Analysis Essendon goalkicking analysis

[calyam]

I created a thread earlier today on the main board looking at the best and worst goalkickers in the league based on some really interesting analysis found here.

The article assesses goalkickers in three distinct ways:

1. Simple accuracy (goals / shot attempts).
2. Points scored compared with expected score.
3. Probability of scoring as many goals by luck.

Simple accuracy

This is the focus of most goalkicking debates on BigFooty but it is also the least useful measure since it doesn't account for the difficult of the shots taken. It's more difficult to kick a goal from the boundary line than it is from the corridor. It is more difficult again to bomb one from the centre square.

1. Hooker 27.15 64%
2. Fantasia 32.19 63%
3. Daniher 45.26 63%
4. McDonald-Tipungwuti 25.15 62%
5. Stewart 12.8 60%
6. Zaharakis 9.7 56%
7. Heppell 7.6 54%
8. Colyer 10.9 53%
9. Merrett 5.5 50%
10. Green 15.15 50%

We have really benefited from the accuracy of our five main forwards who all convert at 60% or above. Green is the outlier among our forwards but if 50% accuracy is a teams worst (among players with 10 or more scoring opportunities) then you are doing pretty well.

Points scored compared with expected score

The second measure compares the points scored by a player against their expected score based on the difficulty of their shots. So each shot is assigned a probability of success (based on the league average from that position) and a good goalkicker would be one who has a positive net difference (actual score > expected score). Fantasia leads the Bombers in this category with a net difference of +28 (Actual score 211 > expected score 183).

1. Fantasia 211 > 183; +28
2. Daniher 296 > 278; +18
3. McDonald-Tipungwuti 165 > 151; +14
4. Zaharakis 61 > 56; +5
5. Merrett 35 > 33; +2
6. Heppell 48 > 47; +1
7. Stewart 80 < 81; -1
8. Colyer 69 > 70; -1
9. Hooker 177 < 180; -3
10. Green 105 < 113; -8

This is really interesting. Hooker leads our team in simple accuracy but it is mainly because he takes relatively easy shots for goal. Adjusted for difficulty and his accuracy is actually below league average. It makes sense; how many of you would want Hooker kicking for goal on a 45 degree angle with the game on the line?

One of the things that jumps out to me is that our major goalkickers are accurate in front of goal because the team has done a great job at creating high quality opportunities. In other words we are scoring on the fast break and through the corridor rather than relying on low percentage opportunities from the boundary line.

Probability of scoring as many goals by luck

The third measure compares a players actual score against the probability of an average player doing just as well from the same opportunities. So if a player had a score of 20 per cent this would mean that there is a 20 per cent chance that this goalkicker is really just an average goalkicker who has become lucky. The better the goalkicker the lower than probability is. Fantasia again leads the Bombers, with a probability of 5.71 per cent.

1. Fantasia 5.71%
2. Daniher 22.30%
3. McDonald-Tipungwuti 24.71%
4. Zaharakis 32.31%
5. Heppell 35.98%
6. Merrett 55.65%
7. Stewart 58.62%
8. Colyer 62.92%
9. Hooker 68.01%
10. Green 79.44%

So in other words there is a 22.3% chance that Daniher isn't actually an accurate kick for goal but just experiencing a lucky phase! Fantasia establishes himself as easily our most reliable kick for goal.

The team will also improve a great deal if Hooker and Stewart can do a little more with their opportunities.

 

[Killer_Mike]

Good work Calyan.

Another thing that would interest me is whether they are taking set shots or shots on the run. I suspect Green suffers in that regard because he's taking a lot more shots on the run rather than a set shot from 30m out.

It highlights how deadly Fantasia is around goal though because he would probably take as many shots around the corner as he would set shots.

 

[lamaros]

Good work Calyan.

Another thing that would interest me is whether they are taking set shots or shots on the run. I suspect Green suffers in that regard because he's taking a lot more shots on the run rather than a set shot from 30m out.

It highlights how deadly Fantasia is around goal though because he would probably take as many shots around the corner as he would set shots.

I think the majority of Fanta's goals have been set shots actually.

 

[Killer_Mike]

I think the majority of Fanta's goals have been set shots actually.

Yeah, would be nice if we had access to that data.

Bloody AFL and protecting stats...

 

[Lore]

Yeah, would be nice if we had access to that data.

Bloody AFL and protecting stats...

It's interesting that GPS data had to be released by the players, either via the clubs (Bulldogs were holding out) or later through the CBA. I wonder how much of it is to do with that, because they don't monetise it (at least to the public). Champion data collects it and shares it with the clubs and some media outlets, presumably they make money that way. Surely they'd make more if it was some sort of subscription based access that anyone could have (even levels with higher prices to get to the more advanced stuff). Atm there's a bunch of really basic stats on the AFL website and with the live pass there's about five other different stats and now GPS data, but there's definitely a lot more that the public can't access at all. Prospectus is rubbish btw, if anyone has considered trying it.

