Now we've squeaked past the halfway point in the Premier League, it's as good as time as any to look back at how players have performed over the course of the season and see if any trends have emerged. We'll split the season so far into two 10-gameweek sections to remove any bias for overall numbers and look at players' points per start to see if who is on the rise or declining, why, and, most importantly, what to do about them!
To keep the graphs relevant, it has been filtered for players that have started at least 3 games in each period (GW1-10 and GW11-20). The graphs will be set out in quadrants that split below and above average points per start (PPS) for each period, while the diagonal line will split players that have been improving (i.e. their PPS in GW11-20 is better than GW1-10), or declining. PPS is being used as this is the best metric to remove any noise from sub-appearances where you will likely never have started the player anyway.
The players in the bottom left quadrant are typically irrelevant or streamer level as their PPS is consistently below average. These are generally to be ignored unless the underlying numbers or previous season's performances provide some hope that things could get better. Players in the top right quadrant are consistently above average and should be rostered everywhere. The players in the top left quadrant got off to a slow start but have improved significantly, while those in the bottom right quadrant got off to a hot start but have been falling well below that ever since. For each position, we will look at some overall trends for players, while also diving a bit deeper into some interesting cases for both the risers and fallers.
Defenders
One of the first things to note is the very low level of PPS for defenders. Players that fit the criteria above average just 5.9 PPS. There are some players in the bottom left quadrant that you may even have punted on in a favourable matchup, such as Issa Kabore (4.0 PPS), and some bigger names that have likely been rostered all season such as William Saliba (5.3 PPS). Unfortunately, this is more of an indication of how bad defenders have been this season, where almost half of relevant starting defenders are consistently not worth rostering. If you read this deep dive article last season on the decline in defenders performance you will understand why this trend is continuing.
There really isn't much to shout about for defenders aside from a couple of interesting outlier cases. Antonee Robinson has impressively increased his ghost point output from 8.8 to 11.8 gPPS and started to provide more creativity with 3 assists. Marcos Senesi was in and out of the team earlier in the season, but he has certainly nailed down his spot now, in GW11-20 he recorded 12.9 PPS! Scoring 2 goals from just 0.7xG and an assist from 0.4xG has helped, but Bournemouth's improved defensive performances with 4 clean sheets is giving him a significant boost too. On the flipside a player that has gone off the boil is Ethan Pinnock, but this has more to do with Brentford's decline than his own. Pinnock's ghost point output is still one of the best in the league; it has only reduced slightly from 10.6 to 9.8 gPPS. More significantly, Brentford have just stopped keeping clean sheets and their goals against have increased from 1.2 to 2.1 per game. Even a legendary ghoster like Pinnock needs his team to be functioning to offer good value.
Risers
Trent Alexander-Arnold
It was a very underwhelming start to the season for Alexander-Arnold, with those of us who drafted him in round 1 wondering whether we had completely sabotaged our season with this pick. Short of the home game against Aston Villa where he put on a deep-lying passing master class, Trent really struggled to find a productive role for Liverpool as Dominic Szoboszlai dominated creatively from midfield. This meant less time in the inverted midfield role that was so productive at the end of last season. Szobo even took over a significant number of set pieces, further stifling Trent's production. From GW1-10, Trent scored no goals, provided just 1 assist and his bread and butter key passes were down to just 1.7 per 90, resulting in a meagre 8.6 PPS.
However, since then Trent has been on fire! As Szoboszlai's influence and energy has waned, Klopp has trusted Trent to step into midfield more consistently and boy has that helped! In GW11-20 he has racked up the most FPts by a significant margin at 192 with a whopping 18.5 PPS. Two goals and 3 assists have helped, but his creativity has skyrocketed, with key passes jumping to 3.8 per 90, boosting his ghost points from 7.4 to 14.1 gPPS! This is also borne out in the underlying data with his xA increasing from 0.14 to 0.4 per 90. Now, will this continue? It's difficult to say as Klopp has seen good results by shifting his tactics and squad around already this season. So while I don't expect Trent to continue to juggernaut his way to over 18 PPS, I'm confident that his performances and growing influence on this Liverpool team will result in over 15 PPS for the rest of the season.
