Just a note to say that this is likely to be my last article and probably my last post for a while at least. There are a few things in my personal life that I need to give full attention to and I don’t want to phone in content or imply FPL advice when I know I’m not giving it the thought that I normally would.
It’s been an absolute blast producing FPL content over the past 10 years - I wanted to try and contribute something to a community which has given a lot to me and I hope I’ve gone some way to doing that. Thanks to everyone, especially @FFScout, for allowing me that opportunity.
I’m more grateful than I can really express for the incredible support and experiences I’ve had, and for all the amazing people I’ve met along the way. I know X isn’t a great place for sentimentality but you guys truly are the best and I didn’t want to disappear without letting you know how much it has all meant to me.
Thank you. Good luck.✌️
https://t.co/qxdfPwMCDF
The FPL Champion’s League Final Top 3:
1. Matthew Martyniak (2012/13)🥇🐐
2. @ASC_FPL (2023/24)🥈
3. @MarchSimon (2014/15)🥉
Looking forward to 2025/26 FPL Champion @TheFPLStudy hopefully joining us next season, huge congratulations to him! 🏆👏
No worries, thanks for the reply!
Ha, it has been over a decade now, so some might say I’m due…
I agree with the data issue, I think one way the models get around it, from what I’ve seen at least, is by favouring particular profiles of players with very repeatable characteristics i.e. nailed on, play 90, set pieces (esp pens), defcon... so the dynamism/scarcity of the data becomes less of a factor because the foundation for scoring points remains solid.
As a result, they tend to like the likes of Szoboszlai, VVD, Rice, Bruno and are generally less keen on the likes of Wilson, Bruno G, Neto etc… who might appeal more if you were going by the eye test.
That said, it seems like the people who use models the best A: *really* commit to the output and B: are incredibly patient, which suits the kinds of incrementally-scoring players the models tend to favour. That data+playing style combination means they often do really well consistently, but actually winning FPL often comes down to a few big logic-defying risks that pay off disproportionately (esp in a game of 12m+), so it'll be interesting to see how that tallies up in the future.
Btw I'm pretty sure is was a 1-1 draw between Saints and Chelsea that I watched so, not the best but at least I didn't feel like I was missing too much!
@FplVeteran_
That particular season, I watched maybe a match and a half, but I should mention that that was mostly down to the circumstances that year, I had watched a lot of football before then (and since, of course).
Nevertheless, it’s been characterised a bit as a victory for ‘numbers over grass’ and it sort of was (the data was there, even in 14/15, albeit a lot less refined than it is today) but I think the bigger advantage was psychological in that, when you’re just looking at the numbers and not actually consuming football as entertainment, it’s a lot easier to make cold and ruthless FPL decisions!
It probably is possible to win FPL using solely a data model but I still think it would be a disadvantage to try and do it without any sort of footballing lens to contextualise things through. Even then you’d still need to be capable at squad/budget management, managing the finer risk margins etc. Then, even after that, there’s still the huge luck factor.
@FPL__Raptor and I had a good discussion about a few of these topics a little while ago so, if you are interested, give it a watch:
https://t.co/y0LB2QZk8O
Is more always better? 🏦
Former FPL champion @MarchSimon examines whether collecting as many transfers as possible is truly the best strategy 🧠
https://t.co/5J3nCXLWW3
🤦♂️ Want to avoid FPL pitfalls before they arise?
🖱️ CLICK | Former champion @MarchSimon explains how ‘pre-mortems’ can make us better FPL managers 👇
https://t.co/WEDu7qLbH2
I’ve never been able to figure out a good answer to this. It probably makes sense to weigh the more recent data higher, whether it contradicts the wider sample or not, but then the question becomes whether either sample size is big enough to be meaningful, especially when it comes to something as dynamic as football?
I mean, you can watch two games of football and feel like you’re watching two different sports at times.🤷♂️
@FPL_Martin68 Cheers, you too! I've been dodging points all season with that '3rd' striker spot so I feel like I have to give him AVL & IPS at least. Can't say the signs are looking great though...
@FPL_Boxer That's tough, based on GWs 10 & 11, the 2nd option. Based on GWs 12 & 13 the 1st option. I'd be tempted to go TAA+R/SL now because you could still switch in a couple of GWs if you wanted to
@mike_jeski Good call! I think my problem is that I've got too much money tied up in Solanke when I could have somebody cheaper who actually scores goals (like Raul)
@BillyBrightFF Solanke would probably go first atm (maybe for Cunha) as he's not even been getting chances - that said, I'm tempted to give him the next 2 and see if things improve.
@RobertoHollis9 It sounds like he probably will. I've got a benching dilemma otherwise though so I've convinced myself that the slight doubt it reason enough to make him 1st sub