Also the fact that the worst team in the league has <5% differential in clearances and half the top 8 is on the negative side of the ledger suggests to me that clearance numbers don’t mean a whole lot.
4 Likes
But what if the Guelf loves a stat on a Saturday night?
I think that differential thing is the point for Worsfold. Winning clearances is outweighed by the quality once you get the ball.
(btw, what’s the ref for these numbers?)
Well I definitely didn’t write a python script to download and scrape every 2017 match page for clearance numbers from footywire. Only a crazy person would do that.
Edit: I might shove into a csv or something tomorrow if I can be bothered.
3 Likes
Tee hee
If you look at footywire, rank teams by average and opposition average by various stat columns, it’s quite stark how much we avoided stoppage clearances this season.
We were actually 5th in the league for centre clearances & -0.6 per game down on our opponents - respectable.
Dead last for stoppage clearances & -1.6 down on our opponents.
4th bottom for hitouts & -1.3 a game down.
I believe a fair bit of this is due to attacking through the middle at all times - very few boundary stoppages. And, dare i say, eschewing grunt midfielders whose sole aim is to dive on the ball and force a ballup. Worsfold wants the side to keep play flowing and win turnovers, not grind out through set pieces and stoppages.
Tigers stoppage numbers were bollocks all year. Maybe not that important.
The key stat is scores from clearances - Nothing else compares.
@yaco55, you have an amazing ability to respond to a recent post in a thread, yet quote the opening post from the thread. I’ve seen it 3 or 4 times now.
Actually have no idea how the preamble appeared in my previous post. I never clicked on the preamble, even though its a well written piece.
If you click the reply at the bottom (not to a post), and then the speech bubble, you quote the first post.
So I reckon if you accidentally hit the big red reply, rather than the reply on the post, then click the speech bubble thinking you’re replying to the post, rather than the topic, and then don’t read the quote, this is what you end up with…
1 Like
Would love to see the script. I have a blog post on “how to scrape AFL stats data” bookmarked somewhere, but have never got around to figuring it out…
Clearances are a funny one, and a stat where I feel you really need the magic champion data stats and not just the box stats. A Dangerfield clearance is usually a different thing to a Tom Mitchell clearance.
This one is a year old (data from 2016 season) but a goody from Ryan Buckland
I wonder what something similar would yield for teams
2 Likes
But i’m not sure you can use the centre clearance totals.
Since we were the 3rd highest goal scoring team and 7th highest ‘points against’ team, we would be attending more centre bounces because our games have more goals.
Like what we saw vs the Swans, we need some clearance bulls so we can help stop momentum with constant i50 entries.
Around the ground it’s okay because we can put more numbers around the ball.
You can use it -in context.
4th highest team for giving up centre clearances, so we weren’t great in there. Just not as terrible as some are painting it.
/fxd4U
3 Likes
Interesting.
I don’t like the arbitrary positioning of the horizontal axis which seems to be a thing, and obviously the chosen data set creates the appearance of a linear trend, with Dangerfield bring this massive other, whereas there’s be so many more players in the bottom left corner if you included everyone.
But it’d be interesting to get the CD dataset and play around with it!
2 Likes
From memory he graphed top 30 clearance winners and top 30 metres gained, the right quadrant are top 30 in both.
Zerrett would be in the top left if graphed for 2017 I’d imagine, which I think is a misuse of an incredible resource. People tend to forget Zerrett when discussing our clearance issues.
1 Like
Just having a quick look at 2017 numbers, Merrett pretty similar, Dangerfield reverted to the mean in metres gained at 441 (Ablett higher than him on both stats)
1 Like
So many of these things could be analysed from a team perspective and game plans built around them.
For instance with Cats now having the two best metres gained from stoppages in the league (danger and Ablett) then as long as you win the stoppages you are in a good scoring position. As such do they then add another defender at stoppages to ensure if they loose it creates another and play a high stoppage game to take advantage of the two?
I wonder how deep they go?
1 Like
Someone on reddit made this site to compare player from stats or something you nerds will figure it out.
Very interesting stats there, particularly the 1% ers.
I still want to know where you can see stats on pressure acts