A 2023 All Australian Team Picked By Stats
Вставка
- Опубліковано 8 вер 2024
- After a year off in 2022, we’re back with our third rendition of the All-Australian by the stats team.
In 2020, inspired by a lack of transparency in all-australian selection, that left many fans frustrated, we created a framework for selecting the team based entirely off stats and ran an experiment to see what kind of results that generated.
Created by Andreas and Zachary Baruhas
===== Support us on Patreon: / footya2z =====
Facebook:
/ footya2z
Instagram:
/ footy_a2z
Kyle Langford played the first 3rd of the year in defence.. He truly had a good year..
Too bad his team is the joke of the league
third? he was playing forward against collingwood anzac day
@@desiredrapfilms
16 flags ain't no joke..
@@PluckYeah yeah back before Jesus was born 😂
@@MagpieR6 Semantics..
Glad that this shows that in a way Neale and Andrews were very stiff not to make it. Especially Andrews. Pretty much the best key defender and all round defender for the season. I couldn't believe 2 key defenders got picked ahead of him, he was Brisbanes best and fairest in a team that had Neale, Dunkley, Clug, Daniher (great season), and Cameron.
Tf is Clug, McLuggage?
Who else would it be?
@@VCS7083pretty common nickname for him among lions fans
Clug the Suitcase
Why would small forwards not have goal assists as one of their stats, they are there to set up forwards or kick their own goals, it is more important for them to have goal assists than a key forward.
that would get grian miers a guernsey, which he deserved...
Glad I'm not the only one who noticed this, why would key forwards be the only one judged on goal assists? It's the second most important stat after goals for ALL forwards.
As a bit of a math nerd, and one who cares little for teams that aren't my own, I was more impressed by your thorough methodology and rigour. I don't typically care for accolades that go to players from other teams.
Fantastic video. Well done mate.
Amazing been waiting for this!
As per usual, great video and great visuals. However, just a note on the methodology:
Using totals instead of averages for the stats makes this AA selection unfair imo - it means a lot of players will get a look in to this AA team simply because they played more games and could rack up higher totals. Should be setting a minimum games played and using averages instead.
E.g. Libba was #1 for average clearances and #3 for average contested possessions but falls down a bit when using totals, ranking goals by average goals per game puts tex at #1 with 3.5 per game while curnow and larkey are 2/3 at 3.1 per game each, which shows that on an individual game basis Tex is significantly more damaging in his ability to score goals - although Curnow would still make this AA team easily even with averages I’d imagine.
Not sure how much it would change the team as I don’t have the pro version of stats, but just looking at the normal version and comparing average vs. totals the ranks can change a fair bit. Was there any rationale behind using totals as opposed to averages?
Yeah to be honest in an ideal world we'd like to experiment with doing it that way, but believe it or not you actually can't filter by games played on afl.com.au (it's not even there as a stat) and it's just a lot of work to filter manually, so we'd need to look at using a different database. They exist, we just haven't really had the time to try it and this video is pretty delayed coming out as it is.
We're ok with using the totals because biassing towards players who had more of an impact across the entire season doesn't necessarily feel like the worst thing, but then it kind of comes back to what you want to AA team to be. Kinda similar to the Brownlow, Daicos has more votes per game but Neale played more games so he gets the medal as the best player across the whole season.
Definitely interested to hear what people with a good minimum games would be though, somewhere around 17 feels about right.
Great comment @mittenss6986
I think averages with a minimum games played is the way to go. I also think that weighted rankings, whilst adding some more complexity and bias to certain categories would be worthwhile. There are two aspects to weightings, for example:
(1) Instead of having an even weighting across the five categories (20% each) you could tweak so that certain categories have more impact on the overall score. Contested marks could have a 25% importance weighting and 1%ers could have a 10% importance weighting for defenders, for example.
(2) If a player were to have much higher numbers in a stat category than the next-ranked player, or conversely pip the next-ranked player by a very small margin, this is not reflected at all. One way to address this would be to have all stats weighted as a percentage of the highest total scored in the stat category. For instance, if Curnow kicked 90 goals and this was the most for the season, Curnow would score 100% and Player B kicking 45 goals would score 50%.
If goals kicked had an importance weighting of 25% for the forwards category, Curnow would have 25% towards goals and Player B would have 12.5%.
