Analysis Season 2023 - Statistics and Analytics Thread

[Tom14]

Expected score ladder post round 10 - top for points for on expected score, third behind Pies and Dogs on expected score against by just 7 points. We've played 5 of the top 7 expected scores for as well (excluding us at #1).

 

[blynd_freddie]

As past time poster, long-time lurker. Thought it best if started throwing some posts at the board to see if they stick before doing an end of home & away preview post. Hopefully the below numbers stack up better than my total post numbers. Had planned to make this my 6,666 post but what the hell.

Thought it might be worth sharing/discussing after looking a little deeper into inside 50 differentials after noticing that the last 4 AFL premiers were ranked either #1 or #2 in that year with the highest inside 50 differential.

And at the top of i50 differential per game we have
#1 – Pies have an 8.9
#2 – Lions 7.8
#3 - Cats 7.7

The differentials of course simply reflect inside 50 For and Against. After 10 rounds in 2023 the AFL team average for inside 50’s is 52.6 per team per game,
The top three are;
#1 – Dees with 57.9
#2 – Pies 56.7
#3 – Lions 56.4

The lowest average opponents inside 50’s
#1 – Pies - with opponents held to an average of 47.8
#2 – Cats - 47.9
#3 – Lions - 48.6

So for discussion purposes let’s remove our round 1 game, where we were ambushed by Port leaving our 9 game differential average at 11.4 per game. For comparison, last year, Geelong had a differential of 12.6. [we were ranked 6th with 3.9]

Returning to the inside 50 numbers – the below table shows all teams average inside 50’s by rank for the first 10 rounds, their i50 versus Lions, that games variance to their average to date, our i50 in that game and our game differential. As things stand, we have been able to hold all but two opponents to LESS than their 2023 average for inside 50’s in all but two games.


Thinking my 6,666 post now ought to be my round 24 preview post. Although I may just use it to slag off at Warwick ...

 

[Nunez]

As past time poster, long-time lurker. Thought it best if started throwing some posts at the board to see if they stick before doing an end of home & away preview post. Hopefully the below numbers stack up better than my total post numbers. Had planned to make this my 6,666 post but what the hell.

Thought it might be worth sharing/discussing after looking a little deeper into inside 50 differentials after noticing that the last 4 AFL premiers were ranked either #1 or #2 in that year with the highest inside 50 differential.

And at the top of i50 differential per game we have
#1 – Pies have an 8.9
#2 – Lions 7.8
#3 - Cats 7.7

The differentials of course simply reflect inside 50 For and Against. After 10 rounds in 2023 the AFL team average for inside 50’s is 52.6 per team per game,
The top three are;
#1 – Dees with 57.9
#2 – Pies 56.7
#3 – Lions 56.4

The lowest average opponents inside 50’s
#1 – Pies - with opponents held to an average of 47.8
#2 – Cats - 47.9
#3 – Lions - 48.6

So for discussion purposes let’s remove our round 1 game, where we were ambushed by Port leaving our 9 game differential average at 11.4 per game. For comparison, last year, Geelong had a differential of 12.6. [we were ranked 6th with 3.9]

Returning to the inside 50 numbers – the below table shows all teams average inside 50’s by rank for the first 10 rounds, their i50 versus Lions, that games variance to their average to date, our i50 in that game and our game differential. As things stand, we have been able to hold all but two opponents to LESS than their 2023 average for inside 50’s in all but two games.
View attachment 1694194

Thinking my 6,666 post now ought to be my round 24 preview post. Although I may just use it to slag off at Warwick ...

Welcome back O' Blynd one.

 

[KissKiss]

I was talking to a friend of a friend on the weekend who has a bit more exposure to the internal workings of another club. He made the comment that Inside50s was not of interest to clubs but they tracked inside30s as that was where it really became a high % scoring opportunity (not available to general public of course). It did make me wonder about that. Where are our scoring zones? Where are we targeting more?

Historically we have kicked to the pockets a lot. But Im not sure that is as true this season

 

[Nathan Barley]

I was talking to a friend of a friend on the weekend who has a bit more exposure to the internal workings of another club. He made the comment that Inside50s was not of interest to clubs but they tracked inside30s as that was where it really became a high % scoring opportunity (not available to general public of course). It did make me wonder about that. Where are our scoring zones? Where are we targeting more?

Historically we have kicked to the pockets a lot. But Im not sure that is as true this season

I think it's partly because if you get it inside 30 it's harder for them to rebound when you press up

 

[Warwick]

As past time poster, long-time lurker. Thought it best if started throwing some posts at the board to see if they stick before doing an end of home & away preview post. Hopefully the below numbers stack up better than my total post numbers. Had planned to make this my 6,666 post but what the hell.

