Premier League Analysis

Nottingham Forest’s 2–2 at Liverpool contained real danger on both sides, including a 0.53-xG Igor Jesus attempt blocked, an Isak goal worth 0.84 xG and a Gibbs-White penalty. The penalty contributed materially to Forest’s 2.30 xG. Their 2–1 win at Villa also involved improved attacking supply after halftime: 17 shots and 1.61 xG, compared with five shots and 0.31 before it. Even there, Delap’s opener came from a chance worth only about 0.02 xG and the third match goal arrived in the 88th minute. These higher totals combined better access with contributions that are not guaranteed to repeat.

Against Coventry, Forest took 24 shots but half were blocked and the whole attack produced just 1.10 xG. Against Leeds, the combined total was only 1.13 xG and the sole goal came from a late direct free kick. Tottenham’s 0–0 was less secure evidence for a low total: Van de Ven missed a 0.82-xG chance from a corner, although Forest themselves created only 0.65 xG. Last season Forest generated 2.32 xG at home to Spurs and won 3–0, so the same fixture label conceals a substantial change in their own threat. Higher totals need Forest to escape the Coventry-style blocked, low-value shooting pattern, with opposition danger or enough quality to carry the scoring themselves.

Manchester United’s 5–2 win over Ipswich contained 4.56 United xG, with 3.82 arriving after halftime. Even after separating a penalty and a Greaves own goal, United repeatedly reached valuable chances. Ipswich also retained a late scoring route through Akpom’s 0.97-xG finish. Palace’s 2–3 defeat had a different shape: Ipswich created 1.67 xG against a side reduced to ten men in the 34th minute, while Palace scored twice from attempts each worth about 0.04 xG. The five-goal result needed difficult Palace finishes as well as Ipswich’s numerical advantage.

Sunderland’s 2–1 defeat stayed below four despite Ipswich missing a 0.61-xG Maeda chance; Ipswich’s second ultimately came from a difficult late Clarke finish. Liverpool’s 2–0 win was lower in underlying danger, with just 1.39 combined xG, whereas Everton’s 1–0 win contained 3.68 and significant missed chances, before Fatawu’s dismissal reduced Ipswich’s late options. For another four-goal match, United-style sustained defensive exposure is the clearer repeatable concern. The Palace score is less useful without its red card and conversion conditions, and a low recent total cannot be assumed to mean Ipswich have stopped allowing chances.

Sunderland’s three goals at City were supported by 3.57 xG, with Brobbey converting two exceptionally valuable chances worth about 0.92 and 0.97. This was meaningful access to dangerous positions, not simply three difficult shots all going in. Their goals in the shared-scoring games at Ipswich and Brentford had different foundations: Angulo scored a direct free kick at Ipswich, while Le Fée equalised from the penalty spot at Brentford. Those replies do not establish the same open-play supply seen at City.

Arsenal’s clean sheet against Sunderland also needs care. Le Fée had a penalty saved and Mukiele’s 0.34-xG corner attempt was stopped, so Sunderland’s blank included missed set-piece opportunities. Against Fulham, Sunderland won with Isidor’s valuable second-half chance after he had already hit the post with another, but Fulham failed to put any shot on target. Last season’s City visit brought Sunderland only 0.71 xG and one target, compared with nine targets this time. For future goals at both ends, an opponent allowing Brobbey-style clear chances is more convincing than dependence on a free kick or penalty, and the opponent still needs its own effective shooting route.

Leeds’ 4–1 win over Newcastle included a Miley own goal and three Leeds scoring shots worth approximately 0.11, 0.20 and 0.05 xG. Leeds generated 1.91 xG, so there was real attacking activity, but the four-goal return was boosted by the own goal and accurate finishing. Brighton–Leeds produced only two goals despite 4.67 combined xG. Leeds missed a substantial corner chance and had Calvert-Lewin’s 0.31-xG effort saved; Brighton later missed valuable opportunities of their own. The larger score did not come from the larger pool of chances.

The other low totals also differ. Forest–Leeds had only 1.13 combined xG and was decided by Stach’s late direct free kick. Brentford–Leeds had 3.13, including a missed 0.50-xG Thiago chance, while Palace–Leeds contained a saved 0.52-xG Calvert-Lewin effort and further Palace misses. Last season’s away Brighton match reached three goals with Leeds generating only 0.50 xG; this time Leeds contributed 2.49 but the match stopped at two. For future totals, distinguish genuine restriction from missed chances. Newcastle’s score alone is weak evidence for another high total unless the opponent again allows substantial danger rather than merely being punished by efficient finishing.

