Hunting for Value in FPL - Overperformers and Underperformers

In this piece, I will be looking at the biggest overperformers and underperformers in Fantasy Premier League (FPL) last year to try and identify some overlooked and overhyped picks ahead of the start of the 26/27 season.

Two quick things before we get into it: Firstly, I have not looked at anyone who had fewer than 1500 mins – that’s because generally these guys won’t be FPL picks as they simply do not play enough.

Secondly, expected Assists (xA) is based on whether a player’s action will lead to what Opta, and basically anyone else, defines as an assist. That is different from what FPL deem to be an assist. For example, winning a penalty is an assist in FPL if it is scored but Opta does not consider that. To keep this simple, I am using Opta assists data and not Official FPL. That will mean that the players below will often have more assists in the FPL game, we just cannot model them easily.

The Overperformers

Name	Team	Cost	App	Mins	Assists	xA	Difference	Goals	xG	Difference	G+A	xGI	Difference
Bowen	WHU	N/A	38	3402	11	5.24	5.76	9	7.26	1.74	20	12.5	7.5
B. Fernandes	MUN	12	35	3062	21	12.32	8.68	9	10.83	-1.83	30	23.15	6.85
Semenyo	MCI	8.5	37	3200	4	3.12	0.88	17	11.09	5.91	21	14.21	6.79
Haaland	MCI	15.5	35	2955	8	2.85	5.15	27	25.43	1.57	35	28.28	6.72
Wilson	FUL	6.5	36	2692	7	4.78	2.22	10	5.83	4.17	17	10.61	6.39
Richarlison	TOT	6	32	2029	4	1.46	2.54	11	8.09	2.91	15	9.55	5.45
Gibbs-White	NFO	8	37	3090	4	2.92	1.08	15	10.84	4.16	19	13.76	5.24
Rogers	AVL	7.5	37	3279	6	4.3	1.7	10	6.78	3.22	16	11.08	4.92
Gravenberch	LIV	6	36	2997	3	2.26	0.74	5	1.9	3.1	8	4.16	3.84
Dorgu	MUN	6	26	1509	4	1.85	2.15	4	2.35	1.65	8	4.2	3.8
Bruno G	NEW	7	29	2459	5	4.98	0.02	9	5.6	3.4	14	10.58	3.42
Matheus Nunes	MCI	6	34	2866	5	2.24	2.76	1	0.35	0.65	6	2.59	3.41
Cherki	MCI	7.5	33	1866	12	8.65	3.35	4	4.13	-0.13	16	12.78	3.22
Kroupi Jr.	BOU	7.5	33	1742	0	1.53	-1.53	13	8.35	4.65	13	9.88	3.12
João Pedro	CHE	7.5	35	2655	5	2	3	15	14.96	0.04	20	16.96	3.04
Woltemade	NEW	6	33	1931	3	1.53	1.47	8	6.59	1.41	11	8.12	2.88
Kudus	TOT	6.5	19	1529	5	2.4	2.6	2	1.93	0.07	7	4.33	2.67

I’ve left Bowen in this list simply so we (and mainly Joe) can lament his loss to the league. Let’s look at the rest of the list then:

The Big Three

Bruno Fernandes – He literally broke the record for assists last year and you can see that is the principal source of his overperformance. I don’t think that necessarily makes him a bad pick – he has great opening fixtures and is a fitness monster. However, he also missed a couple of penalties last year which largely makes up for his underperformance on xG. I don’t think he will get 235 points again this season. It is more likely that he will finish 190- 210 points which is more in line with his historical performance. So, he is still a great pick.

Semenyo – Historically, Semenyo has outperformed his xG. That is normally a sign that a player is an above average finisher. However, the fact that he overperformed his expected goals by such a huge margin last year gives me pause for thought. The best finisher on this list comes up next and he only outperformed his xG by 6% and Semenyo outperformed his xG by about 53%. Long time readers will know that I LOVE Semenyo, but this combined with a new manager means he probably doesn’t make my gameweek 1 team.

