The Run Home

[Bring-Back-Powell 14,865]

12 minutes ago, Sydney_Demon said:

I know Fox Footy's analysis goes a bit deeper than just looking at average percentages of opponents for the rest of the season, but pure averages can be very misleading. In Melbourne's case we have 11 games to go but 6 of them are at the MCG and only 3 interstate (1 of which is a home game in Alice Springs against GWS, 1 is against North Melbourne in Hobart, with the final round of the season against Sydney who probably won't have much to play for).

We have no games in Western Australia, South Australia & Queensland. Other teams will have tough away games against Adelaide, Port Adelaide, Brisbane, Gold Coast & Fremantle . The average percentage of opponents for Melbourne is dragged up by games against Collingwood & Brisbane, both of which we play at the MCG.

I'd say, especially with Western Bulldogs & Adelaide losing yesterday we are very-well placed. Wheelo has us a 96.4% chance Top 8, a 73.5% chance Top 4 as of this morning:

https://www.wheeloratings.com/afl_simulations.html
 

Agree with most of that except for the Sydney bit.

Sydney are right in the thick of the top 8 action and may need to beat us in the last game to make finals.

And if they are out of the race, it will likely be Buddy’s farewell game and emotions will be very high at the SCG. Therefore I expect it to be a very tough assignment.

[Sydney_Demon 1,175]

18 minutes ago, Bring-Back-Powell said:

Agree with most of that except for the Sydney bit.

Sydney are right in the thick of the top 8 action and may need to beat us in the last game to make finals.

And if they are out of the race, it will likely be Buddy’s farewell game and emotions will be very high at the SCG. Therefore I expect it to be a very tough assignment.

Fair point about the Sydney game. Its extremely likely they'll have a much better team on the park than they've had in recent times, but we did comprehensively beat them earlier in the year at the MCG when they were near full-strength. One player they are missing who I think was crucial to them last year (especially against Melbourne) was Sam Reid and he won't be back. Neither will Paddy McCartin. Sydney are currently 12th (with a game in hand) with a tough draw and I would predict the 11 teams above them are all morely likely to stay above them than not. They also have 3 teams below them in Richmond, Carlton & GWS who will their fair share of games for the rest of the year. It all depends on whether the Swans have something to play for. I don't think they'll be competing for a finals spot in Round 24 but a long way to go. Will the Buddy factor make enough of a difference?

[Sydney_Demon 1,175]

Swans current injury list:
 

Joel Amartey	Hamstring	2-4 weeks
Peter Ladhams	Ankle	TBC
Paddy McCartin	Concussion	Season
Tom McCartin	Concussion	TBC
Logan McDonald	Ankle	7 weeks
Callum Mills	Calf	2-4 weeks
Luke Parker	Suspension	Round 14
Dane Rampe	Neck	1-2 weeks
Sam Reid	Hamstring	Season
Matt Roberts	Knee	4-6 weeks
Marc Sheather	Foot	TBC

[Great Northern Summer 708]

1 hour ago, Sydney_Demon said:

I know Fox Footy's analysis goes a bit deeper than just looking at average percentages of opponents for the rest of the season, but pure averages can be very misleading. In Melbourne's case we have 11 games to go but 6 of them are at the MCG and only 3 interstate (1 of which is a home game in Alice Springs against GWS, 1 is against North Melbourne in Hobart, with the final round of the season against Sydney who probably won't have much to play for).

We have no games in Western Australia, South Australia & Queensland. Other teams will have tough away games against Adelaide, Port Adelaide, Brisbane, Gold Coast & Fremantle . The average percentage of opponents for Melbourne is dragged up by games against Collingwood & Brisbane, both of which we play at the MCG.

I'd say, especially with Western Bulldogs & Adelaide losing yesterday we are very-well placed. Wheelo has us a 96.4% chance Top 8, a 73.5% chance Top 4 as of this morning:

https://www.wheeloratings.com/afl_simulations.html
 

Also has us as Premiership favourite by some way.
 

I get how it uses some formula to predict finishing positions after H&A season based on opponents and current ladder position, but how on earth does it determine likelihood of finals matches outcomes?

[Sydney_Demon 1,175]

4 minutes ago, Glorious Day said:

Also has us as Premiership favourite by some way.
 

I get how it uses some formula to predict finishing positions after H&A season based on opponents and current ladder position, but how on earth does it determine likelihood of finals matches outcomes?