Crystal Palace’s blank at Everton was not the result of being kept away from goal. Mateta missed a 0.33-xG fast-break chance and hit the post with another worth 0.37; Nketiah had a 0.60-xG attempt saved. Palace finished with 1.96 xG. Ipswich also created 1.44 xG without scoring, including a 0.42-xG Leif Davis miss. Those clean sheets depended on good chances going unfinished. Tottenham’s blank was more consistent with restricted shooting: eight of their 14 attempts were blocked and their total chance quality was only 0.64 xG.

Everton’s own scoring is the other condition. Their Bournemouth equaliser came from a Tarkowski corner finish in the 90th minute, after Barry had two much better chances saved. Against United, Everton’s goals arrived in the 83rd and 90th minutes from shots each worth less than 0.04 xG. Last season Palace scored at Everton despite producing less xG, 1.53, than in this season’s blank. Another opponent repeatedly reaching valuable chances should make the clean-sheet record less persuasive, but both teams scoring still requires Everton to finish their own opportunities rather than depend on another difficult late rescue.

Hull’s 0–0 with Villa contained 1.74 Villa xG, including McGinn missing a 0.62-xG chance and Buendía a 0.40 opportunity. Hull’s own attack contributed only 0.58 xG. At Coventry, Hull won 1–0 after Awoniyi had a 0.72-xG attempt saved before halftime; Millar eventually finished a fast break in the 82nd minute. Those low totals combined modest Hull supply with significant opposing chances going unfinished. They are not all evidence that Hull prevent dangerous opportunities.

The higher totals took different routes. Chelsea–Hull reached four goals from 2.38 combined xG, with all three first-half scoring chances individually worth less than 0.15 xG. Newcastle–Hull reached three after Newcastle scored twice in seven minutes and Hull then produced 19 second-half attempts while chasing. The 2–0 United win was another warning: United generated 1.83 xG without scoring, and Hull’s second-half attack virtually disappeared after their set-piece goals. For the next total, the question is whether Hull can add sustained threat and whether the opponent keeps reaching clear chances. A narrow recent scoreline is less persuasive when the opposing forwards have been missing opportunities that normally carry much more danger.

Fulham’s 7–1 corner win over Palace came with a 63–33 advantage in final-third entries, and they added five corners after halftime while Palace added none. Sunderland was a comeback in the corner contest: Fulham trailed 1–3 at the interval, then took three unanswered corners as seven second-half shots were blocked and Sunderland made 25 clearances. The final 4–3 margin depended on Fulham maintaining an attacking response, despite losing the match itself.

United exposed what happens when that response disappears. Fulham took only one second-half corner while United took ten; Fulham made 25 clearances and lost the full count 3–13. Liverpool shows that pressure alone does not make that collapse inevitable: Fulham won corners 8–5 even though Liverpool led final-third entries 70–46, because Fulham kept creating their own corner endings and added five after halftime. Last season Fulham also won the corner count at Liverpool, 9–6, whereas their home United result flipped from a 9–6 win to this season’s heavy defeat. For a future positive margin, Fulham need an outlet that keeps producing corners when the opponent pushes; defending more territory is survivable only if their own supply remains active.

Aston Villa put five of seven shots on target against Forest and six of 15 at Tottenham, and both games saw both sides score. At Hull, Villa had 73% possession and 1.74 xG but managed only one shot on target from 13 attempts. McGinn missed a chance worth 0.62 xG and Buendía another worth 0.40. That blank involved opportunities going unfinished, whereas the Arsenal blank came from just 0.33 xG and no target at all. Villa’s scoring risk therefore changes depending on whether the opponent restricts the chances or Villa fail to execute them.

The goals against Forest and Spurs should not be overstated either. Alysson’s Forest goal came from a corner chance worth about 0.07 xG; Jackson and Buendía’s second-half Spurs goals were worth about 0.06 and 0.02. Villa’s improved accuracy helped, but some finishes were unusually difficult. Tottenham’s reply also arrived late, through goals in the 86th minute and stoppage time. Last season’s Brighton visit featured 2.41 Villa xG and ten targets, compared with 0.29 and none this time, when Villa also lost a player at 4–0 down. For future shared scoring, Villa need both a usable chance route and an opponent still capable of threatening; more possession by itself does not solve Villa’s side.