 

[Killer_Mike]

It's interesting that GPS data had to be released by the players, either via the clubs (Bulldogs were holding out) or later through the CBA. I wonder how much of it is to do with that, because they don't monetise it (at least to the public). Champion data collects it and shares it with the clubs and some media outlets, presumably they make money that way. Surely they'd make more if it was some sort of subscription based access that anyone could have (even levels with higher prices to get to the more advanced stuff). Atm there's a bunch of really basic stats on the AFL website and with the live pass there's about five other different stats and now GPS data, but there's definitely a lot more that the public can't access at all. Prospectus is rubbish btw, if anyone has considered trying it.

Stats coverage on our game is the worst in the world I would imagine.

We are treated like second rate citizens and it's frustrating.

 

[eth-dog]

Interesting that Stewart is seen as lucky. Other than the Saints game he's been an excellent set shot IMO

 

[Liberator]

I created a thread earlier today on the main board looking at the best and worst goalkickers in the league based on some really interesting analysis found here.

The article assesses goalkickers in three distinct ways:

1. Simple accuracy (goals / shot attempts).
2. Points scored compared with expected score.
3. Probability of scoring as many goals by luck.

Simple accuracy

This is the focus of most goalkicking debates on BigFooty but it is also the least useful measure since it doesn't account for the difficult of the shots taken. It's more difficult to kick a goal from the boundary line than it is from the corridor. It is more difficult again to bomb one from the centre square.

1. Hooker 27.15 64%
2. Fantasia 32.19 63%
3. Daniher 45.26 63%
4. McDonald-Tipungwuti 25.15 62%
5. Stewart 12.8 60%
6. Zaharakis 9.7 56%
7. Heppell 7.6 54%
8. Colyer 10.9 53%
9. Merrett 5.5 50%
10. Green 15.15 50%

We have really benefited from the accuracy of our five main forwards who all convert at 60% or above. Green is the outlier among our forwards but if 50% accuracy is a teams worst (among players with 10 or more scoring opportunities) then you are doing pretty well.

Points scored compared with expected score

The second measure compares the points scored by a player against their expected score based on the difficulty of their shots. So each shot is assigned a probability of success (based on the league average from that position) and a good goalkicker would be one who has a positive net difference (actual score > expected score). Fantasia leads the Bombers in this category with a net difference of +28 (Actual score 211 > expected score 183).

1. Fantasia 211 > 183; +28
2. Daniher 296 > 278; +18
3. McDonald-Tipungwuti 165 > 151; +14
4. Zaharakis 61 > 56; +5
5. Merrett 35 > 33; +2
6. Heppell 48 > 47; +1
7. Stewart 80 < 81; -1
8. Colyer 69 > 70; -1
9. Hooker 177 < 180; -3
10. Green 105 < 113; -8

This is really interesting. Hooker leads our team in simple accuracy but it is mainly because he takes relatively easy shots for goal. Adjusted for difficulty and his accuracy is actually below league average. It makes sense; how many of you would want Hooker kicking for goal on a 45 degree angle with the game on the line?

One of the things that jumps out to me is that our major goalkickers are accurate in front of goal because the team has done a great job at creating high quality opportunities. In other words we are scoring on the fast break and through the corridor rather than relying on low percentage opportunities from the boundary line.

Probability of scoring as many goals by luck

The third measure compares a players actual score against the probability of an average player doing just as well from the same opportunities. So if a player had a score of 20 per cent this would mean that there is a 20 per cent chance that this goalkicker is really just an average goalkicker who has become lucky. The better the goalkicker the lower than probability is. Fantasia again leads the Bombers, with a probability of 5.71 per cent.

1. Fantasia 5.71%
2. Daniher 22.30%
3. McDonald-Tipungwuti 24.71%
4. Zaharakis 32.31%
5. Heppell 35.98%
6. Merrett 55.65%
7. Stewart 58.62%
8. Colyer 62.92%
9. Hooker 68.01%
10. Green 79.44%

So in other words there is a 22.3% chance that Daniher isn't actually an accurate kick for goal but just experiencing a lucky phase! Fantasia establishes himself as easily our most reliable kick for goal.

The team will also improve a great deal if Hooker and Stewart can do a little more with their opportunities.

Great post mate!

 

[Doss]

I created a thread earlier today on the main board looking at the best and worst goalkickers in the league based on some really interesting analysis found here.

The article assesses goalkickers in three distinct ways:

1. Simple accuracy (goals / shot attempts).
2. Points scored compared with expected score.
3. Probability of scoring as many goals by luck.

