McSpoish

4.3K posts

McSpoish banner
McSpoish

McSpoish

@Spoish

Your favourite FPL players favourite FPL player

เข้าร่วม Nisan 2011
716 กำลังติดตาม148 ผู้ติดตาม
Filip Novák
Filip Novák@11filipnovak·
The truth is, the vast majority of the “analytical” FPL managers do very little actual data analysis of players/teams/match-ups. Trust me, I know when I see one. I’d rather call them “elite planners” or something like that. FH weeks lay this completely bare. Which is not a bad thing – there are several ways to play the game, and many are fantastic players. Just don’t dress it up as analytical work with football data, because it’s not.
English
5
0
42
12.6K
McSpoish
McSpoish@Spoish·
@FPL_Brandon Yeah for example say Spurs def is going to be 200% owned, makes a Wolves attacker more appealing
English
0
0
1
93
Brandon
Brandon@FPL_Brandon·
@Spoish By this do you mean factoring in effective ownership?
English
1
0
0
315
Brandon
Brandon@FPL_Brandon·
Free Hit Gameweek 34 edition anyone?
Brandon@FPL_Brandon

I am Free Hitting this week and decided to write a long piece on increasing your exposure to luck in FPL. This is especially important on a Free Hit week. Insightful read if your attention span is not cooked. Topics covered: • Risk in FPL • Risk on a Free Hit • Correlated Variance (CoVar) Why incorporate more risk on a Free Hit than in your normal team? A Free Hit is a great opportunity to add risk because the decisions do not roll over for multiple gameweeks. Everything is contained to a single week, and a single gameweek naturally carries far more variance than a longer stretch of matches. In probability, risk has equal upside and downside, but not equal strategic value. A Free Hit lets you target an extreme outcome in a week of your choosing, often when low-owned picks have higher expected value. Get it right and you can separate from the crowd in one gameweek, get it wrong and the loss is contained to that week. That is why a Free Hit is the perfect moment to embrace volatility and chase upside. How to expose yourself to risk on a Free Hit The most common approach is to go for low-owned punts. This works, but you will naturally end up with some low-owned players anyway because you are taking one week punts. A less common but extremely powerful method is to embrace defensive and offensive stacking of players with correlated outcomes, which I refer to as Correlated Variance, or CoVar. Defensive CoVar is generally stronger than offensive CoVar because defenders from the same team are heavily correlated through clean sheets. They usually either all get the clean sheet or none do. Stacking two or three defenders from the same team increases the chance of an extreme outcome, both good and bad. This gives you the best odds of a huge return, and the strategy is so strong that many fantasy formats add stacking penalties to limit it. In the table, you can see the clean sheet outcome probabilities for Free Hit 13. It is boom or bust, but the odds of hitting a haul favour stacking. Offensive CoVar means stacking two or three attackers from the same team. A single goal can generate returns for both the scorer and the assister within your team. Midfielders also correlate through a clean sheet point, although bonus points are shared. You are essentially betting on one team performing well and needing fewer goals for a big overall return. With two attackers, there are two ways to get correlated returns from a single goal. Add a third attacker and this increases to six possible scoring-assist combinations. That is why triple attacking stacks massively increase the chance that at least two players profit from every goal. The same logic applies if defenders chip in with attacking returns. It is important to highlight that while stacking is powerful, you should still aim to maximise expected points and avoid sacrificing too much just to force stacks. For example, City have the best clean sheet odds, but a defensive triple-up without Haaland gives up too many expected points. I hope you found this piece insightful.

English
15
1
142
24.9K
Andrew Lawes
Andrew Lawes@FPLNightmare·
@Spoish If you get it right, then yes, but getting it right is so much harder when the consequences of getting it wrong are so much more pronounced
English
1
0
0
66
McSpoish
McSpoish@Spoish·
@FPL_Barbossa I will ask you to retract the use of the emoji in question.
English
0
0
1
95
JC 🚀
JC 🚀@FPL_Spaceman·
Sadly we are now down to Analytics Elite 63* after one manager decided to hang up their spreadsheets and retire from FPL
English
2
0
20
3K
JC 🚀
JC 🚀@FPL_Spaceman·
What did the Analytics Elite 64 and # Elite 64 do in GW29?
JC 🚀 tweet mediaJC 🚀 tweet media
English
2
3
40
6.2K
McSpoish
McSpoish@Spoish·
@FF_Trout Someone needs to share the Solio viz settings with you :D
English
0
0
1
35
Fantasy Football Trout
Fantasy Football Trout@FF_Trout·
Some extra graphical context: In recent seasons, team value has never dipped below £0.2/m for a sustained period of time. With my pathetic team value of 100.6, I'm hoping it stays this way 😭
Fantasy Football Trout tweet media
JC 🚀@FPL_Spaceman

Team value in FPL has never been less important than right now. In previous seasons, you could consistently expect around 0.30 to 0.35 extra projected points per gw per 1mill of budget (peaking as high as ~0.42 at the start of the 22/23 season). Currently it is around ~0.17