Refer my other post under AFL Stats Resource: Wheelo Ratings. There's nothing as simplistic as using a formula. Multiple simulations are run of every game for the reminder of the season with the outputs of those simulations aggregated. I don't really understand your point about predicting outcomes of finals games. The same process I assume as for H&A games.  I'm hoping @WheeloRatings might provide some clarity re chances of a Melbourne Premiership, but I agree with you it doesn't make a lot of sense at face value. 

[old55 23,626]

I'll let @WheeloRatings give the definitive answer but I imagine the simulations use the Team Ratings as a basis, including for finals (which in each season simulation will have finalists allocated based on H&A results) and that's how the results are calculated.  Currently we have the highest overall Team Rating, 1st for Defence and 3rd for Attack.

[Jaded No More 68,967]

Again results have gone our way this week with Dogs and Adelaide losing. I think if we beat the Saints we will go a long way to securing a top 4 spot. 
Dogs and Saints have a very easy run home, but if we can get big wins against North and Hawks again we can ensure they can’t catch us on percentage. 
If you finish top 4 anything can happen. Finish 4th and we most likely play Pies in week one at the G. Finish 3rd and we probably end up in an away final in Brisbane or Adelaide. Finish 2nd, and we are laughing. 

[Sydney_Demon 1,175]

14 minutes ago, old55 said:

I'll let @WheeloRatings give the definitive answer but I imagine the simulations use the Team Ratings as a basis, including for finals (which in each season simulation will have finalists allocated based on H&A results) and that's how the results are calculated.  Currently we have the highest overall Team Rating, 1st for Defence and 3rd for Attack.

I'm surprised that we are ranked so highly in those team rankings. Having said that, I think that our losses to Port & Fremantle and the closeness of our game against Carlton were significantly due to poor goal kicking. Assuming a reversion to the mean I'm hoping that wil turn around!

8 minutes ago, Jaded No More said:

Again results have gone our way this week with Dogs and Adelaide losing. I think if we beat the Saints we will go a long way to securing a top 4 spot. 
Dogs and Saints have a very easy run home, but if we can get big wins against North and Hawks again we can ensure they can’t catch us on percentage. 
If you finish top 4 anything can happen. Finish 4th and we most likely play Pies in week one at the G. Finish 3rd and we probably end up in an away final in Brisbane or Adelaide. Finish 2nd, and we are laughing. 

For me St Kilda are currently not trending well but will go ahead of us if they beat Sydney and we lose to Collingwood. Both those games are 50/50 I think. Our percentage is a big factor because we should be ahead of whoever finishes on the same number of wins as us at season's end. I agree with you that finishing 2nd would be brilliant as we would be playing either Port or Brisbane in Melbourne and then if we win Port or Brisbane again in a Prelim. Also I don't think Brisbane or Port are unbeatable away in a Qualifying or Preliminary Final.

[WheeloRatings 1,163]

3 hours ago, Glorious Day said:

Also has us as Premiership favourite by some way.
 

I get how it uses some formula to predict finishing positions after H&A season based on opponents and current ladder position, but how on earth does it determine likelihood of finals matches outcomes?

 

2 hours ago, Sydney_Demon said:

Refer my other post under AFL Stats Resource: Wheelo Ratings. There's nothing as simplistic as using a formula. Multiple simulations are run of every game for the reminder of the season with the outputs of those simulations aggregated. I don't really understand your point about predicting outcomes of finals games. The same process I assume as for H&A games.  I'm hoping @WheeloRatings might provide some clarity re chances of a Melbourne Premiership, but I agree with you it doesn't make a lot of sense at face value. 

I'll respond to your other post.

2 hours ago, old55 said:

I'll let @WheeloRatings give the definitive answer but I imagine the simulations use the Team Ratings as a basis, including for finals (which in each season simulation will have finalists allocated based on H&A results) and that's how the results are calculated.  Currently we have the highest overall Team Rating, 1st for Defence and 3rd for Attack.

@old55 is correct that the simulations use the Team Ratings as a basis, and my model has Melbourne currently ranked #1. I agree that it doesn't necessarily seem correct, but in my defence (or my model's defence), the aggregate of a number of models on https://squiggle.com.au/power-rankings/ also sees Melbourne ranked #1 and Collingwood #4.

The model simulates the remainder of the season 50,000 times, including simulating the finals based on the outcomes for a particular simulated season. Basically, for a given simulation, I simulate the remainder of the home and away season and then simulate the final series based on the final ladder positions of that particular simulation. I do that 50,000 times and aggregate the results. Melbourne's premiership favouritism is largely due to them being ranked #1.

[Adam The God 30,306]

51 minutes ago, WheeloRatings said:

 

I'll respond to your other post.