Gabriel attempted 74 passes at Villa but only 38 against Chelsea, with both appearances lasting 90 minutes. Most of the difference came in his own half: 42 passes at Villa versus 12 against Chelsea. His advanced passing changed much less, from 32 to 26. The low Chelsea total therefore reflected reduced involvement earlier in Arsenal’s circulation, not simply an inability to take part beyond halfway.

Coventry shows why a large Arsenal team total is insufficient. Arsenal attempted 616 passes there, but Gabriel contributed 57; at Villa, the smaller team total of 530 still produced his 74. Brighton brought 64 with 35 own-half passes, while Sunderland’s smaller team pool of 388 left him on 44. The higher case needs two conditions: enough Arsenal possession to circulate, and repeated use of Gabriel within that circulation. An opponent allowing Arsenal to move the ball without frequently returning it through him can leave his personal count lower even when the team dominates overall.

Bukayo Saka’s 47 passes at Aston Villa came with 77 touches and 29 recorded ball carries: Arsenal were involving him repeatedly on the ball, not only at the end of moves. Thirty-eight of his passes were played in the opposition half. Chelsea and Sunderland produced a different workload despite nearly identical playing time—28 and 27 passes, 49 and 48 touches, and 15 and 18 carries. Saka shot four times in each of those games, compared with twice at Villa. The larger attacking output therefore did not require the larger passing count: he was reaching shooting opportunities while taking part in fewer passing actions.

Brighton makes the team-possession distinction especially clear. Arsenal had 60% possession and 511 passes, close to their 61% and 530 at Villa, yet Saka attempted only 21 passes in 81 minutes, including 14 beyond halfway. His own-half passing changed little, seven versus nine; the missing volume was advanced involvement. He still supplied three key passes, so a low total did not mean no creative contribution. Last season’s matched away fixtures also differed: Saka made 36 passes in 79 minutes at Villa but 14 in 90 at Brighton, while his Sunderland away total was 41 in 90, rather than the current 27. That last comparison prevents treating an opponent label as a fixed workload. A larger Saka passing expectation needs sustained personal involvement beyond halfway as well as minutes; Arsenal having the ball or Saka producing shots is insufficient.

Ipswich supplied the stronger model for another Manchester United game reaching four goals. United generated 3.82 xG after halftime alone, with a Mbeumo effort worth 0.92 hitting the post and a Tielemans chance worth 0.49 saved in the same minute before the later goals. The eventual 5–2 score included an own goal and a penalty, but substantial chances existed beyond those contributions. Ipswich also retained a scoring route, finishing a 0.97-xG opportunity late on. This was a match in which one attack could do most of the work while the other still threatened.

Everton’s 2–2 draw offers less dependable support. The four goals came from only 1.70 combined xG; Everton’s two scoring shots were worth about 0.04 and 0.03 xG, while Mbeumo’s opener was about 0.02. The match did not reach four until the 90th minute. Fulham is the counterweight: 41 combined attempts and 3.22 xG still produced just two goals, one an own goal, as blocked shots and missed chances prevented the volume becoming a large score. For another four-goal match, the convincing route is Ipswich-style sustained high-value supply, with enough opposition threat to keep adding to it. A busy match or Everton’s recent final score alone does not establish that environment.

Jan Paul van Hecke attempted 108 passes against Everton and 82 against Newcastle, playing 90 minutes in both. His own-half total was identical at 40. Every extra pass in the Everton game came beyond halfway, where his count rose from 42 to 68. Tottenham’s larger team passing pool helped, but the specific personal mechanism was repeated advanced circulation through Van Hecke, rather than extra defensive-half recycling.

Villa limits a simple possession forecast. Tottenham had 64% possession, more than against Everton, yet Van Hecke attempted only 68 passes: 28 in his own half and 40 in the opposition half. Brentford also brought just 61 despite 60% possession. His touches moved with that involvement, from 127 against Everton to 83 against Villa. For another very high count, Spurs need to keep using him as a connection beyond halfway while retaining his deeper workload. A territorial advantage that directs the ball through other players can leave him well below the Everton ceiling.