Simple accuracy

This is the focus of most goalkicking debates on BigFooty but it is also the least useful measure since it doesn't account for the difficult of the shots taken. It's more difficult to kick a goal from the boundary line than it is from the corridor. It is more difficult again to bomb one from the centre square.

1. Hooker 27.15 64%
2. Fantasia 32.19 63%
3. Daniher 45.26 63%
4. McDonald-Tipungwuti 25.15 62%
5. Stewart 12.8 60%
6. Zaharakis 9.7 56%
7. Heppell 7.6 54%
8. Colyer 10.9 53%
9. Merrett 5.5 50%
10. Green 15.15 50%

We have really benefited from the accuracy of our five main forwards who all convert at 60% or above. Green is the outlier among our forwards but if 50% accuracy is a teams worst (among players with 10 or more scoring opportunities) then you are doing pretty well.

Points scored compared with expected score

The second measure compares the points scored by a player against their expected score based on the difficulty of their shots. So each shot is assigned a probability of success (based on the league average from that position) and a good goalkicker would be one who has a positive net difference (actual score > expected score). Fantasia leads the Bombers in this category with a net difference of +28 (Actual score 211 > expected score 183).

1. Fantasia 211 > 183; +28
2. Daniher 296 > 278; +18
3. McDonald-Tipungwuti 165 > 151; +14
4. Zaharakis 61 > 56; +5
5. Merrett 35 > 33; +2
6. Heppell 48 > 47; +1
7. Stewart 80 < 81; -1
8. Colyer 69 > 70; -1
9. Hooker 177 < 180; -3
10. Green 105 < 113; -8

This is really interesting. Hooker leads our team in simple accuracy but it is mainly because he takes relatively easy shots for goal. Adjusted for difficulty and his accuracy is actually below league average. It makes sense; how many of you would want Hooker kicking for goal on a 45 degree angle with the game on the line?

One of the things that jumps out to me is that our major goalkickers are accurate in front of goal because the team has done a great job at creating high quality opportunities. In other words we are scoring on the fast break and through the corridor rather than relying on low percentage opportunities from the boundary line.

Probability of scoring as many goals by luck

The third measure compares a players actual score against the probability of an average player doing just as well from the same opportunities. So if a player had a score of 20 per cent this would mean that there is a 20 per cent chance that this goalkicker is really just an average goalkicker who has become lucky. The better the goalkicker the lower than probability is. Fantasia again leads the Bombers, with a probability of 5.71 per cent.

1. Fantasia 5.71%
2. Daniher 22.30%
3. McDonald-Tipungwuti 24.71%
4. Zaharakis 32.31%
5. Heppell 35.98%
6. Merrett 55.65%
7. Stewart 58.62%
8. Colyer 62.92%
9. Hooker 68.01%
10. Green 79.44%

So in other words there is a 22.3% chance that Daniher isn't actually an accurate kick for goal but just experiencing a lucky phase! Fantasia establishes himself as easily our most reliable kick for goal.

The team will also improve a great deal if Hooker and Stewart can do a little more with their opportunities.

That second measure pretty much passes the eye test impression I've formed - that Fantasia is the one you want with the ball if someone is kicking for the game.

 

[Runk]

I created a thread earlier today on the main board looking at the best and worst goalkickers in the league based on some really interesting analysis found here.

The article assesses goalkickers in three distinct ways:

1. Simple accuracy (goals / shot attempts).
2. Points scored compared with expected score.
3. Probability of scoring as many goals by luck.

Simple accuracy

This is the focus of most goalkicking debates on BigFooty but it is also the least useful measure since it doesn't account for the difficult of the shots taken. It's more difficult to kick a goal from the boundary line than it is from the corridor. It is more difficult again to bomb one from the centre square.

1. Hooker 27.15 64%
2. Fantasia 32.19 63%
3. Daniher 45.26 63%
4. McDonald-Tipungwuti 25.15 62%
5. Stewart 12.8 60%
6. Zaharakis 9.7 56%
7. Heppell 7.6 54%
8. Colyer 10.9 53%
9. Merrett 5.5 50%
10. Green 15.15 50%

We have really benefited from the accuracy of our five main forwards who all convert at 60% or above. Green is the outlier among our forwards but if 50% accuracy is a teams worst (among players with 10 or more scoring opportunities) then you are doing pretty well.

Points scored compared with expected score

The second measure compares the points scored by a player against their expected score based on the difficulty of their shots. So each shot is assigned a probability of success (based on the league average from that position) and a good goalkicker would be one who has a positive net difference (actual score > expected score). Fantasia leads the Bombers in this category with a net difference of +28 (Actual score 211 > expected score 183).