@old55 is correct that the simulations use the Team Ratings as a basis, and my model has Melbourne currently ranked #1. I agree that it doesn't necessarily seem correct, but in my defence (or my model's defence), the aggregate of a number of models on https://squiggle.com.au/power-rankings/ also sees Melbourne ranked #1 and Collingwood #4.

The model simulates the remainder of the season 50,000 times, including simulating the finals based on the outcomes for a particular simulated season. Basically, for a given simulation, I simulate the remainder of the home and away season and then simulate the final series based on the final ladder positions of that particular simulation. I do that 50,000 times and aggregate the results. Melbourne's premiership favouritism is largely due to them being ranked #1.

And sorry, I've missed this. How are the team ratings calculated?

[WheeloRatings 1,163]

2 hours ago, A F said:

And sorry, I've missed this. How are the team ratings calculated?

Firstly, a team's rating is the sum of the its attacking rating and its defensive rating. The attacking rating is based on whether teams score above or below average and the defensive metric is based on whether teams concede above or below average. However, instead of using a team's actual score, I calculate an "adjusted" score using a weighted average of their actual score and the score they would have kicked had they kicked at an expected accuracy. This is to account for the luck factor in goalkicking.

The attacking and defensive ratings are updated following each match based on these "adjusted" scores compared to the expected scores. If a team's adjusted score is higher than expected, their attack rating increases (and vice versa). If their opponent's adjusted score is higher than expected, their defensive rating decreases (and vice versa).

The expected scores take into account the teams pre-game ratings and any venue advantage.

I have a page on my site which provides some more detail, noting that I have never actually got around to finalising it:

https://www.wheeloratings.com/afl_methodology.html

In addition, teams carry over ~65% of their rating from the previous season as there's a general regression to the mean from one season to the next.

[Lucifers Hero 40,010]

Also posted this in the Contenders/Pretenders thread .

Recently I posted about remaining 8-pt games and a teams opportunity to gain % boosts ie most likely vs the bottom 6 teams. 

Summary below.  Right hand column shows total 8pt games.  Not very scientific but gives a good picture of the Run Home.

The rationale of the commentators saying we have the hardest draw is flawed imv!!  I reckon we have the easiest draw of contending teams.

Setting aside that any team can win on any given day,

• The 'easiest' draw is Demons, Saints and Bulldogs with six 8pt games only two of which are vs top 4 teams. 
• The 'hardest' draw is Geelong with 10 of the last 11 being 8pt games, 5 of which are vs top 4 teams.  They may not even make the 8 if Sydney and Freo get on a roll. 
• Pies will have to fight out the season with games vs Cats, Lions and Bombers in last 3 rounds.
• Similarly Lions have a challenging last 4 games.
• Cats will have to fight in every game; no resting of players this year.
• Top 4 locked in.  Other contenders not convincing of late.
• Dees will probably be top 2.  We play 5 of the bottom 6 sides to see out the season. (6 of the bottom 7 if we count Sydney).

Suns are the team to watch.  They won't make the 8 but they will have a very big say in who does!!

A very different run home for us than 2022.  We will have the chance to manage injured players enabling us to be one of the 'freshest' contenders going into the finals.

Edited June 4, 2023 by Lucifers Hero

[Demons11 7,032]

On 5/29/2023 at 8:45 AM, Gawndy the Great said:

I thought the run home is worth discussing given that we are close to midway through the year. 12 wins gets you into finals and 16 into the top 4. There is an extra game this year so it’ll be interesting whether this effects the required wins to make it into the 8/4.  We are are 7-4 meaning there is a bit of work to do. Where do our wins come from. My  views below based on current form gets us to 14-9 which ends up being 6-7 territory. There is probably a few more wins in there but I have not the confidence to call em right now.

Geelong are gettable , Adelaide and Lions on our home deck we’d have to have real hot shot… but that is assuming we get our ducks in order. We get Petty and Oliver back after the bye so no excuses.

Melbourne v Carlton - W

Melbourne v Collingwood - L

Geelong Cats v Melbourne - L

Melbourne v GWS Giants - W

St Kilda vs Melbourne - L

Melbourne v Brisbane Lions - L

Melbourne v Adelaide Crows - L

Richmond v Melbourne - W

North Melbourne v Melbourne - W

Carlton v Melbourne - W

Melbourne v Hawthorn - W

Sydney Swans v Melbourne - W

 What a Debbie downer outlook.

we smack saints, crows and are a massive shot in all games we play 

[bush demon 2,047]

21 hours ago, Bring-Back-Powell said:

Agree with most of that except for the Sydney bit.

