BH
643 posts


I am the best FPL manager over the past few years.
Decision-making, not data models, is what determines your outcomes.
I wrote some reflections on what fantasy sports teaches us about that.
decidership.com/p/decision-ski…
English

I would have gone 10ATB if I could

elevenify@elevenify
Optimal Free Hit 34 is 5ATB btw don’t let any of these content creators or models gaslight you into thinking ‘goals’ exist
English

GW34: Free hit ramble
Rank: 10
Last week I played my bench boost in a very template way, as just being on bench boost was the differential I needed to push higher up given the teams around me.
It's a good situation to be in, however in terms of what to do it became pretty complex.
First has a similar team to me, 44 extra points and an extra chip that I don't have (triple captain) so if we are to be realistic you've got to say its unlikely. Infact, all 9 teams above me were also on bench boost last week (so all very similar teams).
Now, I could just play very solidly from now on and coast in this position. I'm top of the elite 1k league ('the top 1000 managers of all time'), and were I a content creator I would be top of that league too. I could coast because I can pretty much guarantee every other player in the top 10 of the elite 1000 league will captain Fernandes. Going Salah if I was only in this league would be reckless. But, at the end of it all, you don't get here very often, so I think I am obligated to still try.
Taking risk at the extremes in rank has asymmetric payoffs. Lets say TC is worth 8 points so I'm down 52. If I gain 50 on those around me, I move up to 2nd for a gain of 8 places. If I have the same risk go the other way equally (so -50 points to those around me), I drop to 200th so 190 places.
For people who play for the highest average rank across seasons (most content creators, most 'who is the best manager' review systems) its obviously a bad idea to take those risks. But I believe that view is too simplistic.
To me the difference between 1st and 2nd is worth more than between 24387 and 24388, even though its still a single place.
So the way I approach the problem is to think of it like a poker tournament, where there are 'prizes' that are heavily weighted to the top of the rankings. This is how you can justify taking choices that would seem 'irrational' to those chasing the best average rank. You are optimising for a different payoff to them.
I have actively sabotaged my standings in various best manager rankings multiple times. For example when I was top of the 3 year rankings, I was taking picks to cover the choices of the teams behind me rather than the picks I really thought were best which dropped my average ranking a bit, but I became more highly correlated with them (and so ended on top). When I was top of the massive data rankings on fplreview, it wasn't because I'm "so much better at optimisation / analytics" than @sertalpbilal , @FPL_Spaceman , @FF_Trout ' etc that people have said before, (I'm not at all, they are all great - at best I would get to the same result they do with way more time spent), its just that I took a hard, slightly less-optimal line on purpose to differentiate myself and over time the model moved towards my picks.
Now, I'm going to say to start, that I dislike my FH team quite strongly. If I was at a middling rank and able to choose freely, I would likely have ended on a 451 or 541, but my job instead became, work out what the average free hit will be (assuming the 9 above me will do this), and then work out where I can deviate / go against it.
Given the fixtures and who had things to play for, I think a triple up on ARS / LIV / TOT was at least the guide. I think Virgil / Gabriel were required, but beyond this:
Predicted thought process --> My thought process
Best captain: Fernandes --> So I go Salah
Best goalkeeper: Raya --> Try to have more ARS attack instead (Havertz / Eze / Rice)
Safe Liverpool minutes: Gakpo / Szobo / Wirtz --> Take the risk with Isak
Best of the rest: WHU DEF --> Go EVE attack (Ndiaye)
So yeah, I would have tinkered more (I basically ran out of time), and the whole team does feel a bit off for me, but I think this is just the part of me that is hard to quiet (knowing that in all likelihood I've made my position worse when I didn't have to) but I hope at least I've been able to explain my reasoning as to why I've done it.
Best of luck to you all this week



English

If you need help planning for the upcoming Blank and Double Gameweek fixtures, you can find out how to get a copy of my FPL Transfer Planner spreadsheet here:
fantasyfootballhub.co.uk/ben-crellins-f…
English

Blank and Double GW schedule:
🔥DGW33 🔒
- 2 fix for BOU, BHA, BUR, CHE, LEE, MCI
❄️BGW34🔒
- 0 fix for BOU, BHA, BUR, CHE, LEE, MCI
🔥DGW36❓
- almost certainly 2 fix for MCI
- very likely 2 fix for CRY
- slim chance of 2 fix for BOU
🔥DGW37❓
- slim chance of 2 fix for CRY
- slimmer chance of 2 fix for MCI
❄️BGW37❓
- slim chance of 0 fix for BOU
So in GWs 36+37, we'll either get:
- MCI & CRY Double in GW36 (around 90% likely)
- MCI & BOU Double in GW36, CRY Double in GW37 and BOU Blank in GW37
- MCI & CRY Double in GW37

English

@BenCrellin horrible planning ben
top 10k off the cards now unfortunately
English

Two teams in the top 50 own Salah.
Two teams in the top 50 with Salah 🧢
I'm one of them..
I'm scared.
FPL Physio@FPLPhysioo
GW29 OR: 46 🌎 Transfers: Enzo & Haaland -> Ekitike & Salah. Captain: Salah 🪦 + = 👑
English

@GianniButtice What models? 👀
I definitely think it's between Haaland, Ekitike & Fernandes at this point. Probably in that order.
English