Unfortunately, we recently received the news that Trent has been injured and will be out for a 'few weeks', which won't help his momentum. But given how spread out the next couple of gameweeks are, he may only miss a couple of matches. So if your team isn't too badly hit by AFCON/Asian Cup/injuries, I would be tempted to buy low on this news if possible.
Emerson
We have seen a big improvement from Emerson having posted just 4.3 PPS in GW1-10 but now has a respectable 8.0 PPS in GW11-20. You might think that he could be a good option going forward, especially given his recent performances. However, we need to look a bit deeper into what's causing this, as his ghost point numbers have stayed pretty similar throughout at 6.6 gPPS in GW1-10 vs 7.3 gPPS in GW11-20 and his two assists for the season came in GW2 & GW8. So what's changed? After keeping just one clean sheet all season up to GW16, West Ham managed to rack up 4 clean sheets in 4 gameweeks, despite giving up a total of 6.7 xGA. Those 4 clean sheets are massively propping up Emerson's total and they were fortunate to nick them given the aforementioned xGA. So I'm much more inclined to trust his season long 6.2 PPS as a good indicator of ROS value, which puts him at streamer level.
Fallers
Joachim Andersen
Andersen was the 2nd highest scoring defender and the 14th highest scorer overall in GW1-10, putting up a ridiculous 12.8 PPS - reaching heady heights indeed given his obscurity on draft day. We all knew this level wasn't sustainable, but how has he fallen so far and so fast? Frankly, he hasn't been worth rostering through GW11-20 with just 4.5 PPS. His ghost points have dropped significantly from 10.5 to 7.7 gPPS, largely from a reduction in key passes and aerials. You can also see this reduction in creativity in his shot creating actions (SCA) numbers too, which are down from 2.1 to 0.7 per 90. In the early part of the season he managed to score 2 wonder goals which are not exactly repeatable statistically. Even his assist vs Spurs in GW10 was a lofted ball into the box that somehow fell kindly for Ayew. Palace have also been getting steadily worse this season as a team, which has resulted in more losses and more goals against (13 vs 16), and while the underlying xGA numbers isn't much worse (15 vs 16.4), the lack of clean sheets has been a killer (4 vs 1) for the big Dane.
So it's been a perfect storm for Andersen - he's being stifled in how expansive he can be with his range of passing and getting forward while letting in goals every game, which is not a recipe for success for a defender in Fantrax! So it appears Andersen's value is intrinsically linked to Palace's fortunes improving, and given the dearth of options in defence, I wouldn't blame anyone for holding onto him in the hope that he improves. I wouldn't expect the lofty heights of 12.8 PPS to return, but around 7ish PPS could be possible going forward if Palace pick up again. If you can sell at a value closer to his season long 8.7 PPS, I would go for it.
Kieran Trippier
It feels like those who had doubts about Trippier in pre-season are maybe starting to feel vindicated given his (and Newcastle's) bad form and his increasing susceptibility to injury. Those worries included fatigue from additional European fixtures, more potential for rotation and the cost this could have on Newcastle's performances and results in the Premier League. Clearly, no one could have foreseen the injury crisis that has occurred, but it could certainly be argued that Howe did not help matters by not trusting his squad players and trotting out his best available XI every game.
Still, Trippier didn't have the best start to the season due to a tough run of fixtures, which meant just 7.8 PPS in the first 4. But he then went nuts with 25.8 PPS in the next 5. An overall PPS of 16.5 in GW1-10, was followed by a tough period in GW11-20 where he scored 'just' 10.0 PPS and missed 2 games through injury. However, this average was propped up massively by a 34.5 point outing against United, without that it was just 6.5 PPS! Now his ghosts have declined a bit, from 12.5 to 11.6 gPPS, but this isn't too significant as this still puts him as one of the best ghosters in the league.