Hope this makes sense and could be explored further in future years editions!
There should be a mid-defender across the half-defender flank like a Scott Pendlebury, Nick Daicos or Elliott Yeo
That was the back pocket, Dan Houston dominated that position this year
Really interesting! Love seeing all of the reasons behind each stat being included and how a system like this compares to the official AFL selection.
Shows how good Rowell actually is. He’s not a flashy player or get as much recognition as other young players such as gulden, Noah anderson, nick daicos or butters, however Rowells stats show how valuable he actually is
Sure, but he can’t kick, especially compared to Gulden or Daicos, who are brilliant by foot …
Amazing channel!! Thanks for the informative content,
Cheers - a bombers fan from New Zealand 🇳🇿
Going off total rankings instead of averages (With a minimum game requirement obviously) means it's impossible for someone like Sicily to make this squad because of his ridiculous suspension.
If averages, what would the minimum number of games be? 20, 15, 10? Maybe a minimum of 15 would make sense.
I legit wait for this vid every year, love it
Such a good analysis and excellent script writing thanks
Excellent stuff. The 2020 video was a revolution
would be interesting to see if this could be done for the brownlow.
also, would love a video on how the bulldogs keep giving up massive scoring chains in the past few seasons! feels like its a problem that bevo just can't seem to solve...
I was gonna try that for sure. But I only did one game
stats don't cover fairness, the first criteria for a charlie...
@@hedonismunderstands2469 isn't that covered by getting banned
bevo needs the sack
This shows how much work Collins and Ballard had to do as the last line of defence. They had fairly mixed seasons for me but the stats credit the fact that they were put under lots of pressure and held up ok.
Criminally underrated channel this! Hoping you can get that 20k soon ❤
I've just started watching the AFL for the first time and I'm hooked. I'm so glad I found your channel to help me better understand the game. more power and respect to you from the Philippines!!! do you have like a document of PDF that will help explain what those stats mean. Thanks
Welcome to the community! Champion Data, the company that records stats for the AFL have a list here thats quite helpful: www.championdata.com/glossary/afl/
nice to see the video topic back, a different and refreshing point of view
Pumped to see this series back!!
Great to see Jordan Dawson on the feild
With the wingers, I wonder what it would have spat out if you included a stat like "total metres covered" to highlight two way running.
I'm still confused on the Net 1 on 1 wins. In the key defender stats Darcy Moore is rated 2nd and Jacob Weitering is rated 13th, but in the All-Round Defender category Darcy Moore falls to 10th and Jacob Weitering rises to 2nd? I understand that someone could fall with more players being included but how did Jacob Weitering go up?
Love this video, great insights and can't argue with the results
Good job, much better than the offical team. If I was to make a suggestion it would be to maybe weight the stats in games vs top 8 sides and discount vs bottom 4.
This is my favourite video to watch every year
love these videos
Would be interesting to see what the AA team would be if you only took the first half of the season or the second half of the season into account, just to see how things change throughout the season.
Sam collins is soooo underrated so
I’m glad you gave him the spot with your analysis.
Awesome video definitely do this and more like it next season
this is great content
Jack Buckley. BackChat Backman of the year, voting based on one-on-ones won. Underrated year, nice too see him here with a good all-round key defensive game.
Love the your AA team keep up the good work
sam Taylor and Harris Andrews put up all the one percent stats in the backline. sure they're advanced stats but really they play so much of a bigger role than the basic box score says. glad you included it.
edit: daicos not being include is a crime. idec about stats and im a blues fan. that's just wrong. obviously stats don't lie but the stats are lying
Great video 👌
Very interesting exercise. I wonder how this AA stats team would have gone in fantasy and how far over budget you'd have to go to make this team lol
Grian got robbed!
Always an interesting video, though -- keep up the great effort!
(If you ever want to go full stats-nerd though, I've got one suggestion: calculating then adding z-scores would work much better than adding placements)
I loved this, thank you so much for putting it together.
What data source do you use for this? Is it paid access?
I’d love to put together the same outcome for a specific team and note the ranking for each player comparatively to the rest of the competition, using that to indicate which positions/roles teams need the most improvement in.