Thought it might be worth sharing/discussing after looking a little deeper into inside 50 differentials after noticing that the last 4 AFL premiers were ranked either #1 or #2 in that year with the highest inside 50 differential.

And at the top of i50 differential per game we have
#1 – Pies have an 8.9
#2 – Lions 7.8
#3 - Cats 7.7

The differentials of course simply reflect inside 50 For and Against. After 10 rounds in 2023 the AFL team average for inside 50’s is 52.6 per team per game,
The top three are;
#1 – Dees with 57.9
#2 – Pies 56.7
#3 – Lions 56.4

The lowest average opponents inside 50’s
#1 – Pies - with opponents held to an average of 47.8
#2 – Cats - 47.9
#3 – Lions - 48.6

So for discussion purposes let’s remove our round 1 game, where we were ambushed by Port leaving our 9 game differential average at 11.4 per game. For comparison, last year, Geelong had a differential of 12.6. [we were ranked 6th with 3.9]

Returning to the inside 50 numbers – the below table shows all teams average inside 50’s by rank for the first 10 rounds, their i50 versus Lions, that games variance to their average to date, our i50 in that game and our game differential. As things stand, we have been able to hold all but two opponents to LESS than their 2023 average for inside 50’s in all but two games.
View attachment 1694194

Thinking my 6,666 post now ought to be my round 24 preview post. Although I may just use it to slag off at Warwick ...

What annoys me is that you’ve demonstrated an ability to show that level of depth and insight, one questions what you’ve been doing for the other 6664.

 

[blynd_freddie]

What annoys me is that you’ve demonstrated an ability to show that level of depth and insight, one questions what you’ve been doing for the other 6664.

The ox is slow, but your mirth is slower.

 

[blynd_freddie]

I was talking to a friend of a friend on the weekend who has a bit more exposure to the internal workings of another club. He made the comment that Inside50s was not of interest to clubs but they tracked inside30s as that was where it really became a high % scoring opportunity (not available to general public of course). It did make me wonder about that. Where are our scoring zones? Where are we targeting more?

Historically we have kicked to the pockets a lot. But Im not sure that is as true this season

I think pockets have been a lock in option, if the ball cannot be marked then the boundary line and/or stoppage creates a secondary option. The inside 30 thinking is that a possession there greatly increases the scoring efficiency.
Individual player heat maps used to show possessions and I think scoring from positions but sadly quite a labour intensive task to create any decent game by game fact information document.

 

[Tom14]

I think pockets have been a lock in option, if the ball cannot be marked then the boundary line and/or stoppage creates a secondary option. The inside 30 thinking is that a possession there greatly increases the scoring efficiency.
Individual player heat maps used to show possessions and I think scoring from positions but sadly quite a labour intensive task to create any decent game by game fact information document.

Somewhat on the same lines as this - I put myself through listening to Sportsday while making dinner last night, and Daniel Hoyne from Champion Data made the point that in Melbourne's two games against Port and us, they'd conceded 5 goals from forward 50 stoppages, when their season average outside those games was about a goal conceded from that source.

The main reason? There were just far more F50 stoppages in their two games against us and Port. Managing to contest in the air and get the ball to ground and locked inside F50 is such a big thing these days against great intercept defenders. I almost think it is worth tracking our intercept marks against stat on its own to see how much it impacts our scoring and our defence on rebound.

 

[Grasshopper17]

I was talking to a friend of a friend on the weekend who has a bit more exposure to the internal workings of another club. He made the comment that Inside50s was not of interest to clubs but they tracked inside30s as that was where it really became a high % scoring opportunity (not available to general public of course). It did make me wonder about that. Where are our scoring zones? Where are we targeting more?

Historically we have kicked to the pockets a lot. But Im not sure that is as true this season

I think we are definitely entering 50 much deeper, which is the main reason our G-B scorelines look respectable this year. Daniher and Hipwood aside I certainly don't think our set shot goal kicking has improved an enormous extent.

 

[Elixuh]

Melbourne just got a David King special on 360.

King analysed different Melbourne stats but removed their games against Hawks, norf and eagles. He then compared it against other teams in the comp without removing those teams lol.

 

[Tom14]

Melbourne just got a David King special on 360.

King analysed different Melbourne stats but removed their games against Hawks, norf and eagles. He then compared it against other teams in the comp without removing those teams lol.

Incredible, can't wait to (hate) watch.

 

[buttermuffs]

Melbourne just got a David King special on 360.

