Analysis Goalkicking

[Interloperer]

I manually collected all the crows games from the AFLxScore twitter account:

(This excludes rushed behinds)

Round​	Opponent​	Home​	WIN​	Actual Score*​	Expected Score​	Efficiency​	Opponent Score*​	Opponent Expected​	Opponent Efficiency​	Ratio​
1​	GWS​	FALSE​	LOSS​	90​	95.1​	94.64%​	104​	105.5​	98.58%​	0.96​
2​	RICH​	TRUE​	LOSS​	74​	85.9​	86.15%​	107​	89.5​	119.55%​	0.72​
3​	PORT​	FALSE​	WIN​	117​	96.6​	121.12%​	86​	79.8​	107.77%​	1.12​
4​	FRE​	TRUE​	WIN​	110​	90​	122.22%​	69​	69.1​	99.86%​	1.22​
5​	CARL​	TRUE​	WIN​	118​	108.5​	108.76%​	59​	58.5​	100.85%​	1.08​
6​	HAW​	FALSE​	WIN​	76​	82.3​	92.35%​	75​	83.9​	89.39%​	1.03​
7​	COLL​	TRUE​	LOSS​	55​	75.8​	72.56%​	55​	59.7​	92.13%​	0.79​
8​	GEEL​	FALSE​	LOSS​	71​	65.5​	108.40%​	96​	86.3​	111.24%​	0.97​
9​	STK​	TRUE​	WIN​	121​	101.1​	119.68%​	69​	75​	92.00%​	1.30​
10​	WB​	FALSE​	LOSS​	40​	52​	76.92%​	85​	102​	83.33%​	0.92​
11​	BL​	TRUE​	WIN​	94​	84.6​	111.11%​	74​	84.9​	87.16%​	1.27​
12​	GCFC​	FALSE​	LOSS​	87​	79.9​	108.89%​	107​	97.2​	110.08%​	0.99​
13​	WCE​	TRUE​	WIN​	174​	138.2​	125.90%​	52​	51.5​	100.97%​	1.25​
14​	BYE​	​	​	​	​	​	​	​	​	​
15​	COLL​	FALSE​	LOSS​	77​	86.3​	89.22%​	80​	87.5​	91.43%​	0.98​
16​	NMFC​	TRUE​	WIN​	138​	122.7​	112.47%​	72​	65​	110.77%​	1.02​
17​	ESS​	FALSE​	LOSS​	95​	71​	133.80%​	110​	102.7​	107.11%​	1.25​
18​	GWS​	TRUE​	LOSS​	57​	61.5​	92.68%​	68​	69​	98.55%​	0.94​
19​	MELB​	FALSE​	LOSS​	92​	85.5​	107.60%​	97​	95.9​	101.15%​	1.06​
20​	PORT​	TRUE​	WIN​	112​	109​	102.75%​	64​	69​	92.75%​	1.11​
21​	GCFC​	TRUE​	WIN​	89​	103.7​	85.82%​	60​	59​	101.69%​	0.84​
22​	BL​	FALSE​	LOSS​	91​	101​	90.10%​	99​	91.6​	108.08%​	0.83​
23​	​	​	​	​	​	​	​	​	​	​
24​	​	​	​	​	​	​	​	​	​	​
TOTAL​	---​	---​	​	1978​	1896.2​	104.31%​	1688​	1682.6​	100.32%​	1.04​

Basically, Adelaide kick very slightly above expectations. While their opponents average almost dead-on expectations, which is to be expected and can't really be changed by anything Adelaide do.

If you can see any pattern beyond this, then you're better than me.

edit: fixed backwards Bulldogs game

 

[Scorpus]

I wonder what the split would be between set shots and snaps.

Anecdotally we look better at difficult snaps thanks to players like Walker and Rankine but we choke some really simple set shots

 

[Golumless]

I wonder what the split would be between set shots and snaps.

Anecdotally we look better at difficult snaps thanks to players like Walker and Rankine but we choke some really simple set shots

My guess would be very similar, seeing we've had enough snaps and set shots to really get rid of anything weird.

Though I'd imagine that the expected scores from set shots is lower than snaps.

 

[xjamescleanx]

Love it, mate. I used to work in a role where I was fed a ridiculous amount of data/KPI's... kind of miss it now I just live in the mountains like a strange, part-gay hobbit. Your posts always scratch an itch!!!

And yes... the important thing is the difference between "goal kicking cost us today" vs. "shot location" and "shot quality" etc. etc.