Brighton’s four first-half goals against Villa included a Lindelöf own goal, but the attacking danger was substantial beyond that contribution. De Cuyper scored a chance worth 0.96 xG, and Hinshelwood finished opportunities worth 0.87 and 0.44. Brighton generated 3.03 xG before halftime. All four goals preceded João Gomes’s 40th-minute dismissal, so the early scoring cannot be attributed to the later numerical advantage. Arsenal’s two-goal halftime deficit had a very different basis: Groß and Kostoulas scored chances worth roughly 0.05 and 0.06 xG, while Brighton’s whole half totalled only 0.36 xG. Both halves cleared two goals, but only Villa’s provided repeated high-value opportunities.

The same-venue comparisons reinforce the distinction. Last season’s Villa game reached 2–2 at halftime with dangerous attacks on both sides; this time Villa created just 0.16 first-half xG, so Brighton had to carry the scoring. Against Arsenal last season, Brighton generated 0.42 xG before halftime and failed to score—slightly more chance quality than in this season’s two-goal half. A quick Brighton lead is therefore more credible when clear chances recur, as against Villa. The Arsenal result demonstrates what excellent finishing can achieve, but it does not establish a reliable early scoring rate from similarly difficult attempts.

Liverpool’s 2–0 win at Ipswich was decided by Alexander Isak’s assisted finishes in the sixth and ninth minutes, from attempts worth approximately 0.08 and 0.14 xG. Liverpool put six of seven first-half attempts on target, even though only two came from inside the box and the half’s chance value was 0.43 xG. Their expected goals on target reached 1.70. The difference was execution: relatively modest opportunities became much more demanding shots once struck.

That route should be separated from repeatedly creating clear chances for a second goal. Liverpool’s second-half activity at Ipswich dropped to three shots and 0.23 xG, leaving the full-match return at two goals from 0.66 xG. Fulham then allowed a higher 1.13 xG but held Liverpool scoreless: five attempts were blocked, six missed the target, and only three reached it. Isak’s best late opening, a 0.31-xG corner chance, was missed in stoppage time. Liverpool scored twice in three of the five recorded league games, but Ipswich demonstrates a finishing-led success rather than a volume baseline. An opponent allowing repeated accurate efforts offers a route to two goals even without several major chances; if the accuracy falls toward the Fulham pattern, a similar amount of attacking activity need not produce the same return.

Manchester City reached two goals against Bournemouth through a late corner finish by Marc Guéhi and a 90th-minute Joško Gvardiol goal. The chances had existed earlier—Guéhi missed a 0.77-xG corner chance and Haaland had a 0.49-xG effort saved—but the second goal still required a late breakthrough. At Palace, City instead turned four relatively modest chances, each below 0.17 xG, into four goals. Sunderland brought another combination: a direct free kick and a corner goal helped City reach three before halftime from only 0.67 xG, before substantially better chances arrived after the interval. These successes do not establish one repeatable, early scoring pattern.

Coventry’s 1–0 defeat shows why a single City goal need not mean their attack was contained. City produced 2.12 xG, including a 0.46-xG Haaland chance saved and a 0.62-xG Foden attempt recorded as saved. The derby’s 1–0 win, after Foden’s 23rd-minute dismissal, had much thinner supply: six shots, only two from inside the box, and most of the danger concentrated in two Haaland opportunities. The useful distinction for another multiple-goal game is whether City can repeatedly reach good shooting positions, as against Coventry and Bournemouth, or need unusually efficient finishing from a small pool of chances. Palace’s four and Sunderland’s five should not be treated as normal returns from those same shot profiles.

Fulham opened the scoring against Crystal Palace through King’s 11th-minute fast break and went on to create 1.72 first-half xG. City also scored first and produced 11 first-half shots, six on target. Those games support concern about opponents repeatedly reaching Palace’s goal. Everton was different: Palace had already missed a 0.33-xG Mateta break, seen a 0.60-xG Nketiah chance saved and hit the post through Mateta before Dewsbury-Hall scored a roughly 0.04-xG effort in the 42nd minute. Palace conceded first despite creating the better opportunities.

Ipswich shows that Palace can take the lead without resolving that vulnerability. Khalaili scored first from a chance worth about 0.04 xG; Ipswich equalised before Disasi’s 34th-minute dismissal, then scored again before halftime. Leeds produced no opener at all, with both teams failing to finish their better second-half chances. Last season at Everton, Palace led at halftime from only 0.63 xG, whereas this season’s 1.57 yielded no goal. For a future first-concession judgment, separate opponents likely to dominate chance creation from games in which Palace’s own wastefulness keeps the opening goal available to either side.