One of Trippier's strengths is his open play and set pieces deliveries, which of course gives him an easy route to key passes but also opportunities to provide assists. In GW1-10 he provided 6 assists (3.9 xA), but this reduced to just 1 (2.4 xA) in GW11-20, so a reduction in his chance creation but maybe a bit of reverting to the mean in terms of his teammates' finishing. However, probably the most significant portion of this can be attributed to Newcastle shocking decline in their underlying stats. It's no wonder their defensive assets have struggled lately. In GW1-10 they conceded just 11 goals (9.8 xGA), which ballooned in GW11-20 up to 18 GA (24.1 xGA). Fortunately for Trippier he wasn't present during the 4-2 mauling at Anfield, but he also missed the 3-0 win over Fulham, which means that 4 clean sheets in GW1-10 was reduced to just 2 in GW11-20. Goals against also increased from just 1.1 to 1.75 per game.
While it's not the prettiest of pictures for Trippier, there is some light at the end of the tunnel for those rostering him. He's now fit again, as we saw with him playing 90mins in the FA Cup against Sunderland. The extended winter break affords Newcastle the opportunity to rest their overworked players and get their injured players back into the fold. Additionally, they are out of the Champions League (and Europe altogether), removing the additional games they had to contend with earlier in the season. If anything it could be a great time to buy low on Trippier given his perceived drop in value. If Newcastle can get back to firing again we could see him comfortably back around the 14+ PPS mark. It's also worth noting that their fixture run looks very favourable, after they get City out of the way in GW21, so maybe wait until after that matchup to pounce.
Midfielders
It's interesting to note how much better the average PPS is for midfielders than defenders at 8.0 PPS, and while streamers in and around this value typically don't have a great ceiling, they certainly have a better floor than the average defensive streamer and are typically a much safer bet. There are a few unheralded names like Brownhill, Lemina, Cook and Christie, that have come into relevance, while Luton boys Barkley and Doughty have been tearing it up and just seem to be getting better. We've also seen some big names fall away such as Mitoma, Jensen, Doucoure, Ward-Prowse and even Bruno Fernandes who has struggled with United's lethargy. Given the lack of quality midfielders, at times defensive midfielders like Rice, Palhinha and Norgaard have been useful MID4/5s but their latest form has been pretty disappointing and will make you think twice about the value of holding them.
Martinelli has improved ever so slightly, which shows just how bad he's been this season. His ghosts haven't changed but he's just not generating as much xG+xA as last season (down from 0.59 to 0.34 per 90) and he's slightly underperforming those numbers too. Currently a super buy low, I'm not sure I necessarily want to go there especially if people are holding too close to draft day value. It's also worth noting how good Gross has been: 6th overall in FPts, an amazing ghoster and has only got better despite Brighton's struggles and being played all over the pitch.
Risers
Cole Palmer
While Palmer took a while to break into the Chelsea XI, his talent couldn't be held back for too long and he has been a mainstay ever since. His early performances were aided substantially by scoring 4 penalties in just 5 games. At the time, his ghost point numbers had been sketchy, at just 6.8 gPPS in his first 4 GWs (up to GW10). However, they have significantly improved since then, increasing to 8.7 gPPS, which added to his goal and assist threat. As a result, he has grown into an elite fantasy asset at 17.1 PPS in GW11-20.
Interestingly, his underlying data hasn't changed too much over the GW11-20 period. He has just been taking more of his open play chances (4 goals from 2.7 npxG) and his teammates have been putting away the chances he has created (3 assists from 3.3 xA). While it would be folly to expect Chelsea to keep getting as many pens as they have so far this season (7) to keep up Palmer's current output, his increasing productivity in open play and Chelsea's reliance on him as one of their only producing attackers is a huge boost for his prospects. I don't expect him to continue at this rate given his overperformance on his xG and an expected drop off in pens, but he should be good for at least 13+ PPS rest of season.
