Robert Younger

Backseat footballer, mathematician and ex-blogger. Performance Data Scientist @melbournefc

[Robert Younger @figuringfooty]

Fair old night!

[Andrew Whelan @andrew_whelan]

New Expected Scores page now available

🟣Explore goalkicking accuracy and shot rating (2021-)
🟣View by player, team or AFL-wide
🟣View player shot charts
🟣Filter shots by season, shot type, distance and corridor/pocket

https://t.co/UthVLuvPjZ

[Robert Younger @figuringfooty]

@UselessStatsAFL @CapitalCityCody Can’t unfortunately. But enjoying the analysis

[Robert Younger @figuringfooty]

@EmlynBreese The issue is the difference in definition between the statistical “expectation” which is the average value of a random variable, and the everyday use which tends to imply some level of fate or deserving. I’ve played with things like “shot value”, but fear the ship has sailed.

[Cody Atkinson @CapitalCityCody]

This week we dove relatively deep into centre clearances, how much they are really worth and how teams move the ball away from the middle.

To do this of course we focused on West Coast.
https://t.co/w123A96cnH

[Michael Lopez @StatsbyLopez]

📢📢
Ladies and gentlemen, the 2024 Big Data Bowl #BigDataBowl
📢📢

https://t.co/HyXr3q0Oeu

[Chew The Stat @ChewTheStat]

Does the outcome of tight matches play mind games on the umps as to which side they should give the votes to? The size of the gap between a 1pt win and a 1pt loss in the below suggests yes, it does.
Using the margin is an obvious yet critical factor in predicting Brownlow voting.

[Max Barry @SquiggleAFL]

Full results are in for 50 expert & media pundit preseason ladder predictions. Summary: Over the long term, Squiggle is going quite a lot better than any of them. I will be reposting this link once per day forever. https://t.co/Sfwbr8IFW5

[SEN 1116 @1116sen]

Champion Data's Daniel Hoyne provides some context around HOW Expected Score as a statistic should be used.

@SENSportsday | #AFL

[Robert Younger @figuringfooty]

@AflGlicko So the team ratings you post on Monday are independent of selected team?

[AFL Live Ladder @AflLadder]

Quite simply the most wide-open battle for finals in #AFL history. The computer found over 230,000 possible permutations of the Final 8.
Expect 5 teams to finish with 54+ and 8 teams to end up on 44-52 points. 48 won't be enough, unless you have both percentage and luck 1/5

[Robert Younger @figuringfooty]

@crow_data_sci @JTGiuliano @WheeloRatings @DeeArmy @The_DeeBrief @attntodeetail @melbournefc @demonland @MelbourneFooty @melbournefans @BreeseEmlyn How much of this is because the ball is in F-half more than the average? These are naturally higher threat, yes?

[DTS @DTS098]

Am really pleased to share this paper as my first piece of authored work. Hopefully it can add to commentary surrounding teams' forward line strategy in the AFL. You can check it out at the link below.
#sportsanalytics #sportsdata #sports #analytics #afl

[Ben Cochrane @BenCochraneR]

What does the average #AFL player look like?

Layering semi-transparent AFL headshots of all 2022 players yields this bloke:

[Robert Younger @figuringfooty]

@CapitalCityCody @alittlefitness @BreeseEmlyn Ah yeah, that might be it. It was coded as if Brayshaw held the ball for half an hour for time in poss stats

[Robert Younger @figuringfooty]

@alittlefitness @BreeseEmlyn How’s this calculated? Derived from event data? I can’t seem to replicate anything that has Melbourne anywhere near this low. Any particular games that stand out?

[Dom Samangy @DSamangy]

🚨 Guide to Sports Analytics 📊

New design & we now cover 4! softwares in-depth (R, Python, Tableau, SQL). In total, links to 500+ tutorials, 250+ data sources, 125+ accounts to follow, 50 books, 40 cheatsheets, 40 papers & more. All in ONE place!! ⬇️

https://t.co/ZeFuzwVPmY

[Robert Younger @figuringfooty]

@MattCowgill @RDublyaBuckland They could break the 2 year drought for Joondalup AFL flags.

[Robert Younger @figuringfooty]

@alittlefitness @crow_data_sci @CapitalCityCody @liammannix A smoothed model is your first step in building an equity based player ratings model ala PR. But you’re improvements for this will definitely be marginal. Love what you’re doing!

[Robert Younger @figuringfooty]

Hard to imagine a worse performance (both onfield and off) than that by the aussies. You can write it off as India being good, but England won’t be pushovers by any means.

[Robert Younger @figuringfooty]

So it’s pretty clear now that England are the ones to reap the first mover advantage of analytics in short form cricket.

They’re incredibly well drilled, innovate (weakest bowler opening the bowling, fewer yorkers at death etc) and set plans to opposition so well.

[Liam 🐦 @crow_data_sci]

2022 is done and dusted! Thought I'd put together a super thread of all the analysis, visualisation and data exploration I got up to this year.