@jaseziv.bsky.social
1.6K followers 410 following 60 posts
Creator of various #rstats libraries including {worldfootballR}. Various analytical musings. Squiggle competitor. Views my own dontblamethedata.com | https://github.com/JaseZiv
Posts Media Videos Starter Packs
Pinned
To all the users of {worldfootballR} and the amazing connections I’ve created through that vehicle, I’m sorry to say that {worldfootballR} will no longer be maintained. I’m incredibly proud of it and incredibly thankful all the support I’ve received over the years! github.com/JaseZiv/worl...
Retiring from maintaining `worldfootballR` · Issue #454 · JaseZiv/worldfootballR
I have no spare capacity to continue to maintain this library so I will be taking a step back. Happy for anyone to take this as the bones of a potential version 2 of it and make it their own. Thank...
github.com
Thanks for the kind words James! Glad to hear it helped.

So far no takers, but hopefully someone picks it up. The challenge will be managing the increasing aggressive blocking of web scraping of some of the sites worldfootballR was getting data from
A few of us are going again this #nbl season and seeing how our models perform against each other. You can track our progress here www.dontblamethedata.com/tipping_comp...
Don’t Blame the Data - NBL Tipping Competition
www.dontblamethedata.com
Huge thanks to Max @squiggle.au for yet another massive year's effort! He puts in a tonne of thankless work so some of us footy modellers can test our models against each other. Was pretty stoked to take out the accuracy award and come second in the margin this season squiggle.com.au/leaderboard/
Leaderboard - Squiggle
squiggle.com.au
Nah that might be a me error.. will investigate over the next day or so
Sorry for your loss Tony ❤️
So happy it could provide value! Thanks for sharing ❤️
Hey mate, it will continue for now, but they’ve made some backend changes so I won’t have a fix for that until a week or two in to the season unfortunately
Love hearing these stories Chris and it’s no doubt what kept us going for so long!
Thanks so much Mark and glad we have been able to add some value!
Thanks John, and thanks for the very public support you showed!
To all the users of {worldfootballR} and the amazing connections I’ve created through that vehicle, I’m sorry to say that {worldfootballR} will no longer be maintained. I’m incredibly proud of it and incredibly thankful all the support I’ve received over the years! github.com/JaseZiv/worl...
Retiring from maintaining `worldfootballR` · Issue #454 · JaseZiv/worldfootballR
I have no spare capacity to continue to maintain this library so I will be taking a step back. Happy for anyone to take this as the bones of a potential version 2 of it and make it their own. Thank...
github.com
What version of worldfootballR are you running?
Fbref are aggressively blocking scraping activity now… nothing we can do unfortunately. Need to respect their terms.
Fbref are aggressively blocking scraping activity now… nothing we can do unfortunately. Need to respect their terms.
The dates issue should now be fixed in version 0.6.8.0001. Install the latest version from github. Reach out if there's any issues.
Looks like a bug with some of the date fields being wiped due to TM displaying their dates differently. Will have a fix over the weekend. Sorry for the inconvenience
I will however look into tm_player_bio() hopefully tomorrow to see what’s happened there
Hey Chris, no it’s not you… TM have become a bit more aggressive in blocking scraping activity so it’s likely that… I’ve found creating loops and increasing time between player page loads to 15 seconds goes some way to address this, but it’s also not perfect. Reach out if you need a working example
If you’re getting that error, you’re likely in their ‘dog house’ for a period of time, due to excessive page loads. I think historically this might have been 24 hours. Wait it out.
That certainly could be interesting. I suppose my follow up question is how to estimate alpha? 'General good kick' seems like you'd also need to control for a number of factors, which could also lead to small sample sizes?
Have also thought about the effectiveness of this but small sample size keeps me from doing any heavy lifting
Haven't posted an update for a while but given the angst getting player valuations are causing, we've pushed an update to version 0.6.8, allowing users the ability to loop through each team instead to fix issues a little easier as they arise #worldfootballR

jaseziv.github.io/worldfootbal...
Have hacked something together (not yet ready to be pushed to a version update of worldfootballR) but could share some data from this season... DM me where you want me to send the data and whether .csv or .rds