English
3
4
36
11K
JC 🚀
JC 🚀@FPL_Spaceman·
@FFScout What is the source of these odds? Are they FFScout model projections?
English
4
0
47
3.8K
Fantasy Football Scout
Fantasy Football Scout@FFScout·
🧱 𝘾𝙇𝙀𝘼𝙉 𝙎𝙃𝙀𝙀𝙏 𝙊𝘿𝘿𝙎 📍 GW21
Fantasy Football Scout tweet media
English
2
6
108
31.5K
McSpoish
McSpoish@Spoish·
@FPL_Spaceman A lot of words for "I want to qualify for The Spicy League next season"
English
0
0
0
37
JC 🚀
JC 🚀@FPL_Spaceman·
Projection modelling is much more of a framework than a prescription on what decisions to make. With my own beliefs, Haaland projected slightly ahead of Bruno. Many people captained Bruno because they thought he was better. I did it because Haaland (c) is not how I want to play
English
1
0
7
1.4K
JC 🚀
JC 🚀@FPL_Spaceman·
Bruno, Haaland, Saka, Salah, Mbeumo were all viable captaincy picks this week imo If you bring your own beliefs that MCI are better and MUN are worse than this then sure, Haaland will project more points on average. But you still don't *have* to captain the top projected player
Rob T@robtFPL

📊 Solio projected #FPL points and picks | GW14 A look at some of the top projected players on our live projection model & an example 'free hit' team Very close for captaincy this week - market numbers will continue to update until deadline 👉 fpl.solioanalytics.com

English
13
2
36
9.6K
McSpoish
McSpoish@Spoish·
@FPLFella @fplblackbox_az It's not this, the models don't update that quickly. And in that week only four of the AE64 captained Haaland, the rest were split between Saka (31), Bruno (23), and some poor misguided fools who still had Salah (5).
English
0
0
0
17
Fella
Fella@FPLFella·
@fplblackbox_az Recency bias. He got 11 points in GW13 and everyone jumped on him. Data probably changed a lot because of that.
English
1
0
2
1.2K
Az
Az@fplblackbox_az·
Wild stuff. Struggling a bit to understand why Fernandes was so favoured this week? He had Everton (H) 2 weeks ago and Haaland played Newcastle and the Bruno hype was nowhere near what it was this week
FPL Jurgen (Sam)@FPL_Jurgen

I was the only manager in the whole of the analytics elite 64 to captain Haaland. Genuinely couldn’t believe it when the game refreshed. Not a brag as Bruno still to go, but just find it crazy how strict on the model people must be for not even 1 other person to captain him today

English
27
1
140
75.2K
McSpoish
McSpoish@Spoish·
@calm_strategy I fortunately was busy the entire Saturday morning meaning my captain stayed on Saka. Now that's skill.
English
0
0
0
52
Calm
Calm@calm_strategy·
@Spoish Yeah, true, I wrote in a rush. Principle is still, larger EV differential than comfortable with
English
1
0
0
132
Calm
Calm@calm_strategy·
GW11: 52 Rank: 68 I had a few close decisions this week. Dubravka v Vicario in goal. Saka v Haaland for captaincy. I solved the goalkeeping one by choosing the one who wasn't facing one of my attacking players (mentioned in a previous GW review) The captaincy one can bring me to a deeper discussion. I said in one of my responses to questions last week that I was likely to captain Saka, but that I thought Haaland was the slightly better choice. So there are 2 questions: 1) Why did I change? (addressed here) 2) Why would I ever captain someone who I don't believe to be the best choice? (I'll cover this next GW) The main takeaway / tldr is that close decisions don't matter (they are controlled by luck), but by the time the deadline came along, it was no longer as close. We speak of FPL being a game that is a mixture of skill and luck. The skill portion is what do we do before the GW starts. How do we set our team up? And 90% of this skill benefit is going to be in the obvious things. Spending your money, playing your chips at the right time, picking players from good teams that start, replacing injured players, not taking lots of -4 hits, captaining your best player each week*. You can critique yourself based on these choices. The luck aspect is then, what actual outcomes occur. This is never something to be upset / angry at yourself about. You're a team manager, not a player controller. So we can't control what happens here, but we *can* control how much we are subjected to it, based on how different our team is from the next person. To win FPL / any game with a large player pool we want to be heavily subjected to luck to avoid being stuck in the crowd. So to really simplify, Saka and Haaland had equal expected points earlier in the week, and this was giving an opportunity to have more risk / more 'luck' v the field when they would mostly be Haaland (c). However, by the time the deadline came around the EV difference was about 0.5 (which doubles to 1 when captained), and I deemed this to be too much of an EV / 'skill' sacrifice just for the sake of being different. (I will cover this in much more detail next GW). Looking forwards with 2FT, Vicario remains an issue and I can now add Reijnders to that given his consecutive benchings. There are always possibilities of injuries during the international breaks so I won't make any early moves. Assuming injuries don't happen, I'm likely to try to maximise the value in my starting 11, and that would mean a goalkeeping transfer. I'm most likely to go to Petrovic, but Verbruggen or even Alisson if confirmed fit are options.
Calm tweet mediaCalm tweet media
English
31
3
136
39.9K