 

[NSolis]

Yeah, Adelaide's goalkicking is fine, the whole 'if only we could learn to kick we'd win X more games' thing is just a meme on here. A large portion of our set shots are taken by excellent kicks, which is a big plus. We do obviously miss shots and have had some poor goalkicking games this year, but this is a situation where people remember the bad stuff and forget the good, like the games we've had this year where we've kicked extremely well. The first Showdown in particular was a good game for that.

 

[Interloperer]

I wonder what the split would be between set shots and snaps.

Anecdotally we look better at difficult snaps thanks to players like Walker and Rankine but we choke some really simple set shots

There are maps of all the shots at the twitter account but it would take way more effort than its worth to analyse them all. But if you want to check individual games they are there to check

The current charts split into five types (set shot, ground kick, general play, loose ball get, hard ball get). But early season ones are less specific.

This is last round's game. Loose ball and hard ball get would likely be snaps (or weird ad hoc checksides).

The 5.28 Set Shot from 25m seems to be the most egregious miss (one of Laird's I think). 4.28 points below expectation. Or if you accept Champion Data's claim that the team kicking in has 0.5 points disadvantage, 3.78 points worse off, which makes the nearby 3.85 out on the full the worst miss.

 

[FR0GGY]

Yeah, Adelaide's goalkicking is fine, the whole 'if only we could learn to kick we'd win X more games' thing is just a meme on here. A large portion of our set shots are taken by excellent kicks, which is a big plus. We do obviously miss shots and have had some poor goalkicking games this year, but this is a situation where people remember the bad stuff and forget the good, like the games we've had this year where we've kicked extremely well. The first Showdown in particular was a good game for that.

Not sure I put much stock in expected shots, it we are missing either we aren’t kicking from the right spots or we aren’t good kicks.
We have kicked more points in 9.5 games (we were level in the second showdown) out of 21 games.
But what is telling is that 8 of those came in our 10 losses.
So basically kick straight and we win.

 

[NSolis]

Not sure I put much stock in expected shots, it we are missing either we aren’t kicking from the right spots or we aren’t good kicks.
We have kicked more points in 9.5 games (we were level in the second showdown) out of 21 games.
But what is telling is that 8 of those came in our 10 losses.
So basically kick straight and we win.

Yeah but at that point you may as well just say 'score more goals and you win'. The real question is whether it's executing in front of goal that is holding us back or other things. I'd have thought there were two losses this year where goalkicking held us back: the first Collingwood loss and this week against Brisbane. And that's supported by the stats above, we missed some very gettable shots those games and lost narrowly. As you say, kick straight and we win. On the other hand if you kick 11.12 and a bunch of the misses are rushed behinds or snaps from the boundary or whatever blaming goalkicking is just confusing the issue.

Comparison to other clubs is illustrative here.

 

[Scorpus]

Nah bad accuracy leading to losses is definitely costing us more than other teams.

We've let teams beat us 5 times from a position of fewer scoring shots.

I'm sure to get an overall average result we are compensating for this by being more accurate in games we are overall dominant in (eg. 21.12 and 27.12 against North and West Coast)

 

[Interloperer]

Not sure I put much stock in expected shots, it we are missing either we aren’t kicking from the right spots or we aren’t good kicks.

My main point in posting this was that was no discernable pattern, good or bad, to Adelaide's goal kicking.

But what little pattern there is tends towards the 'we aren’t kicking from the right spots' claim. Either through bad entries, bad choices to shoot instead of pass, or the third option; having the wrong player take the shot.

One other thing that does stand out is the two matches that were in anomalous conditions were both losses. Adelaide kicked obscenely well in the perfect conditions of the Docklands vs Essendon, but lost due to having so few chances.

And, even though Adelaide ended up kicking worse in the end, the Bulldogs were able to build a big lead in Ballarat when it counted by 'spraying and praying' with the wind and hoping a few of them actually went through the big sticks.

 

[robboaustralia]

Bit more of this maybe?

 

[Coopers]

There are maps of all the shots at the twitter account but it would take way more effort than its worth to analyse them all. But if you want to check individual games they are there to check

The current charts split into five types (set shot, ground kick, general play, loose ball get, hard ball get). But early season ones are less specific.

This is last round's game. Loose ball and hard ball get would likely be snaps (or weird ad hoc checksides).

The 5.28 Set Shot from 25m seems to be the most egregious miss (one of Laird's I think). 4.28 points below expectation. Or if you accept Champion Data's claim that the team kicking in has 0.5 points disadvantage, 3.78 points worse off, which makes the nearby 3.85 out on the full the worst miss.
View attachment 1773621

Those are a really good way of looking at it.

Brisbane game stands out because we missed some howlers that could have won it for us. I think the Pies games maybe as well, same thing.