How to pick the perfect Fantasy Premier League team for the 2024/25 season

Fantasy Premier League is back – and that means it’s time to start analysing price lists, studying the fixtures and slowly figuring out our squad for the opening weekend. It’s a daunting task and one that even experienced players struggle with. Happily, our resident expert is here to help you get your FPL teams off to a flier.

Matthew Gregory finished 1,693rd in the world last season and he’ll be here with weekly advice columns – but he’s kicking things off with some handy pre-season articles to help you get your team in a good place from Gameweek One. Today, he’ll look at how you should put together your starting team, suggest some ideas for templates and offer his pointers for identifying the best players with which to start the season.

Once you’ve read through that, it’s also worth checking out his pieces on the best budget buys and on how the new transfer and chip rules will change FPL strategy in the 2024/25 season. But for now, over to Matt…

Picking premiums and testing templates

Picking your FPL team is basically a process of following a single, absolutely colossal decision tree, one that looks completely overwhelming at first glance – but there is a right place to start, and that’s by taking a few different possible templates and playing around with them.

By template, I mean a set-up which combines formation and the ‘premium’ players that you want to include. For instance, there will be a template which sets up to play 3-4-3 with and without the wildly expensive Erling Haaland, and one which ditches him to add an extra top-tier player. There will be a 3-5-2 which puts all of the cash into midfield, and another which loads up in the attacking line and skimps on someone like Mohamed Salah.

Of course, coming up with a first draft for every possible formation and combination of premium players just to compare them would take ages, so we need to start by figuring out which ideas we should ignore entirely.

For starters, I would only build teams which want to start with three defenders every week – in other words, only spend serious money on a maximum of three players for your back line and invest smaller amounts in two defenders who will ride the bench most weeks. This is because defenders simply score fewer points. Attackers and midfielders can get past 200 points over the course of a good season, but only a few defenders will even hit 150. Ben White and William Saliba were the only ones that managed it last year.

So we want to look at building teams which plan to either set up as 3-5-2 or 3-4-3. The next decision is whether we can afford players like the £15.0m monster that is Erling Haaland, £12.5m Mohamed Salah, £10.0m Bukayo Saka and so on. Start by putting them into your team and playing around with things – then trying versions which take them out and see what you can afford in their stead. We’ll get onto specific things to look out for with different players shortly, so factor that in as you go, and pick out the budget ‘enablers’ you want in order to afford better players up the pitch.

A lot of people will instinctively look at Haaland and feel that he’s too expensive, and on the basis of last year’s return he would be. But in his debut season at Manchester City he put up a massive 272 points over the course of a 52-goal season and if he does that again, he’d be essential even his price was even higher. It’s a judgement call that players have to make for themselves, but we’ll look at the tools to make those decisions easier later too.

Similarly, Trent Alexander-Arnold, the most expensive defender in the game at £7.0m, didn’t do enough last season to justify that price tag – but he’s scored heavily before and there’s no reason to think he can’t again. If he hits his best form, he’s a necessary inclusion.

The best way to think of it is that you’re looking to pick out the handful of highest-scoring players over a season who you will pick every week regardless of opposition, and then you are trying to surround them with the cheaper players that make it financially possible to own those super high-scoring players while still scoring plenty themselves.

A final early pointer I’d make is that whether you decide to start 3-4-3 or 3-5-2, don’t skimp too much on the midfielder or striker you’re expecting to have on the bench. There are some great cheap options going like John McGinn and João Pedro, but don’t just take the cheapest possible player for the sake of your team’s economy – there will be injuries, players rested unexpectedly and those bench players will come on eventually. You want one defender and one attacker/midfielder who will be able to pop up with some points. Only the back-up goalkeeper and fifth defender are truly expendable.

Stats matter

The big question then is – how do we know which players are going to go big and which will struggle? Will players that scored a ton of points last year do it again?

There are no guarantees, of course, but stats help us make these decisions. We can take a look at players’ expected goals and assists and try to figure out which are underappreciated and which are overvalued. We can’t know which players will hit form and which will struggle for it – Marcus Rashford last year, for example, was a good buy by the stats but lost his confidence and underwhelmed – but we can play the odds by going for players who are likely to score us points.