Thanks mate, we find them all here, free to use: www.afl.com.au/stats/leaders
No players made it from West Coast, North Melbourne, Hawthorn. Shows why they got the 2023 top draft picks.
The only position in the forward line that took account of goal assists was key forward? I don't understand that at all.
Dropping the worst category distorts selections significantly. Effective disposals for any mid is essential, but was dropped for first ranked Matt Rowell who was only ranked 53rd in that key stat.
Bounces for wingers is crazy
Yeah it's rogue, but it helps filter out most the inside mids so helps favour actual wingers funnily enough.
Defenders probably needs to be tweaked to allow for, specifically, James Sicily. Didn't he lead the league in marks taken? I'd look at his stats to determine what's missing, because I assume it's obvious to most that he definitely deserved AA honours this year and was probably an oversight last year. I know this is manipulating the stats to a degree, but it also probably highlights what's not being considered. Or it could just be that Sicily was fighting against the tide of being in a weaker team, so his stats took a hit.
Great work! I think the captaincy selection needs a second look. Leadership qualities are tough to quantify, but I don't think Curnow and Rowel are the obvious choices when you've got guys like Bontempelli, Moore, Dawson and Greene in the team. Perhaps it should go to the highest ranked player who currently occupies a leadership role for their team?
he explained it as the player with the best stats in their area, otherwise it might go down to bias or how well a team performed in a season and not by the stats, as the team is made to be
@@papleylivesrentfreeinurhead So Curnow kicking more goals than the other forwards is a good metric for being captain of a side? Bontempelli won the AFLPA best captain award for the second time this year, edging out Moore and Greene. Taylor Walker is also a former 2 time winner of that award. Yet Curnow kicked three more goals then him for the season so he becomes captain? If the point it to make it by the stats, then there needs to be some other way to apply or filter the stats, otherwise the result is clearly bunk.
@@JohnHenrySMH again, the video is called “by the stats”, i understand in no real situation would curnow or rowell be the best option realistically. u make an interesting point tho, wat stats would u see as fit for a captaincy role?
@@papleylivesrentfreeinurhead I think you can still use the individual stats, but have a qualifier that they have to have had AFL leadership experience. So either a captain, vice captain or if there are none in the squad, then someone who has been in the designated 'leadership' group that most teams seem to have. Then whoever has the best stats out of that sub group?
Love the videos, but a mid-def or warranted as winger would be good. Toby Greene should still be captain though ;)
Like a forward type winger + defensive type winger maybe
James Sicily...
Interesting that most of the midfield's dump stat was efficiency. Not sure why exactly we need a dump stat, just go lowest score across all the stats selected, otherwise why select the stat
This side beats the regular AA side every day of the week
Good team.
“There are lies, damn lies and statistics” -Mark Twain
"krhkrgkrkrkkrkrkrrrrkrrrrrgg" -Perry the Platypus
@@FootyA2Z 😂
But love your work
Hey @FootyA2Z I was also in the tipping comp my name is Mia!
Stats done by total or average? Sorry if u mentioned I may have missed it
Goes to show you can get stats to mean whatever you want then to
this guys got something against gryan miers
Dawson didn't play on the wing tho.
Yeah hurts
Already was, wasnt it? AA selectors only seam to.
Honestly this is how it should be can’t help but thinking most of the selections are biased and not particularly deserved
Cal Wilkie play badminton does he?
did u actually read the title or watch the video or did u just have a little hissy fit bc the vid explains the point of it pretty well
@@papleylivesrentfreeinurhead yeah it was a good video. it’s a good way to demonstrate that stats aren’t everything. i was just having a bit of a tongue in cheek joke, not everything is serious champ :)
@@manormuppet44 sorry mate my fault can never tell in these yt comments
miers stiff not to get selected in either all oz side...
Still no Taranto
Sicily stiff down back
Super impressive, until Matt Rowell 🤦♂
You misspelt Errol Gulden’s name
Matt Rowell clearly the best player of all time
😂
This is the All Australian team without VIC bias. Victorian commentators are so brainwashed
gryan robbed 😭😭😭
Love not seeing people who take 5 kick ins every game
Contested Marks is a useless stat
The coaches wouldn’t agree with that ridiculous comment ..
would of thought you’d mention how midfielders sometimes get rested on half back