King analysed different Melbourne stats but removed their games against Hawks, norf and eagles. He then compared it against other teams in the comp without removing those teams lol.

Did he just compare it against competition average or specifically compared it to team who have already played all three?

The thing to note is that I’m not sure if there’s been another team in the league who’s actually versed all three other than Freo. When there’s a pretty big gap between those three and the rest that actually presents an interesting effect on how the team is perceived. Collingwood hasn’t played any of those three (North on the weekend) while we have only played North.

Looking at Melbourne’s stats without those games paints a better picture of how they stand against the rest of the league who haven’t had that luxury. I think it’s similar to how we remove the Port game when looking at stats sometimes because it’s such an anomaly compared to how we’ve played the rest of the season.


Sent from my iPhone using BigFooty.com

 

[martinson]

Did he just compare it against competition average or specifically compared it to team who have already played all three?

The thing to note is that I’m not sure if there’s been another team in the league who’s actually versed all three other than Freo. When there’s a pretty big gap between those three and the rest that actually presents an interesting effect on how the team is perceived. Collingwood hasn’t played any of those three (North on the weekend) while we have only played North.

Looking at Melbourne’s stats without those games paints a better picture of how they stand against the rest of the league who haven’t had that luxury. I think it’s similar to how we remove the Port game when looking at stats sometimes because it’s such an anomaly compared to how we’ve played the rest of the season.


Sent from my iPhone using BigFooty.com

Sometimes the anomaly is the problem you have.

 

[Elixuh]

Did he just compare it against competition average or specifically compared it to team who have already played all three?

The thing to note is that I’m not sure if there’s been another team in the league who’s actually versed all three other than Freo. When there’s a pretty big gap between those three and the rest that actually presents an interesting effect on how the team is perceived. Collingwood hasn’t played any of those three (North on the weekend) while we have only played North.

Looking at Melbourne’s stats without those games paints a better picture of how they stand against the rest of the league who haven’t had that luxury. I think it’s similar to how we remove the Port game when looking at stats sometimes because it’s such an anomaly compared to how we’ve played the rest of the season.


Sent from my iPhone using BigFooty.com

This would still be flawed, but wouldn't it have been logical to remove WCE, Norf and Hawks from the dataset altogether. The competition wide data would be skewed by having those three in it and Melbourne's data would be skewed the other way by having it removed. It made no sense.

 

[martinson]

This would still be flawed, but wouldn't it have been logical to remove WCE, Norf and Hawks from the dataset altogether. The competition wide data would be skewed by having those three in it and Melbourne's data would be skewed the other way by having it removed. It made no sense.

But North were fairly good for a few games.

Hawthorn have had transient joys.

The drover's dog would've beaten Geelong at the start of the year.

As soon as you start tampering with this sort of data it has no logical validity . And we know its empirical validity is flawed by what we know and see.

In other words most of the starts put up by footy 'experts' are just a load of junk to justify whatever point they want to make.

It always amuses me to see coaches reaching for the stats sheet in their post match pressers to illustrate whatever point they're making when they don't need the stats sheet to bolster what their observation and experience has already told them.

 

[Elixuh]

But North were fairly good for a few games.

Hawthorn have had transient joys.

The drover's dog would've beaten Geelong at the start of the year.

As soon as you start tampering with this sort of data it has no logical validity . And we know its empirical validity is flawed by what we know and see.

In other words most of the starts put up by footy 'experts' are just a load of junk to justify whatever point they want to make.

It always amuses me to see coaches reaching for the stats sheet in their post match pressers to illustrate whatever point they're making when they don't need the stats sheet to bolster what their observation and experience has already told them.

Yeah the David King special comment was more speaking to him manufacturing stats to suit whatever narrative he wants to spew out.

 

[buttermuffs]

This would still be flawed, but wouldn't it have been logical to remove WCE, Norf and Hawks from the dataset altogether. The competition wide data would be skewed by having those three in it and Melbourne's data would be skewed the other way by having it removed. It made no sense.

I just watched it and it’s worse than I expected. He looks at total scores from stoppages and where they rank in the AFL from it and then took out the three games worth of scores. I gave him the benefit of the doubt that he used averages but he used totals.

I don’t think the competition wide data would be skewed from them because those three teams are also a part of the competition average. North getting outscored from stoppages counts just as much as the team that outscores them. Like how we destroyed Melbourne in clearances: the average clearances for the team would’ve been 46 with no indication as to who technically lost them.

Anyway, Kings follows up with an actual useful stat showing Melbourne are +90 from stoppage scoring against those three teams and -50 against the other seven teams they’ve versed this year. Pretty damning that the regular average would have them winning clearance scores 4 points but it swings to losing by a goal when you take out the easy beats.