Sydney are right in the thick of the top 8 action and may need to beat us in the last game to make finals.

And if they are out of the race, it will likely be Buddy’s farewell game and emotions will be very high at the SCG. Therefore I expect it to be a very tough assignment.

Buddy is moving around ATM like Talos from Jason and the Argonauts..

https://youtu.be/dD9qc44oMqU

Edited June 4, 2023 by bush demon
Add link.

[tiers 2,842]

It will all be to no avail if we don't hurry up and find a settled, reliably performing forward line.

Using Maxie and Grundy as key forward targets will not get it done in finals and JVR is proving to be a better roving CHF type than FF type.

[old55 23,626]

1 hour ago, tiers said:

It will all be to no avail if we don't hurry up and find a settled, reliably performing forward line.

Using Maxie and Grundy as key forward targets will not get it done in finals and JVR is proving to be a better roving CHF type than FF type.

#35

[layzie 33,872]

3 hours ago, Lucifers Hero said:

Also posted this in the Contenders/Pretenders thread .

Recently I posted about remaining 8-pt games and a teams opportunity to gain % boosts ie most likely vs the bottom 6 teams. 

Summary below.  Right hand column shows total 8pt games.  Not very scientific but gives a good picture of the Run Home.

The rationale of the commentators saying we have the hardest draw is flawed imv!!  I reckon we have the easiest draw of contending teams.

Setting aside that any team can win on any given day,

• The 'easiest' draw is Demons, Saints and Bulldogs with six 8pt games only two of which are vs top 4 teams. 
• The 'hardest' draw is Geelong with 10 of the last 11 being 8pt games, 5 of which are vs top 4 teams.  They may not even make the 8 if Sydney and Freo get on a roll. 
• Pies will have to fight out the season with games vs Cats, Lions and Bombers in last 3 rounds.
• Similarly Lions have a challenging last 4 games.
• Cats will have to fight in every game; no resting of players this year.
• Top 4 locked in.  Other contenders not convincing of late.
• Dees will probably be top 2.  We play 5 of the bottom 6 sides to see out the season. (6 of the bottom 7 if we count Sydney).

Suns are the team to watch.  They won't make the 8 but they will have a very big say in who does!!

A very different run home for us than 2022.  We will have the chance to manage injured players enabling us to be one of the 'freshest' contenders going into the finals.

Great chart.

Our run is still not easy but no-where near as hard as the media lead people to believe at the moment. 

[Sydney_Demon 1,175]

On 5/29/2023 at 8:45 AM, Gawndy the Great said:

I thought the run home is worth discussing given that we are close to midway through the year. 12 wins gets you into finals and 16 into the top 4. There is an extra game this year so it’ll be interesting whether this effects the required wins to make it into the 8/4.  We are are 7-4 meaning there is a bit of work to do. Where do our wins come from. My  views below based on current form gets us to 14-9 which ends up being 6-7 territory. There is probably a few more wins in there but I have not the confidence to call em right now.

Geelong are gettable , Adelaide and Lions on our home deck we’d have to have real hot shot… but that is assuming we get our ducks in order. We get Petty and Oliver back after the bye so no excuses.

Melbourne v Carlton - W

Melbourne v Collingwood - L

Geelong Cats v Melbourne - L

Melbourne v GWS Giants - W

St Kilda vs Melbourne - L

Melbourne v Brisbane Lions - L

Melbourne v Adelaide Crows - L

Richmond v Melbourne - W

North Melbourne v Melbourne - W

Carlton v Melbourne - W

Melbourne v Hawthorn - W

Sydney Swans v Melbourne - W

12 wins might get a team into the 8 as long as they have a decent percentage. Geelong will probably make it with that many wins as their current percentage is 118.9%, 13 will definitely get you in. Similarly I think 16 wins would get Geelong, Brisbane or Melbourne in to the Top 4 given their respective percentages, 17 in general. I think your take is pretty negative. At this stage I think the only game where Melbourne's opposition would be favourites is KB. Geelong game 50/50. Games against St Kilda, Brisbane & Adelaide we will be favourites. I'm not saying we definitely win them all, but 14 9 is definitely a worst case scenario.

[titan_uranus 25,020]

As @Lucifers Hero has posted above, we actually have a comparatively kind draw.

That stems largely from the six double-up opponents we were given. Of them, two are North and Hawthorn, the next two are Carlton and Richmond who pre-season were considered top 8 chances but right now are both more likely to be bottom 6. Then there's Sydney, a pre-season flag fancy who IMO can still challenge and by Round 24 could be well in the thick of it, and then Brisbane.

