Everything you need for FPL Gameweek 4...
Rankings, team guide, fixtures, player data and more...
Hello and welcome to the latest edition of Fantasy Gameweek. This week we have included everything you need to make decisions for Gameweek 4 of Fantasy Premier League. This includes the FGPE 100 - our ranking of the most valuable players so far - but also a look at forthcoming fixtures plus a dive into all the underlying team and player data now that we have three Gameweeks of matches to analyse. Thank you to Statsbomb for providing the data.
We have tried to provide what you need to judge the probability of a player getting you points but also the pay-off they will give you. It’s all in a short and sharp format and includes:
-The FGPE 100 - the most valuable players in FPL so far
-Team analysis - which attacks and defences should you target
-Fixture guide - who has the best fixtures to target
-Player data - who is under or over-performing with their FPL point relative to their performance
-Important facts from FPL including player ownership and price rises
-Players of interest given all of the above
Let’s begin with the FGPE 100. You can use this to monitor who is delivering value for your team, who is hurting it, and what sort of pay-offs you should be targeting from players. It measures points per million and performance relative to everyone in a player’s position. It is our equivalent of a dividend yield or PE ratio in finance, which are vital to analysing the value of a share price. A score of 1 equates to a player delivering 1 point a week for every million pound and delivering the most points in their position every Gameweek they have played. Rodrigo’s score is very, very good. He is the overall top scorer in FPL so far and features in the picture above.
Next up, let’s compare how teams are performing so far so we can get more of a flavour of which attacks and defences to target. The first chart is a guide to the best and worst attacks by looking at goals scored but also xG - expected goals, a measure of the quality and number of chances they have created.
The x-axis is goals scored per game and the y-axis is xG per game. Tracking the two means we can see who is scoring goals in line with the chances they are creating and who is not. The best attacks are in the top right quadrant - they have scored the most goals and have the highest xG, ie they have created the best and most chances. The worst attacks are bottom left - they have the least goals and smallest xG.
Teams in the top left are those scoring goals at a lower rate per xG than the rest of the league, those in the bottom are scoring goals at a higher rate per xG. There could be a reason for this - bad or good finishing, for example - or they are due some regression to the mean. So, in simple terms, Manchester United have not scored as many goals as they should have so far and Leicester have scored more. A word of caution, it’s obviously still early days in the season so there is still a lot to learn about these teams. Bournemouth, for example, have played two of their three games against Manchester City and Arsenal.
Next it’s the same for defence. This is a look at goals conceded and xG conceded. The numbers on the axis run in reverse this time so the best defences are still in the top right and the worst in the bottom left.
To summarise these tables, this is what they suggest and what it could mean for FPL purposes:
Best attacks (buy their attackers, avoid defenders against them): Man City, Arsenal, Leeds, Tottenham, Newcastle, Brentford
Worst attacks (avoid their attackers, play defenders against them): Aston Villa, Chelsea, Wolves, Forest, West Ham, Bournemouth
Could be about to get better: Manchester United, Everton, Liverpool, Crystal Palace
Could be about to get worse: Fulham, Southampton, Leicester City
Closest to league average: Brighton
Best defences (buy the defenders, avoid attackers against them): Arsenal, Man City, Brighton, Wolves, Tottenham, Leeds
Worst defences (avoid their defenders, play attackers against them): Manchester United, Bournemouth, Southampton, Aston Villa, Leicester City
Could be about to get better: Brentford, Chelsea, Leicester City
Could be about to get worse: Nottingham Forest, Newcastle
Closest to league average: Everton, Crystal Palace, Fulham, Liverpool, West Ham
Now let’s look at the fixtures for this week and which could be worth targeting. Tom touched on the flaws in the official Fantasy Premier League fixture guide in his latest superb piece, which you read here. Using the guide above we can see where there are interesting match-ups, for example where the best attacks are facing a bad or average defence. Target the teams in bold, avoid the italics
Man City attack v Palace defence
Arsenal attack and defence v Fulham attack and defence
Liverpool attack v Bournemouth defence
Tottenham attack and defence v Forest attack and defence
Brentford attack v Everton defence
Man United attack v Southampton defence
Leeds defence v Brighton attack
There are some interesting and popular positional groups not on that list: Man City defence, Brighton defence, Leeds attack, Wolves defence and Newcastle attack.
Let’s now look at some player data.
First, here are the top 10 players for xG and xG assisted according to Statsbomb. This shows the total number of goals these players are expected to have been involved in either through goals or assists. Alongside that is the actual number of goals they have been involved in.
Source: Statsbomb
Next is post-shot xG (PSxG). This measures the quality of a goal-scoring opportunity after a shot is taken, unlike xG, which is before. This therefore considers the quality of the shot or finish, a useful factor when considering attacking players in FPL. Alongside that is the actual goals they have scored so far.
Source: Statsbomb
Finally, these are the top players for what Statsbomb calls OBV, On Ball Value Added, which tracks how much a player contributes to a team’s chance of scoring a goal and stopping the opposition through their play, including passing, dribbling, shooting and defensive contributions. It’s a great way to track how much players are contributing overall. I have excluded goalkeepers from the list.