Haaland – I don’t care about his overperformance on xG because he is the best striker in the league and one of the best finishers in the world. The overperformance on xA is more concerning but then he also plays for one of the best teams in the world and is surrounded by footballing greats and he has done this in the past too. I think it seems fair to suggest his assists go down next season but they are the cherry on top of this great FPL pick and not the reason why you own him.

The Rest

Morgan Rogers has outperformed his underlying stats basically every year he has played in the PL. Outsmarting underlying stats is basically why Unai Emery is an amazing manager – I don’t know whether Xabi Alonso is that guy. Overall, I think the player can continue to outperform his stats but I am just a little wary of his numbers. Of the 150 best players last season, he was the joint 14th worst in terms of blanks with 22 blanks in 37 starts. His blanking numbers are similar to that of a goalkeeper, for example Pickford blanked 24 times. He tends to get attacking returns against weaker opposition which is why, despite his high upside, he is a fixture driven pick for me.

I think there are legitimate reasons to be concerned about whether Harry Wilson can match his overperformance from last season – 10 goals off 5.88 xG is big and we don’t really have historical evidence of him overperforming to this extent. However, at 6.5 he is still pretty cheap and could therefore be a good pick.

Richarlison will likely remain a good pick if he stays and Solanke is injured and Spurs don’t sign another striker. But, he isn’t a guy you will hold all season long. Dorgu is similar – he isn’t nailed so I can’t see him as a consistent option for us.

Gibbs-White on the other hand is completely nailed. He has been handed a decent price hike and we know he is a bit streaky – he went 8 games at the start of last season with just one assist and no goals. His overperformance on xG is kind of the same as Wilson’s and even though he is the main man, I am worried about the price tag. This doesn’t mean I won’t ever own him but it is a question in my head.

Gravenberch and Bruno G. can be dealt with together in my opinion. They are essentially defensive midfielders who benefited from a little variance (and some pens for Bruno G) last year. I don’t see that happening again and will be avoiding them. Probably the same for Matheus Nunes though I am more inclined to have him because of his likelihood to get clean sheet points if he is nailed.

Both Kroupi and Cherki did what they did in basically half a season’s worth of minutes which will make the numbers less trustworthy. The former is injured but you can see why people were excited about him if he were nailed. And, the same is true for the latter – there is a little overperformance but also A LOT of minutes to gain if he becomes nailed which makes him good value IF HE STARTS. Similarly, Woltemade only had half a season’s minutes but who knows who starts for Newcastle up top.

I have already set my stall on Joao Pedro but basically I like the pick in good fixture runs, I am just not sure playing Brighton and Arsenal in the first three is that.

Finally, Kudus is a dribbly boy who has never actually had great attacking output and comes into the season injured. He’s not someone I am looking at who can step up.