Marcus Tavernier
On his day Tavernier is one of the most explosive players in the Premier League! However, he started the season injured and had his minutes managed on his return in the early part of the season, all while Bournemouth struggled for form and results. But it has all started coming together recently, with Tavernier increasing his PPS from 7.8 to an eye-popping 13.7. Tavernier's ghost point production per 90mins has stayed the same (9.3 gPP90), but he's now getting more minutes, going up from 72mins to 85mins, which gives him a boost of 1.3 ghost points per start. He's now getting attacking returns to compliment his ghosting ability with 2 goals (2.2 xG) and 3 assists (1.9 xA) in GW11-20. The increase in his underlying numbers is quite something, with xG/90 up from 0.06 to 0.42, xA/90 up from 0.15 to 0.36, and SCA/90 up from 1.68 to 7.25. The Bournemouth attack has been on fire and Tavernier has been a big part of that. While I think he can keep up this form over the second half of the season given his numbers, he doesn't have the best injury record and Bournemouth have a very rough set of fixtures coming up, so I'd consider trading out on a high depending on what you could get for him.
Fallers
Bukayo Saka
It was interesting to see Saka drop off quite significantly from 18.5 to 13.8 PPS during the period in question. Even at 13.8 PPS he's still a great asset, but he has seemingly lost a step in his points production, so how can we account for this? His ghosts have declined a bit from 11.0 to 10.2 gPPS but that's relatively minor. The underlying stats aren't much different either, his npxG/90 (0.29 to 0.27) and xA/90 (0.26 to 0.33) and his SCA/90 (5.1 to 6.3) are remarkably similar. Simply put, Saka had a few things go for him in the early part of the season that just haven't gone his way lately.
First, he scored 2 penalties in GW1-10 and somehow managed to give away or avoid the other 4 penalties that Arsenal have been awarded. He also managed to record 2 'fantasy assists' - actions that aren't particularly sustainable - that bolstered his tally to 5 in GW1-10 compared to 3 for GW11-20. So as we can see a bit of variance significantly helped him in the early part of the season. So while I don't expect him to continue to be as low as 13.8 PPS, the odd penalty or uptick in his ghosting should see him more aligned with his season long 16 PPS, which seems a fairer reflection of what he should produce going forward.
Dominik Szoboszlai
Szoboszlai's decline in form has been quite shocking, dropping from 13.1 PPS to 6.7 PPS, the largest drop off we've seen of all midfielders! One of the main factors is his decline in creativity and attacking; the key passes and assists have just dried up. His xA/90 has fallen from 0.35 to 0.08 and his SCA/90 has declined from 5.1 to 2.87 per 90. Now this could partly be down to Trent's growing influence as he has moved into midfield more often, but more likely this can be attributed to fatigue. He has gone from playing 90 mins every game to averaging just 73 mins as Klopp has tried to manage his workload. Yet he still ultimately succumbed to a hamstring injury against Newcastle last time out.
Even if you look at his ghost point numbers on a per 90mins basis, these are significantly down, from 9.3 to 6.2 gPP90, pretty much every category is down and he is getting fewer minutes, a double whammy for his value. On the plus side, Trent's injury leaves a window of opportunity for Szoboszlai to get back to his best. He will have had a period of rest and recovery and it appears he has a chance to be back for the next Premier League match vs Bournemouth given it is more than a week away still. It all depends on what value you can get him for but he could be an interesting buy low trade target.
Forwards
Now we move onto forwards, where we see a big advantage over defenders and midfielders, with the average forward scoring 9.5 PPS. Due to the limited number of forwards, a few players below average can be considered as long term holds. We have seen the likes of Wissa, Edouard, Morris, Ferguson and McBurnie provide good value at the beginning of the season, but you'll likely see all of them on the waiver wire at this point. It has been amazing to see the difference that Gary O'Neil has made to the Wolves attack, with Hwang and Cunha putting up ridiculous numbers. But as you can see from the graph, Cunha is the one who just keeps pushing on, recording a ridiculous 12.9 PPS in GW11-20. He now ghosts at a good rate (7.4 gPPS) and is taking the chances he's getting with 6 goals from 6.0 xG, an absolute steal off the waiver wire for anyone lucky enough to grab him this season!
Somehow Rashford and Jesus are improving despite being particularly underwhelming given their high ADP. Quite a few of the underperformers there can be attributed to injury (Mbeumo/Isak/Wilson/Jota) or greater rotation like Diaby. Some players are finally getting their chance and thriving, such as Mateta, Adebayo, Richarlison and Joao Pedro.