It shows that what may as well been a perceived advantage come finals is actually is a shortfall for them.


Sent from my iPhone using BigFooty.com

 

[buttermuffs]

I just watched it and it’s worse than I expected. He looks at total scores from stoppages and where they rank in the AFL from it and then took out the three games worth of scores. I gave him the benefit of the doubt that he used averages but he used totals.

I don’t think the competition wide data would be skewed from them because those three teams are also a part of the competition average. North getting outscored from stoppages counts just as much as the team that outscores them. Like how we destroyed Melbourne in clearances: the average clearances for the team would’ve been 46 with no indication as to who technically lost them.

Anyway, Kings follows up with an actual useful stat showing Melbourne are +90 from stoppage scoring against those three teams and -50 against the other seven teams they’ve versed this year. Pretty damning that the regular average would have them winning clearance scores 4 points but it swings to losing by a goal when you take out the easy beats.

It shows that what may as well been a perceived advantage come finals is actually is a shortfall for them.


Sent from my iPhone using BigFooty.com

I might add that even when comparing averages without games against the bottom three that it would be bad to rank them on a hypothetical ladder against the other teams. However, there is value is seeing how some stats hold up without needing to compare to the rest of the comp.


Sent from my iPhone using BigFooty.com

 

[Tom14]

Yeah the David King special comment was more speaking to him manufacturing stats to suit whatever narrative he wants to spew out.

In this case though it suits my narrative that the Dees are overrated so I'm inclined to let him run with it.

 

[M Malice]

Brenton Sanderson

Brenton Sanderson

Age

52

Ht

185cm

Wt

90kg

Pos.

Def

Career

Season

Last 5

D

15.7

4star

K

11.0

4star

HB

4.6

3star

M

4.8

5star

T

1.4

3star

No current season stats available

D

22.4

5star

K

16.0

5star

HB

6.4

4star

M

6.4

5star

T

1.6

4star

put up some stats on SEN this morning >

In our 8 wins we average 104 marks.

In our 2 losses we average 73 marks.

We are 18th in the competition for tackles.
______________________________________

He said that being last in tackles could be a worry against the really good sides, obviously he mustn't rate the Magpies and Demons as "really good sides". He tipped the Crows to win on Sunday.

Being 18th for tackles is not a worry and a consequence of our game style IMO.

 

[Tom14]

Brenton Sanderson put up some stats on SEN this morning >

In our 8 wins we average 104 marks.

In our 2 losses we average 73 marks.

We are 18th in the competition for tackles.
______________________________________

He said that being last in tackles could be a worry against the really good sides, obviously he mustn't rate the Magpies and Demons as "really good sides". He tipped the Crows to win on Sunday.

Being 18th for tackles is not a worry and a consequence of our game style IMO.

Wow, in our losses we haven't had the footy as much. Shocker!

 

[Tom14]

Brenton Sanderson put up some stats on SEN this morning >

In our 8 wins we average 104 marks.

In our 2 losses we average 73 marks.

We are 18th in the competition for tackles.
______________________________________

He said that being last in tackles could be a worry against the really good sides, obviously he mustn't rate the Magpies and Demons as "really good sides". He tipped the Crows to win on Sunday.

Being 18th for tackles is not a worry and a consequence of our game style IMO.

I spent 5 minutes looking this up and the marks stat is even stupider than I thought. We took 97 marks in round 3 against the dogs - the reason we 'average' 73 marks is we only had 52 in round 1. A clear and massive outlier because we couldn't get our hands on the footy at all.

If you have lost 2 games out of 10, you can't make sweeping stat-based statements only looking at the losses.

As for the tackles stat - we average the 6th least tackles against us. Pretty clear that the games we play don't involve high numbers of tackles.

 

[Elixuh]

Brenton Sanderson put up some stats on SEN this morning >

In our 8 wins we average 104 marks.

In our 2 losses we average 73 marks.

We are 18th in the competition for tackles.
______________________________________

He said that being last in tackles could be a worry against the really good sides, obviously he mustn't rate the Magpies and Demons as "really good sides". He tipped the Crows to win on Sunday.

Being 18th for tackles is not a worry and a consequence of our game style IMO.

Top quality analysis from Mr Sanderson there. You’d think seeing 18th for tackles would prompt you look a little deeper into a team that is a contender.

 

[M Malice]

Top quality analysis from Mr Sanderson there. You’d think seeing 18th for tackles would prompt you look a little deeper into a team that is a contender.

Definitely, as Tom14 pointed out... I don't think Sanderson is an investigative journalist.