26 6Carlton1574▲106 7Adelaide1569▲7 8Melbourne1544▲70 9Collingwood1531▼43 10Western Bulldogs1513▼84 11GWS1493▼87 12Gold Coast1471▼75 13St Kilda1444▲5 14North Melbourne1423▲45 15Port Adelaide1396▼40 16West Coast1275▲21 17Essendon1255▼83 18Richmond1253▼59

Arrows show the change since the start of the club's latest season.

How good is the model?

Winners picked: 67.4%
Average margin miss: 29.2
Games tested: 5,329

Since 2000, using only what was known before each game. Points near the diagonal mean the stated chances were honest: teams given 70% won about 70% of the time.

Points per game since 1925

Combined score of both teams, home-and-away games.

Home-team win rate

Share of home-and-away games won by the nominated home team (draws count half).

How the model works

Every club carries a rating, starting at 1500 in its first season. Before each game the gap between the two ratings, plus a home-ground allowance for interstate trips, gives each team its chance of winning. After the game the winner takes rating points from the loser: more for a big win, and more again for an upset.

Between seasons every rating drifts 15% of the way back towards average, because lists change. The settings were chosen by testing every season from 1990 to 2026 and keeping the combination whose stated chances best matched what actually happened.

The model knows nothing about injuries, selections or weather. Treat it as a well-calibrated starting point, not a tip sheet. Explore the raw numbers yourself in the stats explorer.