Luke Jerdy is an actor with a (sometimes!) unhealthy obsession with FPL. He has been playing seriously for the last 2 seasons, finishing 9k in 18/19 and 30.7k in 19/20. He also hosts co-host Triple Captain, an FPL podcast which started last season. He loves stats based research and the ‘moneyball’ approach to picking his team.

The winner of Fantasy Premier League 2019/2020 scored a total of 2557 points. That’s an average of 67.29 points per Gameweek (you divide 2557 by 38). To get to that total, you need each of your 11 players to score an average of 6.12 points per game. The million dollar question is: how can you ensure of this happening? Well, you can’t of course. You can’t control football for goodness sake! Are you mad? No? Ok…well…how can you skew the odds of each of your players achieving those points per week as much as you possibly can? By picking the right players of course! Easy enough? Clearly not.

To give myself the best chance of success this season, I’m going for a Maths based approach when picking my team. At least nine of the players I pick to begin with will be down to how many points per 90 minutes they have scored on average in FPL. Consistency is key, the more data the better – so three seasons of consistent scoring would be a great amount. If a player has scored consistently highly over three seasons or more, the probability of them scoring highly for a fourth or fifth season in a row is increased. Ideally I want as few variables as possible.

Let me give you the formula. Jamie Vardy is a perfect example: over the last 5 seasons (huge amount of data!), he has played a total of 14,940 minutes and scored a total of 939 points. To get his points per 90 minutes, first of all we divide 14,940 by 90, which equals 166. Next, we divide his total points (939) by 166, which equals 5.66. So, over 5 seasons of football, Vardy scores an average of 5.66 points per 90 minutes. Not quite hitting the 6.16 mark, but over 5 seasons that is an incredibly consistent performer. If we take his data over the last 3 seasons, he scores 5.67 points per 90. What can we expect him to score this season? Round about 5.67 points per 90 minutes. See where I’m going with this? Ideally I want those players I can rely on as much as possible. Vardy is the ideal candidate.

Where it gets interesting of course is with the captain pick each week. But more on that later. So, at least nine of my players will be picked according to this approach, then I’ll probably chuck in one or two ‘wildcard’ picks – perhaps a player from a promoted club, a new signing, or a player who has a high points per 90 score with limited data. We shall see…

Why am I choosing to do points per 90 minutes instead of points per game? Simply because some players haven’t played enough minutes, so over a 38 game season their ‘points per game’ (if I divided their total points by 38) would be quite low. A perfect example of this prior to 19/20 was Anthony Martial – his points per game over the two seasons before was 3.25, due to him playing less minutes. His points per 90 however was a whopping 6.98. If you knew he was going to start every week (like we did after the first few Gameweeks of last season), you could be pretty sure of a good return. As was proved last season with his 200 point return. Just remember the more data the better, so someone who has only played say 10-15 games in the Premier League and scored highly, the data wouldn’t be as reliable as a player who has played consistently for three seasons. I don’t want to have four or five players in my team who aren’t guaranteed lots of minutes.

So, who do I pick?! I’m going to run down the highest points scorers (per 90 minutes) for each position at each price point and *try* to come up with a team for the season who *should* score me around 65-68 points per 90 minutes. Of course, nothing is certain in FPL and this isn’t necessarily a winning formula. I just wanted to cut out as much ‘risk’ as possible.

For the purposes of this approach, I will be ignoring the fact Manchester City, Manchester United, Burnley and Aston Villa blank in Gameweek 1. I’m looking at a team for a whole season of football. A ‘set and forget’, if you will. I will be using this approach for GW1 too but this isn’t about worrying how I will get Kevin De Bruyne and Bruno Fernandes in my team for Gameweek 2 or 3. There are plenty of articles elsewhere for that. I will also be ignoring fixtures – again it is fairly easy to adjust the approach with fixtures in mind. Here we go…

Goalkeepers (highest scorer for each price point plus notable others)

6.0m

Allison – 4.5 points per 90 (over 2 seasons)

5.5m

Pope – 4.44 (2 seasons)

5.0m

Guaita – 4.00 (2 seasons)

Fabianski – 3.85 (3 seasons)

4.5m

Ryan – 3.5 (3 seasons)

4.0m

Nyland – 2.51 (6 matches)

As you can see, there isn’t much to choose between the Goalkeepers (which we kind of knew already). There aren’t many options at 4.5M at all this year and I haven’t included the options from the promoted clubs. Unknown quantities aren’t what we are focusing on right now!

Defenders

7.5m

Alexander-Arnold – 6.31 (2 seasons)

7.0m

Robertson – 5.6 (2 seasons)

6.5m

Van-Dijk – 5.11 (2 seasons)

6.0m

Alonso – 5.57 (4 seasons – caveat with Alonso is that he is no longer a guaranteed starter unfortunately!)