It's interesting to see the improvement in Jackson's numbers who has been much maligned due to his astonishingly bad finishing (7 goals from 10.6 xG) yet is still generating a high number of quality chances, which cannot be ignored. However, if you remove his four goal game against 9-man Spurs, the picture isn't so rosy, with just 8.25 PPS in GW12-20. He might have more competition when he gets back from AFCON, or he could be trusted going forward given Chelsea's FFP issues. It's one of the big gambles that can be taken at this point and a likely not too expensive one either if you have the roster space. On the flip side it's shocking to see just how low Hojlund has fallen. He barely ghosts and unlike Jackson he just isn't generating enough chances with just 3.1 xG, while still underperforming with 1 goal. A monumental waste of FAB, surely his prospects improve in the 2nd half of the season. It's just scary how much they'll have to improve for him to be relevant.
Risers
Dominic Solanke
Given his ADP of 101, clearly hopes weren't too high for Solanke in pre-season despite the optimism over Iraola bringing a more attacking style to Bournemouth. All the attention was given to the wingers and attacking midfielders given how essential they were to his Raya Vallecano team. Solanke didn't exactly set the world alight in the early part of the season though with just 8.5 PPS, and he was incredibly goal dependent given his floor of 4.5 gPPS in that period. But he was generating chances and finishing them, with 4 goals from 3.7xG, although this wasn't enough to make him stand out.
But with Bournemouth's amazing improvement in form, 0.37 xG/90 skyrocketed to 0.81 xG/90 in GW11-20 - those are serious numbers especially when he was reliably finishing them off (8 goals from 7.2 xG). The Bournemouth attack just clicked and Solanke was benefitting massively. He even improved his ghost points to 6.5 gPPS, mainly from a boost in SOT, TkW, and AER. This resulted in a seemingly absurd (for him) 14.5 PPS, which catapulted him to17th in total FPts, quite the boon for anyone who drafted him Round 7/8+. I don't see him keeping up 14.5 PPS - this form is difficult to sustain and the next 6 fixtures look pretty tough for Bournemouth. So I might avoid him for now. However, he could be an interesting buy low candidate after the Man City game in GW26, especially if Bournemouth struggle during this period. I could see him finish in the 11-12 PPS range rest of season and likely even better than that from GW27 onwards.
Raul Jimenez
Jimenez is one of the feel good stories of the season that I don't think anyone was expecting, especially given his start to the season. After recovering from a horrific head injury, he never really looked the same, eventually leading to a move to Fulham this summer, basically a cheap depth signing after losing star striker Mitrovic. Things did not get off to a good start. Through GW1-10 he didn't register a single goal or assist and was regularly subbed off early for Vinicius, averaging just 67 mins per start. His 2.6 PPS tells you everything you need to know about how dire the situation was.
But then Fulham clicked into gear and started scoring for fun, with Jimenez a huge beneficiary. Fulham were generating good chances for him, as his xG/90 ballooned up from 0.2 to 0.5, and he did his part by putting away 5 goals in just 7 starts, with the extra shots increasing his ghosting to 4.9 gPPS. This is not amazing form but the goals helped him to 11.4 PPS. It's also worth noting the bizarre red card he received at Newcastle, which gave him a -5.5 point game in that match. Take that away and he's at 14.3 PPS. Fulham certainly seemed to miss the Mexican while he was banned for a couple of games with embarrassing losses to Burnley and Bournemouth. His return saw them upset Arsenal at home with Jimenez scoring the equaliser. Now a mainstay in the team and a productive striker, things are looking up again for Jimenez. If Fulham can keep giving him those chances, we should see him hit at least 10+ PPS rest of season. In short, he is well worth rostering.