1. Fantasia 211 > 183; +28
2. Daniher 296 > 278; +18
3. McDonald-Tipungwuti 165 > 151; +14
4. Zaharakis 61 > 56; +5
5. Merrett 35 > 33; +2
6. Heppell 48 > 47; +1
7. Stewart 80 < 81; -1
8. Colyer 69 > 70; -1
9. Hooker 177 < 180; -3
10. Green 105 < 113; -8

This is really interesting. Hooker leads our team in simple accuracy but it is mainly because he takes relatively easy shots for goal. Adjusted for difficulty and his accuracy is actually below league average. It makes sense; how many of you would want Hooker kicking for goal on a 45 degree angle with the game on the line?

One of the things that jumps out to me is that our major goalkickers are accurate in front of goal because the team has done a great job at creating high quality opportunities. In other words we are scoring on the fast break and through the corridor rather than relying on low percentage opportunities from the boundary line.

Probability of scoring as many goals by luck

The third measure compares a players actual score against the probability of an average player doing just as well from the same opportunities. So if a player had a score of 20 per cent this would mean that there is a 20 per cent chance that this goalkicker is really just an average goalkicker who has become lucky. The better the goalkicker the lower than probability is. Fantasia again leads the Bombers, with a probability of 5.71 per cent.

1. Fantasia 5.71%
2. Daniher 22.30%
3. McDonald-Tipungwuti 24.71%
4. Zaharakis 32.31%
5. Heppell 35.98%
6. Merrett 55.65%
7. Stewart 58.62%
8. Colyer 62.92%
9. Hooker 68.01%
10. Green 79.44%

So in other words there is a 22.3% chance that Daniher isn't actually an accurate kick for goal but just experiencing a lucky phase! Fantasia establishes himself as easily our most reliable kick for goal.

The team will also improve a great deal if Hooker and Stewart can do a little more with their opportunities.

Killer post Cal! Well done

 

[Mercurial89]

Amazing post calyam

Question, is the luck component of Danihers shots because he takes so many from long range? There would probably only be a few players that do it reliably, Buddy, Daniher, Walker and now Hipwood are developing reputations for it.

So they are not the average but for Joe, his long range bombs are his most accurate. Whilst Raz is getting relatively simple shots and converting them?

 

[Lance Uppercut]

until about a month ago I didn't think Fantasia could actually miss a set shot. From the moment I started mouthing off to my wife about how this guy never misses he started ****en missing easy shots. Ugh. But by far still one of the absolute best shots in the comp

 

[Mercurial89]

until about a month ago I didn't think Fantasia could actually miss a set shot. From the moment I started mouthing off to my wife about how this guy never misses he started ****en missing easy shots. Ugh. But by far still one of the absolute best shots in the comp

My reasoning for some of the set shot misses is that he has started providing a fair amount of run out of the middle and back line. take a look when he has been lining up the last few weeks, the guy looks absolutely spent, and they aren't just slight misses, he is spraying them horribly (for him).

 

[calyam]

Amazing post calyam

Question, is the luck component of Danihers shots because he takes so many from long range? There would probably only be a few players that do it reliably, Buddy, Daniher, Walker and now Hipwood are developing reputations for it.

So they are not the average but for Joe, his long range bombs are his most accurate. Whilst Raz is getting relatively simple shots and converting them?

It's probably more a product of the fact that his accuracy isn't exceptional for the opportunities he creates. The same guy that did the analysis contained in this thread did another blog post focusing on the Coleman Medal contenders (Every shot in the 2017 Coleman Medal Race).

It's a few rounds old now but it shows that the expected accuracy of someone taking Daniher's shots is around 57 per cent. Daniher's accuracy is 63 per cent so he is an above average goalkicker and that shows up in the data in the OP.

It actually appears as though Fantasia takes more difficult shots (probably wider angles; fewer set shots) but converts at the same rate as Daniher. The data is basically saying that if Fantasia and Daniher had the same opportunities than Fantasia would kick more goals.

 

[blitzer]

It's probably more a product of the fact that his accuracy isn't exceptional for the opportunities he creates. The same guy that did the analysis contained in this thread did another blog post focusing on the Coleman Medal contenders (Every shot in the 2017 Coleman Medal Race).

It's a few rounds old now but it shows that the expected accuracy of someone taking Daniher's shots is around 57 per cent. Daniher's accuracy is 63 per cent so he is an above average goalkicker and that shows up in the data in the OP.

It actually appears as though Fantasia takes more difficult shots (probably wider angles; fewer set shots) but converts at the same rate as Daniher. The data is basically saying that if Fantasia and Daniher had the same opportunities than Fantasia would kick more goals.

In fairness there is a reason why Daniher gets the benefit of easier shots at goal. Its called being 2m tall and being able to leap about another 2 metres on top of that. Fantasia whilst being pretty good in his own right in the air for a smaller player will never get the advantage of getting lots of shots close to goal. Plus no matter how good you are if you are crumbing goals at full tilt your accuracy will suffer unless you can get sneak in behind the opposition defense.