Laporte – 5.2 (2 seasons – 19/20 injured for most of it)

Doherty – 4.68 (2 seasons)

5.5m

Aurier – 4.32 (3 seasons, limited minutes each season but this is promising if he continues as first choice)

Van-Aanholt – 3.93 (3 seasons)

Stevens – 3.82 (1 season)

Boly – 3.79 (2 seasons, injured for most of 19/20)

5.0m

Davies – 4.03 (3 seasons, very few minutes in last 2)

Egan – 3.76 (1 season)

Bellerin – 3.7 (3 seasons, small number of minutes in last 2)

James – 3.64 (1 season of limited minutes)

Saiss – 3.51 (2 seasons, 1 with limited minutes)

Keane – 3.34 (3 seasons)

4.5m

Walker-Peters – 4.1 (15 matches over 2 seasons)

Lamptey – 3.78 (7 matches)

Vinagre – 3.74 (15 matches over 2 seasons – should start as Jonny injured for Wolves)

Lascelles – 3.3 (3 seasons)

Justin – 3.12 (11 matches)

Fredericks – 3.04 (2 seasons, limited minutes)

Webster – 2.97 (1 season)

Taylor (Burnley) – 2.77 (3 seasons, limited minutes for 2)

Unsurprisingly Trent Alexander-Arnold comes out on top for defenders with Andy Robertson behind him. Aymeric Laporte could be one to watch at 6.0m this season as he returns to the Manchester City first-team properly. For consistency of starts and goal/assist threat, Matt Doherty is your man in the 6m bracket.

When looking at the 5.5m options Serge Aurier stands out as a potential candidate but for consistency Patrick Van-Aanholt is knocking on that 4 points per 90 minutes mark (he is injured for now however so probably won’t play Gameweek 1).

Over in the 5.0m’s Ben Davies looks tasty if he can become a regular starter again, as does Hector Bellerin. The ‘safer’ assets would be John Egan, Roman Saiss and (surprisingly) Michael Keane.

It is in the 4.5m category where it becomes interesting with Kyle Walker-Peters, Tariq Lamptey and Rubens Vinagre looking good. We must take these numbers with a pinch of salt however as they have played so few minutes so far. It will be interesting to see if they can keep it up with regular game time. Jamal Lascelles is the stand-out option here, with 3.3 points per 90 minutes over 3 seasons of football at Newcastle, although he scored 3 goals in 2017/2018 which perhaps skews his data a little.

Midfielders

12.0m

Salah – 7.92 (3 seasons)

Aubameyang – 7.07 (2 seasons, points adjusted for if he was a MID)

Mane – 6.73 (3 seasons)

Unsurprisingly the Egyptian King Mohamed Salah tops the list with a HUGE 7.92 points per 90 mins. He really is a beast. Pierre-Emerick Aubameyang looks like a great option now he is a midfielder. Sadio Mane lags slightly behind despite costing the same.

11.5m

Sterling – 7.5 (3 seasons)

De Bruyne – 6.84 (3 seasons, injured for most of 18/19)

Raheem Sterling is close to Salah in terms of points per 90 with a massive 7.5. Being 0.5m cheaper than Aubameyang makes him look very exciting indeed. You could maybe argue he was a little unlucky last season so this one could be a big score. Kevin De Bruyne comes in 4th place amongst these top priced midfielders. He had his best ever season in 19/20 – can we expect him to repeat it this time around or regress slightly?

10.5m

Fernandes – 8.87 (13 matches)

Bruno Fernandes is an absolute MACHINE with that score. It’s the small amount of data we have to be conscious of but even if he regresses a bit he will still be well up there in terms of points.

9.5m

Rashford – 6.22 (3 seasons but limited minutes in 17/18, points adjusted for if he was a midfielder)

9.0m

Son – 6.9 (4 seasons)

8.5m

Mahrez – 6.84 (3 seasons, limited minutes in last 2)

Pulisic – 6.66 (limited minutes)

8.0m

Willian – 5.58 (3 seasons)

Heung-Min Son is the interesting option amongst the 8.0-9.5 midfielders because of his consistency. Four seasons of scoring 6.9 points per 90 is fantastic and just what I am looking for. All the hype around Christian Pulisic is particularly interesting to note when Son actually beats him, despite playing a full 3.5 seasons more.

7.5m

Greenwood – 8.22 (around 15 matches, points adjusted for if he was a midfielder)

Bernardo Silva – 5.39 (3 seasons, limited minutes in 19/20 and 17/18)

Mason Greenwood is up there with Fernandes in terms of his points per 90 minutes – what a debut season for the youngster! I’ll be keeping a close eye on the transfer market to see if United sign someone for his position. If they don’t, he could be gold at that price. Limited data is of course the caveat with Greenwood!