The Underperformers

Name	Team	Cost	App	Mins	Assists	xA	Difference	Goals	xG	Difference	G+A	xGI	Difference
Yeremy	CRY	5.5	34	2121	2	6.11	-4.11	2	4.51	-2.51	4	10.62	-6.62
Amad Diallo	MUN	6	32	2363	3	4.78	-1.78	2	5.57	-3.57	5	10.35	-5.35
Evanilson	BOU	6	36	2744	1	1.49	-0.49	6	10.59	-4.59	7	12.08	-5.08
Kamada	CRY	5	28	1918	0	3.24	-3.24	0	1.65	-1.65	0	4.89	-4.89
Enzo	CHE	7	36	3115	4	7.25	-3.25	10	11.29	-1.29	14	18.54	-4.54
Kadioglu	BHA	4.5	37	3149	0	3.33	-3.33	1	2.03	-1.03	1	5.36	-4.36
Jensen	BRE	5.5	36	2296	1	6.02	-5.02	3	2.27	0.73	4	8.29	-4.29
Wirtz	LIV	7.5	33	2393	3	5.17	-2.17	5	7.02	-2.02	8	12.19	-4.19
Gudmundsson	LEE	4.5	32	2625	0	2.75	-2.75	0	1.1	-1.1	0	3.85	-3.85
Mateta	CRY	6.5	32	2244	0	1.31	-1.31	12	14.51	-2.51	12	15.82	-3.82
Schade	BRE	6	35	2748	3	2.08	0.92	8	12.18	-4.18	11	14.26	-3.26
Mbeumo	MUN	8	33	2613	3	5.04	-2.04	11	12.19	-1.19	14	17.23	-3.23
Hughes	CRY	4.5	31	1641	0	1.89	-1.89	0	1.09	-1.09	0	2.98	-2.98
Mitoma	BHA	6	25	1746	1	2.84	-1.84	3	4.06	-1.06	4	6.9	-2.9
Wharton	CRY	5.5	34	2552	5	7.38	-2.38	1	1.52	-0.52	6	8.9	-2.9
Castagne	FUL	4.5	30	1887	0	1.46	-1.46	0	1.42	-1.42	0	2.88	-2.88
Smith-Rowe	FUL	5.5	37	1992	0	1.11	-1.11	3	4.77	-1.77	3	5.88	-2.88
Struijk	LEE	5	34	2936	1	0.78	0.22	0	3.02	-3.02	1	3.8	-2.8
Scott	BOU	6	37	2861	1	2.95	-1.95	3	3.81	-0.81	4	6.76	-2.76
Ampadu	LEE	5.5	35	3078	1	2.25	-1.25	1	2.5	-1.5	2	4.75	-2.75
Richards	CRY	5	33	2839	0	1.38	-1.38	1	2.36	-1.36	1	3.74	-2.74
Rodri	MCI	6.5	21	1526	0	2.59	-2.59	1	1.13	-0.13	1	3.72	-2.72
Maatsen	AVL	4.5	30	1654	1	2.41	-1.41	0	1.3	-1.3	1	3.71	-2.71
Saka	ARS	9.5	31	2240	5	7.16	-2.16	7	7.55	-0.55	12	14.71	-2.71

There are some really intriguing names on this list. Five players on this list play for Palace and that probably shows how much they struggled to maintain fitness throughout the season with such poor squad depth. That will probably happen again I guess but it’s interesting that there is a lot of pre-season chat about Sarr but nothing about Yeremy. If he can nail down a spot, he offers similar xGI numbers for a million cheaper. Another reason why I like Yeremy is that he is coming into his second PL season now. We know that players tend to struggle in their first PL season because the league is just so good and they get used to it in the second season, so why not this guy? There’s enough question marks for me to not start the season with him but I would be tempted in a deep draft league and will keep an eye on him outside of that.

Second season syndrome is a principal reason I have been toying with Wirtz in my drafts too. We can see that he has more to give in terms of expected numbers and he seems to be nailed on number 10 for Iraola’s Liverpool.

Evanilson has not come anywhere near my thinking because he has now underperformed xG two seasons in a row – I think he might genuinely be just a below average finisher. I’ll still probably fall into that trap again this season…

Much like I will fall for my love – Schade. I really thought he was gonna have a breakout season last year after watching him play in 24/25 and admiring his stats. Last season was super hit and miss – he blanked LOADS and then put up a hat-trick. As you can see, he substantially underperformed his xG so if he can bring the actual number up he could offer great value at 6.0.

Ampadu is another guy I have had my eye on. He’s a DefCon merchant really but if he ups his attacking output to match his expected output he would have finished last season around 145 points which is incredible value for 5.5. Of course, that also means he needs to get a similar number of FPL assists to last year – 3. But, at that price, he’s a guy I am willing to make my 7th or 8th attacker/bench player.

Maatsen has emerged as a darling for the analytics models because Digne has left for PSG and his expected minutes are now really high. I don’t think we can trust those expected minutes because Emery likes to rotate and they are in the market for someone. But, I think he could be okay to start the season and I really like his attacking threat.

Finally, Saka finished last season as one of the top points scorers per start. He didn’t start enough which was problematic but he also underperformed his stats. If we can own him when he starts he is great.