Chelsea’s five recorded league games all reached three goals, yet their own attacking contribution changed sharply. Fulham and Brighton allowed Chelsea 2.24 and 2.93 xG, including valuable early fast-break and restart chances. Arsenal, Hull and Brentford then held Chelsea below one xG each: 0.39, 0.89 and 0.83. The high match totals survived because Arsenal and Hull scored twice and Brentford scored three, rather than because Chelsea maintained the same chance creation.

Brentford shows the distinction most clearly. After a goalless first half, Chelsea accumulated nine corners but had six of eight attempts blocked and no shot on target. Brentford instead produced 2.26 second-half xG, including Jaidon Anthony’s 0.96-xG corner finish and Igor Thiago’s 0.60-xG fast-break goal. The opponent was reaching decisive chances while Chelsea’s visible pressure was ending at defenders. Last season’s same fixture finished 2–2 with Chelsea at 1.25 xG and Brentford at 1.43; the current 0–3 came with a weaker Chelsea attack and a much stronger Brentford chance supply. That is why another high total needs an opponent capable of creating its own valuable chances if Chelsea remain restricted. Chelsea corners or the five-match score sequence alone cannot supply the missing goals.

Brentford’s early opener against Tottenham followed substantial danger: Collins missed a 0.31-xG corner chance in the fourth minute before Lewis-Potter scored a 0.73-xG opportunity in the 12th. Leeds offered a different advantage, giving Brentford 11 first-half shots while producing only two worth 0.04 xG themselves; Schade eventually scored in the 41st minute. Those were openings built on either a valuable early chance or sustained suppression of the opposing attack.

The Chelsea and Sunderland games remained goalless until the 61st and 57th minutes. Anthony’s Chelsea opener came from a 0.96-xG corner chance, showing how a restart can resolve a previously balanced, low-quality first half. Bournemouth is the caution: Brentford scored first through Schade after the hosts had already missed a better António Silva chance. Last season’s home Tottenham fixture stayed 0–0 with only 0.92 Brentford xG, compared with 3.87 this time. Scoring first is therefore not a fixed timing trait. The future case strengthens when Brentford have early clear chances or can keep the opponent quiet until their set-piece opportunity arrives; neither automatically means a goal before halftime.

Bournemouth scored first at Newcastle through Tavernier’s 0.60-xG chance in the ninth minute, a strong example of turning an early opening into the lead. At City, Tavernier scored in the 26th minute despite Bournemouth having only 31% first-half possession. Against Everton, Scott’s 41st-minute fast-break goal followed a much better Everton chance: Barry’s 0.82-xG attempt had already been saved. The opening-goal successes therefore came through different routes—valuable early service, efficient use of limited possession, and surviving the opponent’s better opportunity before breaking forward.

Brentford reversed that order. Bournemouth missed a 0.33-xG set-piece chance in the seventh minute, then Schade scored first in the 34th. Liverpool also scored first only after Bournemouth missed Christie’s 0.43-xG chance four minutes earlier. These failures show why getting the first good chance and scoring the first goal are separate. An opponent that allows Bournemouth an early clear chance offers a credible route, but a recent sequence of opening goals cannot establish dominance or guarantee conversion. If Bournemouth are mainly relying on saves at their own end before a break, the first-goal outcome remains much more finely balanced than their possession or result alone suggests.

Tottenham’s 0–0 with Everton produced 11 corners while both attacks struggled to get clean shots away. Eight of Spurs’ 14 attempts and six of Everton’s 12 were blocked: 14 blocks across 26 shots. Spurs reached 37 penalty-area touches but created only 0.64 xG; their pressure repeatedly met a defender before it could become an accurate finish. Everton added six corners to Tottenham’s five, so neither side needed an exceptional individual count to take the match into double figures.

The same fixture last season finished 1–0 with a 7–7 corner count. Tottenham again had many attacks stopped—nine blocked attempts from 20 shots—and Everton added three blocks from ten. This is a two-match comparison, not proof of a permanent matchup identity, but both meetings show how defensive resistance can coexist with enough attacking activity for a large corner total. Brentford’s current 3–0 win over Spurs reached ten corners through a different balance, seven from Brentford and three from Tottenham. That matters for the next opponent: a Spurs attack being blocked can still contribute corners, but the Everton example depended on the other side adding six or seven. If opposition attacks disappear and Spurs do not make up the missing count, the same low-scoring conditions need not produce the same corner total.