Let’s use Alexander-Arnold as an example. His haul of 122 points is poor for a £7.0m defender, and barring a £4.0m back-up defender we’d want all of our players to score more. But looking at the stats tells a different story (I mostly use FBREF’s Opta-powered stats, by the way, as the official FPL website’s numbers often seem a bit questionable).

Due to injuries, Alexander-Arnold was ‘restricted’ to an xG of 2.6 (scoring three) and an xA of 7.2 (actually getting four). If we run the numbers to extrapolate what he’d notch up over a 38-game season rather than the 28-game season he actually had, it tells us that he’d be expected to score three or four goals and bag around 10 assists – numbers not far off those he achieved when he notched up 200 points in 2019/20 and 2021/22.

Furthermore, Liverpool had the third-best defence in terms of goals conceded last year but only kept 10 clean sheets – a good number, bettered only by three teams, but still perhaps less than the third-best defence in the country would normally keep. In other words – the chances are that Alexander-Arnold will, on average, get more clean sheets and assists this season than he did last, and can be expected to score more points. He’s been included in every team I’ve drafted so far.

Erling Haaland’s stats, by contrast, are complicated. Although he scored nine fewer Premier League goals in 2023/24 than the year before, his xG was actually higher – implying that perhaps his first season score was slightly inflated compared to what he can typically expect to achieve. On the other hand, it’s the first year of top-flight football in which he hasn’t very comfortably exceeded his xG, and he usually scores four or five goals more than an average striker would with his chances.

The conclusion? Even well off his best form, he was a 200-point player and seems incredibly likely to do that again. He also missed games with injury, and looking at his overall numbers there is a very strong chance that he goes for 250 points or so once again. So he’s an auto-include, right?

Well, not necessarily. He’s so expensive that dropping him for, say, Ollie Watkins (£9.0m) will allow you to play a couple of other players with 200-point potential (Saka, De Bruyne, Salah and so on), and that will probably be worth more points in the final reckoning. The question is whether there are enough cheap players who will score 150-200 points or so to be able to afford Haaland. If there are, then a team led by the Norwegian is almost certainly the best way to go, but you have to identify those breakout players very accurately.

Playing Haaland is, as it stands, a high-risk strategy, and it may be best to save transfers to bring him in from around week six, when Manchester City’s fixtures get a bit easier, and when you will be more likely to know which cheap enablers you can buy to bring him in. I don’t blame anyone for starting Haaland, but as it stands I’m leaning gently away from it. It’s probably the toughest decision for your Gameweek One team, however.

One more word on stats – don’t fall into the trap of treating xG and xA as gospel. There are several players who have demonstrated over the course of several years that they won’t score as many as they ‘should’ do. Darwin Núñez and Gabriel Jesus are fine examples. Stats can tell us which players are likely to do better, but track record counts as well.

Avoid excessive risks

There are plenty of players in FPL who we can treat as pretty safe bets – Haaland and Saka, for instance, have very high baselines and are unlikely to let you down too hard. They’ll also start just about every match that they’re fit for, which means you won’t get too many zeroes on your points screen when you select them.

But there are plenty of players who are harder to assess but which have high ceilings. Christopher Nkunku is a good example, as is the aforementioned Rashford. Play at their best, and play regularly, and their prices of £6.5m and £7.0m respectively are exceptionally cheap – but there are big question marks over how involved and how good they will both turn out to be.

No FPL team is guaranteed to score points, but you can’t afford to have too many players who won’t play every last game and score regularly. If you gamble on both Nkunku and Rashford, for example, and neither start well (or start at all) and you’ll have at least one player with a great ugly egg next to their name every week. Risking one is fine, as long as you have a midfielder to come in who you feel able to rely on when they’re benched or blank. Playing both is asking for trouble until you know that they’re going to do the business.

In other words, play it safe to start with. The best theoretical team will probably contain multiple players who weren’t expected to score as many as they did and may have a few new signings in there – but guessing which new or risky players will score big is very likely to backfire. Stay safe to start with and pick players you can trust. Some new signings will shine, others will take a while to warm up, and some seemingly good buys will flop entirely, and there’s no meaningful way to know which is which.

Just make sure that every player you have is as close to being a nailed-on starter as possible to avoid too many zeroes, and don’t take more than one big flyer in defence or in attack and midfield combined. One flop won’t kill you and can be transferred out easily enough. Multiple will cost you points and could force you into an early wildcard unless you get lucky. Pick guaranteed starters and pick players you know will score.