That's a much kinder slate of repeat games than would have been expected pre-season.

[ManDee 7,162]

I struggle when I see or hear "8 point game". There is no such thing. The outcomes are 0 points, 2 points or 4 points.

If you and your opponent start a round on 30 points the greatest difference after the round is 4 points.

My rant for the day 🙂 

[YearOfTheDees 3,266]

13 minutes ago, ManDee said:

I struggle when I see or hear "8 point game". There is no such thing. The outcomes are 0 points, 2 points or 4 points.

If you and your opponent start a round on 30 points the greatest difference after the round is 4 points.

My rant for the day 🙂 

AS much a I agree with you "all games are worth 4 point" Win on Monday and we are 8 points behind Collingwood  Lose and we are 16 points behind them.

[Sydney_Demon 1,175]

20 minutes ago, ManDee said:

I struggle when I see or hear "8 point game". There is no such thing. The outcomes are 0 points, 2 points or 4 points.

If you and your opponent start a round on 30 points the greatest difference after the round is 4 points.

My rant for the day 🙂 

i'm not sure why you're struggling with this. Technically you're correct but no-one is meaning that one team actually gets 8 points by winning the game. As you are aware, in your scenario, the 2 outcomes (assuming no draw) are Team 34, Opponent 30 or Team 30, Opponent 34. There's an 8-point difference here, +4 vs -4 (the -4 is the missed opportunity if you like). It was important in 2021 that our losses generally were against lower teams because although we missed the 4 points for a win it didn't affect our finishing position like it would have for a loss against a team we were actually competing for ladder position with (4-point game vs 8-point game).  

[Sydney_Demon 1,175]

1 hour ago, YearOfTheDees said:

AS much a I agree with you "all games are worth 4 point" Win on Monday and we are 8 points behind Collingwood  Lose and we are 16 points behind them.

Well, this doesn't exactly explain the difference betwen an 8-point game as it's meant commonly and a 4-point game, in the sense that your example above applies to every game. 8-point games imo can really only be between 2 teams legitimately competing for ladder position. We'll only know in hindsight which games were 8-point games. If Collingwood finish ahead of us despite losing on KB it won't be an 8-point game 😀 . 

Edited June 5, 2023 by Sydney_Demon
Typo

[Redleg 40,995]

8 hours ago, Lucifers Hero said:

Also posted this in the Contenders/Pretenders thread .

Recently I posted about remaining 8-pt games and a teams opportunity to gain % boosts ie most likely vs the bottom 6 teams. 

Summary below.  Right hand column shows total 8pt games.  Not very scientific but gives a good picture of the Run Home.

The rationale of the commentators saying we have the hardest draw is flawed imv!!  I reckon we have the easiest draw of contending teams.

Setting aside that any team can win on any given day,

• The 'easiest' draw is Demons, Saints and Bulldogs with six 8pt games only two of which are vs top 4 teams. 
• The 'hardest' draw is Geelong with 10 of the last 11 being 8pt games, 5 of which are vs top 4 teams.  They may not even make the 8 if Sydney and Freo get on a roll. 
• Pies will have to fight out the season with games vs Cats, Lions and Bombers in last 3 rounds.
• Similarly Lions have a challenging last 4 games.
• Cats will have to fight in every game; no resting of players this year.
• Top 4 locked in.  Other contenders not convincing of late.
• Dees will probably be top 2.  We play 5 of the bottom 6 sides to see out the season. (6 of the bottom 7 if we count Sydney).

Suns are the team to watch.  They won't make the 8 but they will have a very big say in who does!!

A very different run home for us than 2022.  We will have the chance to manage injured players enabling us to be one of the 'freshest' contenders going into the finals.

Do you make charts or graphs for a living?

I couldn't do it and wouldn't have the patience either.

Well done. 

[layzie 33,872]

32 minutes ago, Sydney_Demon said:

i'm not sure why you're struggling with this. Technically you're correct but no-one is meaning that one team actually gets 8 points by winning the game. As you are aware, in your scenario, the 2 outcomes (assuming no draw) are Team 34, Opponent 30 or Team 30, Opponent 34. There's an 8-point difference here, +4 vs -4 (the -4 is the missed opportunity if you like). It was important in 2021 that our losses generally were against lower teams because although we missed the 4 points for a win it didn't affect our finishing position like it would have for a loss against a team we were actually competing for ladder position with (4-point game vs 8-point game).  

One of the great footy metaphors!