Fallers
Ollie Watkins
Watkins is the 4th highest scorer in total points, and has been a wrecking ball for Aston Villa this season. Yet we have seen a big drop off in his production, with his PPS falling from 17.1 to 10.5. It's worth mentioning the elephant in the room first - the home game against Brighton where he posted 59.5 points, 3 goals (0.8 xG), 2 assists (0.4 xA), 5 KPs and 4 SOT. That likely singlehandedly won managers their GW! I bring this up as it skews his data significantly. Watkins is good for his total number of goals, with 9 from 9.0 xG, but that game led to a significant overperformance in that period and also boosted his ghost point numbers significantly. Watkins isn't a great ghoster (most forwards aren't), but if you remove that match his numbers in GW1-10 are nearly identical to GW11-20 at around 5.7 gPPS. While Watkins is good for his xG, the same can't be said about his xA, he has 8 assists from 4.2 xA, but he also has 3 fantasy assists (2 of which were for winning penalties), which all came in GW1-10.
As we know, these are not very predictable events, so you can see how Watkins looked like a world beater early on in the season, but has fallen away since then. It's worth noting his expected points is at 12.0 xPP90 (with an overperformance at 14.0 PP90), so I would predict that the 10.5 PPS we've seen more recently is on the low end for him. Going forward, a range of 11-13 PPS seems reasonable. If you have him, I'd hold, as Villa have a nice fixture run coming up over the next 10. But I wouldn't be averse to trading him out if you can get someone to bite for a player around his current production - a swap for someone like Son in a gameweek or two could make sense if you have the roster space.
Julian Alvarez
Many are plotting their next move with Alvarez given De Brunye's impending return. The Argentine has had quite the season, making a mockery of his 57 ADP, although he was typically taken in the 4th round in most competitive leagues. Last campaign, the issue was when/if he would get starts. But he has been a mainstay this season, starting every game and putting up serious numbers. In the absence of KDB, he has taken on more of a creative role and has commanded a significant portion of set pieces, which has given him a nice floor of 8.0 gPPS.
Interestingly, this hasn't changed across the two periods. What has changed is his attacking return output, although it's maybe not as you would have expected. In GW1-10, he notched 4 goals from 3.0 xG and provided 5 assists (2 fantasy) from 1.3 xA - a significant overperformance. Yet in GW11-20 he scored just 2 goals from 3.5xG and 2 assists from 3.7 xA, even his SCA/90 went up from 3.6 to 5.6. Alvarez has had more chances to score and has been more creative, putting on more chances for his team mates than earlier in the season, yet he has much less to show for it at 11.7 PPS. Maybe a bit of reverting to the mean, but still his 13.5 xPP90 puts him in elite company and I wouldn't be worrying about his production. However, with KDB's return to the starting lineup looming, managers have some difficult decisions to make regarding Alvarez.
How will KDB affect Alvarez's points? First of all, we have to consider how many games KDB will play. Given his long injury lay-off and the Champions League returning, we could see his minutes heavily managed. A projection of 50-60% starts feels reasonable if he stays fit. Now what happens to Alvarez while they're on the pitch together? First is likely a loss of set pieces, which will affect his floor. Alvarez relies on set pieces for 25% of his points, which is significant at over 3 points per game. But if we halve that given KDB's availability, we're only losing about 1.5-2 points on average per game. We'll likely not continue to see Alvarez start every game either. We could see him get something like 80-85% starts going forward. His role on the pitch may also change when he plays together with KDB but it's yet to be seen how that will affect his output. Conservatively downgrading him to 11-12 PPS and a few less starts seems reasonable. So those that roster Alvarez right now, I'd advise not to sell him off for cheap, he'll still be a great asset, but just know there are factors which don't appear favourable to his points or minutes going forward. It's still unclear how it will play out, so just wait it out and see, anything can happen in football, and KDB could get reinjured in his first start back (again)!
Wrap Up
So there you have it. I hope this has provided some food for thought on player value. It's very easy to get bogged down by overall season numbers or very immediate form. So remember you can check out the data in the Useful Draft Metrics page. Sorting by Form (last 6 PPS) is always a good place to start to see who might be on a hot or cold streak when compared to season long production. To give yourself another edge, it can also be useful to look at the xPP90 data to see how players are performing based on their underlying data, which also removes nonrecurring events like red cards and own goals. This data can help you identify players that are over or underperforming and see if they are good buy low or sell high candidates.
For all the latest from The Draft Society, follow @Draft_Society on Twitter!
And for more in-depth and exclusive resources, become a member of The Inner Circle.
Komentáře