Check the fixtures

Another factor you need to consider is fixture difficulty, and you should avoid having too many players in your starting team that run into tough games. For example, Wolverhampton Wanderers have three players that are extremely well-priced – Matheus Cunha, Hwang Hee-Chan and Pedro Neto, all at £6.5m. Over the course of the 38-game season, all three could be great includes in your team, but Wolves have an absolutely hideous run of games to start with, and that is likely to reduce the points they pick up until things ease off.

As a result, none of my draft teams have any Wolves players even though I sincerely believe that I’ll include several of those players, and perhaps all, in my squad down the line. Eventually, they’ll rack up points, but based on their first seven or eight games, it’s unlikely to be right off the bat.

By the same token, Liverpool have a pretty soft start to the season. There may well be the expected uncertainty over how they will go in the post-Klopp era, but their first half-dozen games or so aren’t too brutal and players like Alexander-Arnold and Salah look pretty appealing.

So with Manchester City having a relatively tricky start to the campaign and Liverpool less so, I’m leaning towards a team which includes Salah but not Haaland as the way to maximise my points for the first six weeks or so. When the fixtures flip over, which they do around week seven or eight (depending on how good Crystal Palace and Chelsea turn out to be), I can switch them around, something made much easier now that we can roll more than two free transfers over.

Goalkeepers and an early template

One last, small thing to consider before I show you all the team I’ve got as my favoured first draft so far – ignore goalkeepers worth more than £5.0m. The Alissons of the world simply don’t score enough points more than cheaper goalkeepers to be worth spending the money on.

Generally, I favour a £4.5m goalkeeper but the best options at that price point, like Alphonse Areola, have rocky fixtures, and that makes me lean towards Jordan Pickford at £5.0m as it stands. Everton keep a lot of clean sheets under Sean Dyche and Pickford makes a lot of saves – and on average the record books suggest that Pickford should be worth 30 points or so more than, say, Areola or Mark Flekken over the course of a season based on past form of the clubs and managers involved. That’s enough to be worth an extra £0.5m, but perhaps not if it stops you signing a premium attacking player instead.

With everything I’ve mentioned so far in mind, here’s where I’m at to start the campaign now. I’ll be making plenty more drafts and digging deeper into the stats before the season actually starts and it’s very possible my final team looks completely different (especially once more signings come in, reducing the likely minutes of some of my players and providing tempting high-risk options), but this is a solid baseline team for now, and hopefully gives you a pointer or two:

In order to get Alexander-Arnold, I’ve picked out some rotational £4.5m defenders who I can switch around according to fixtures and a £4.0m in Taylor Harwood-Bellis who will play every week and can hopefully fulfil the Jarrad Branthwaite role from last season. I’ve gambled on Nkunku and, to a certain extent, Rogers, in order to fit multiple premium players in. At £9.5m, Kevin de Bruyne looks exceptionally well-priced - he missed a lot of football last season but his stats per game suggest no drop-off from the point at which he was scoring huge numbers of points. He just needs to stay fit.

I actually think Ollie Watkins won’t be quite as good this season, but Aston Villa have relatively soft early fixtures – so my current plan is to start with Watkins and move him to Haaland, with the requisite sacrifices made elsewhere, around week seven or so. But all that may change if, say, Newcastle United sign a new forward meaning Alexander Isak will play less, or if Ivan Toney looks great in pre-season, things which might make me look for a cheaper option to allow more investment elsewhere.

I could, for instance, move Watkins to Toney and spend £6.5m on a more reliable midfielder than Rogers. Or just move Pickford to Areola and upgrade Rogers to the more nailed-on and reliable John McGinn. If I’m convinced Takehiro Tomiyasu will continue to start at left-back for Arsenal, I’ll probably add him and find £0.5m from somewhere else. These are all things I’ll turn over in my mind as pre-season wears on.

This is just one early draft – make several, based around having Haaland or around having five starting midfielders and so on and so forth, and with different numbers of premium players. They will let you see where you can tweak the team to fit different players in, and give you the basic ideas to work out your ideal team come the start of the season. There are plenty of things that may change, players will get injured or look awful in friendlies, but if you know the basic templates you like now then it will make hitting that submit button for the final time so much easier. Good luck!