7.0m

Barnes – 5.15 (2 seasons, limited minutes)

Lucas Moura – 4.78 (2 seasons, limited minutes)

Sigurdsson – 4.43 (4 seasons – huge scores in 16/17 & 18/19, very low in 17/18 & 19/20 when minutes were limited)

Maddison – 4.32 (2 seasons)

Mount – 4.3 (1 season)

Zaha – 4.15 (4 seasons)

Harvey Barnes is an exciting prospect with a score of 5.15 per 90 but needs to nail down a place in the team. If you want reliability and consistency, Wilfried Zaha is the man to go for with 4 seasons of decent scoring under his belt.

6.5m

Foden – 6.96 (2 seasons, very limited minutes)

Lallana – 5.34 (4 seasons, very few minutes in last 3 but expected to be a starter for Brighton). Points per 90 from 13/14-16/17 = 5.02

Jota – 5.07 (2 seasons, limited minutes, points adjusted for if he was a MID)

Perez – 4.79 (3 seasons, limited minutes in 19/20)

Traore – 4.5 (2 seasons, very few minutes in 18/19)

6.0m

Yarmolenko – 6.07 (2 seasons, very limited minutes)

Trossard – 4.67 (1 season, limited minutes)

Milivojevic – 3.9 (3 seasons, big drop off in 19/20)

Ward-Prowse – 3.76 (3 seasons, limited minutes in 17/18 & 18/19, played every game in 19/20)

McNeil – 3.62 (2 seasons)

Look at those numbers for Phil Foden and Andriy Yarmolenko! Of course, the reliability of minutes just isn’t there but well worth keeping an eye on, particularly Foden now David Silva has left City. Adam Lallana is also one to watch at Brighton and possibly Diogo Jota now he is a midfielder again. Not many others to be excited about I’m afraid!

5.5m

Neto – 5.58 (around 10 matches)

Armstrong – 4.98 (2 seasons, limited minutes)

Trezeguet – 4.56 (around 21 matches)

St. Maximin – 4.45 (around 21 matches)

Henderson – 4.1 (2 seasons, limited minutes in 18/19)

Shelvey – 3.67 (3 seasons, very few minutes in 18/19)

Saka – 3.66 (1 season, limited minutes)

Westwood – 3.45 (2 seasons)

McArthur – 3.3 (3 seasons)

5.0m

Ritchie – 3.57 (3 seasons, 16/17-18/19)

Soucek – 3.55 (around 12 matches)

Noble – 3.52 (3 seasons)

The early love on Twitter for Stuart Armstrong appears to be justified as he has 4.98 points per 90 minutes. Providing he is more of a regular starter this season for Saints, he could be a great bargain. Pedro Neto worth noting with a big 5.58 but his minutes just aren’t guaranteed. Trezeguet for Aston Villa will surprise some I’m sure but I can’t see him nailing down a starting spot. Another Twitter favourite Tomas Soucek actually only beats West Ham stalwart Mark Noble by 0.03 points per 90 minutes. Noble’s numbers are over a full 3 seasons so wins this category for value and reliability.

4.5m

Gordon – 3.51 (around 5 matches)

Stephens – 2.35 (3 seasons)

Oriol Romeu – 2.28 (3 seasons, limited minutes last 2)

There really isn’t anything in the 4.5m midfielder category. If you are going for a four man midfield with a 4.5 bench warmer, then Dale Stephens at Brighton is the most reliable.

Forwards

10.5m

Aguero – 7.36 (6 seasons, limited minutes in 19/20 due to injury)

Kane – 6.31 (6 seasons)

Right up there with the big boys is of course Sergio Aguero – a whopping score of 7.36 per 90 minutes. It remains to be seen how his minutes will be managed this time around but well worth a punt at his new price of 10.5. Harry Kane also worth monitoring to see if he finds top form again.

10.0m

Vardy – 5.66 (5 seasons)

9.5

Jesus – 6.73 (3 seasons, limited minutes)

Firmino – 5.35 (3 seasons)

9.0m

Martial – 5.79 (1 season, numbers adjusted for if a forward)

8.5

Ings – 5.83 (2 seasons, limited minutes in 18/19)

Jimenez – 5.31 (2 seasons)

Not that much to choose between the 8.5m-10.0m options here. Consistency and reliability wise look no further than Jamie Vardy who has done it for 5 seasons in FPL. Gabriel Jesus would be gold if he was regular starter, outscoring Kane (albeit with much much less minutes). Danny Ings looks like the best choice but we only have one full season of data unfortunately. Anthony Martial will need to better his goals total this season now he is a forward to make him worth the money.