Source: Statsbomb
Before we get into the round-up of what all this shows, just to flag that you can find important FPL information about player ownership rates, player price changes and the number of teams transferring players in and out here. Ownership is important for understanding which players are differentials for you while weekly transfers in and out show who is popular, who is not and who could change in price (price changes are linked to demand).
So, to finish, some quick thoughts from all this on players of interest:
Budget-options Rodrigo, Pascal Gross and Aleksandar Mitrovic all merit their place high up the FGPE index according to the underlying data (Gross and Mitrovic are also just outside the top 10 on xG and xG assisted too). However, they all face tricky fixtures this week. For Rodrigo I would recommend holding off on bringing him in until Leeds play Everton in Gameweek 5. For Gross, wait until Brighton travel to Fulham in Gameweek 5. As for Mitrovic, I would avoid until a Gameweek 8 trip to Forest. They play Arsenal, Brighton, Spurs and Chelsea before then.
Liverpool players have been frustrating so far but you should stick with Mo Salah, Trent Alexander-Arnold and Luis Diaz for at least one more week if you have them. Their underlying data suggests points are coming. Andy Robertson’s data is not great. He ranks 39th among defenders for xG and xG assisted according to Statsbomb. There is also a risk of rotation given that Konstantinos Tskimas has come on for him in the last two games. However, taking him out before they play Bournemouth at home is a big risk.
Wilfried Zaha has looked good for Palace and his quality shines through in the data as well. Palace’s fixtures look pretty tough over the next few weeks but they play Brentford at home in Gameweek 5, which could be a moment to bring him in.
Newcastle offer intriguing FPL options through Callum Wilson, Allan Saint-Maximin as a midfielder and Kieran Trippier as a defender, according to the underlying data at least. Fabian Schar also remains high up the FGPE after his Gameweek 1 exploits but has offered little attacking threat since his opening day screamer (his xG and xG assisted is just 0.08 for the season while Trippier’s is 0.43). However, Wilson is now injured and his future role uncertain given the likely signing of Alexander Isak. For the other two, wait until at least Gameweek 6 before bringing them in. Newcastle play Wolves and Liverpool away next
Some names standing out in the data who are yet to deliver significant points but should be on your radar: Bruno Fernandes, Anthony Gordon, Tomas Soucek, Neco Williams, Joachim Andersen, Diogo Dalot, Aaron Cresswell, Said Benrahma, Rayan Ait Nouri, Morgan Gibbs-White, Trent Alexander-Arnold, Mason Mount, and Bukayo Saka. Fernandes and Gordon are the real underperformers according to the data. Fernandes has an xG and xG assisted of 1.51 with 6 key passes and 5 shots, but has no goal or assist. Gordon is at 1.17, 3 key passes and 10 shots but also has no returns. Only Mitrovic, Rodrigo and Haaland have had more shots than Gordon and they have 10 goals between them. Fernandes is priced at £9.8 million but we can see from the FGPE that given the quality of midfield scoring he needs to deliver significant returns at that price to offer value right now. One assist, for instance, would not be enough. Gordon is really interesting given his £5.5 million price. However, the uncertainty around his future - he has been linked with Chelsea - plus Everton’s brutal run of fixtures (Brentford away, Leeds away, Liverpool at home, Arsenal away) means he is just one to keep an eye on at the moment.
The leading players who have over-performed the data include Mathias Jensen, James Maddison, Granit Xhaka, Josh Dasilva, Jack Harrison, Ivan Toney, Che Adams, Rodrigo and Gabriel Jesus. An important note here - many good players will outperform precisely because they are good players in good teams. A good finisher, for example, is likely to outperform their xG and help the players that pass to him outperform their xG assisted. Maddison is over-performing because he scored a stunning free-kick on Saturday, but that quality is part of the player he is. The same players to under-performers, Fernandes, for instance, has been a victim of the malaise around him at Man United for the last year and some wild shooting of his own.
William Saliba and Che Adams have soared up the FGPE after their goals in Gameweek 3. Adams has played only 120 minutes all season and is now being linked with a move to Everton, so keep him on your radar for now. Saliba offers a cheap way into the Arsenal defence than Zinchenko but his xG and assisted xG is just 0.05 (a reflection of how good his goal against Bournemouth was). Zinchenko’s is 0.39.
Sticking with Arsenal, Jesus and Gabriel Martinelli have been popular picks so far and feature in more than 80 per cent and 40 per cent of teams respectively. However, Martin Odegaard is now just 4 points behind Jesus and 1 behind Martinelli after his two goals against Bournemouth. His xG and xG assisted is still well behind those two, however. It is 1.03 compared to 1.74 for Martinelli and 2.57 for Jesus. The data suggests that they remain the better options.
Manchester City are good at football, have good players and a nice run of fixtures - Palace at home, Forest at home, Villa away. A triple-up of their players is a sensible play if possible. Erling Haaland still makes sense over Harry Kane
That’s all for this week. If you enjoy Fantasy Gameweek please help us to grow by sharing it and spreading the word.