8.0m

Richarlison – 4.19 (1 season, numbers adjusted for if a forward)

7.5m

Abraham – 6.24 (1 season, around 25 matches)

7.0m

Giroud – 6.42 (6 seasons, limited minutes in all but 15/16)

Calvert-Lewin – 4.4 (2 seasons)

When you look at Olivier Giroud’s numbers it is really surprising that he has been a back-up Striker for so long. When he plays, he scores points. Good points! Tammy Abraham also scored well but with Timo Werner now at Chelsea I can’t help but feel he won’t get the minutes required.

6.5m

Wood – 5.03 (3 seasons, limited minutes in 17/18)

Antonio – 4.99 (1 season, adjusted for if a FWD)

Maupay – 4.27 (1 season)

6.0m

Adams – 6.41 (around 13 matches)

Ashley Barnes – 4.21 (3 seasons, limited minutes in 19/20)

Mitrovic – 3.95 (15/16 & 18/19)

Ayew – 3.77 (1 season)

5.5m

Connolly – 4.39 (around 14 matches)

4.5m

Davis – 4.22 (around 6 matches)

Slim pickings from the rest of the forwards. Chris Wood looks like good value at 6.5m, with Che Adams an exciting punt scoring 6.41 per 90 so far. As we’ve seen with most of the other options though, the more games they play the more their score *should* level out. Pre-season favourite Aleksandar Mitrovic with a rather unappealing 3.95 points per 90. As is always the case, there are many more defensive options at 4.5m than there are strikers.

My ‘set and forget’ Team:

Phew! Still with me?! Now I can work out a team. Remember, this first team I pick let’s imagine that I don’t plan on making any changes throughout the season. I’m certainly not against transfers (and I’ll be making lots!) but hopefully this will give an idea of who, according to the system, are the best options for the 20/21 FPL season. I will be picking my team for Gameweek 1 using this along with making my transfers.

What about the captaincy?! It is perhaps the biggest variable in FPL – you have to get it right more often than not to do well. For the purposes of this, I’ll double the score of my highest, most reliable pick.

Target points – 67.76

GK – Ryan £4.5m (3.5 pp90)

DEF – Alexander-Arnold £7.5m (6.31)
DEF – Lascelles £4.5m (3.3)
DEF – Vinagre £4.5m (3.74)

MID – Salah (C) £12.0m (7.92 x 2 = 15.84)
MID – Sterling £11.5m (7.5)
MID – Fernandes £10.5m (8.87)
MID – Greenwood £7.5m (8.22)
MID – Armstrong £5.5m (4.98)

FWD – Ings £8.5m (5.83)
FWD – Wood £6.5m (5.03)

Subs – Button £4.0m, Walker-Peters £4.5m, Ferguson £4.0m, Davis £4.5m

Total points per 90 mins = 73.12

A very nice score, 5.36 points more than what I was aiming for. Of course there are lots of caveats here: Vinagre, Fernandes, Greenwood, Armstrong and Ings haven’t played enough minutes to be deemed truly ‘safe’ picks. However, you get the idea. Let’s look again and make another team with only one ‘risky’ pick…

GK – Ryan £4.5m (3.5)

DEF – Alexander-Arnold £7.5m (6.31)
DEF – Egan £5.0m (3.76)
DEF – Lascelles £4.5m (3.3)

MID – Salah (C) £12.0m (7.92 x 2 = 15.84)
MID – Sterling £11.5m (7.5)
MID – Son £9.0m (6.9)
MID – Greenwood £7.5m (8.22)
MID – Noble £5.0m (3.52)

FWD – Kane £10.5m (6.31)
FWD – Wood £6.5m (5.03)

Subs – Button £4.0m, Ferguson £4.0m, Douglas £4.0m, Davis £4.5m

Total points per 90 mins = 70.19

Perhaps not as exciting on paper but Greenwood is the only real ‘risk’ in terms of my system. The rest of the team have played consistently, for more than one season (apart from Egan) in the Premier League. The probability of this team scoring those points per 90 is greater than the first team as we have more data to work from.

I hope crunching all these numbers helps you pick your team for this season. As I keep saying, this isn’t an exact science (far from it!) but as scoring points is the most important part of FPL, I thought why not see who the most reliable points scorers are.

Good luck everyone! I look forward to engaging with you all this season.

25% Off Hub Membership

Member Benefits

New app

Become a member!

Join our FPL WhatsApp groups

Join the Hub and chat all things FPL in one of our popular WhatsApp groups:

WhatsApp Chat Groups for FPL, Sky, Telegraph, SDT & FanTeam

Follow the Hub

Twitter | Facebook | YouTube | Instagram | Tiktok
And view all our latest fantasy football tips on our Homepage

Goalkeepers (highest scorer for each price point plus notable others)

Defenders

Midfielders

Forwards

My ‘set and forget’ Team:

More FPL Content

25% Off Hub Membership

Member Benefits

New app

Join our FPL WhatsApp groups

Follow the Hub

Picking My FPL Team – The Maths Approach

Win at FPL

Community & tools