What’s On It

The winner receives a minimum of £110m in Premier League broadcast revenue, the so-called ‘richest match in world football’, plus three years of parachute payments if relegated. For the loser, the cost is not only financial but emotional: Hull owner Acun Ilicali has hinted at legal routes should his side lose, while Southampton, the expelled team, will begin next season on -4 points in the Championship. Boro’s owner Steve Gibson, who pushed for the expulsion, now sees his club with a second chance at promotion. The financial gap between the Premier League and Championship is vast: the winner’s broadcast revenue alone exceeds the entire annual turnover of most Championship clubs. Hull face a particularly stark trajectory question — Sergej Jakirović took over at the start of 2025-26 with the club facing the prospect of relegation to League One, and the play-off run has rewritten that arc. Boro’s expected-goal profile (+27.94 xGD, the best in the play-off field) suggests a side built for the top flight, while their on-pitch elimination by Saints leaves a complicated emotional narrative now resolved at the EFL level rather than over 180 minutes.

The model headline

The logistic regression calibration (LR cal) narrowly favours Hull at 44.7%, with Boro at 33.3% and the draw at 22.0%. The Poisson model, which does not incorporate lineup ratings, has no clear favourite at result level — Hull 37.6%, draw 25.7%, Boro 36.6% — the two sides level on the night with an unusually low draw share. The conditional-modal scoreline, restricting to Hull-win outcomes, is Hull City 1-0 Middlesbrough. The LR model’s draw probability is likely understated—it has not predicted a draw in any Championship fixture this season—so the true draw chance is probably closer to Poisson’s 25.7%.

The Final Championship Table 2025-26

Coventry and Ipswich up automatically. Saints (4th, 80 pts) were expelled from the play-offs. Boro (5th, 80 pts) reinstated. Hull (6th, 73 pts) the lowest seed.

Hull’s Road

Hull City entered the play-offs as the lowest seed, having finished sixth with 73 points and a goal difference of +4. Their semi-final against third-placed Millwall began with a tense 0-0 draw at the MKM Stadium, where Hull made four attacking substitutions after 69 minutes. In the second leg at The Den, Mohamed Belloumi scored the opener in the 64th minute and assisted Joe Gelhardt for the second in the 79th, securing a 2-0 aggregate win. Hull assistant manager Dean Holden noted the team’s adaptability: ‘We’ve adapted well to teams we’re playing against, to injuries and we’ve found ways to win.’ Sir Tom Courtenay, the 89-year-old actor and Hull City supporters’ club president, praised manager Sergej Jakirović’s tactical adjustment between legs: ‘I was impressed at the way the manager adjusted the team after the first leg. It really worked.’ The clean sheet at The Den, and the 2-0 aggregate margin, came against the third-seeded side in the play-off field. Hull’s xG in the two legs was 0.56 and 1.63, while Millwall’s was 0.56 and 0.94 — a slim chance-creation edge that translated cleanly. Jakirović, appointed at the start of 2025-26 after replacing Ruben Selles, took over a Hull side facing the prospect of relegation to League One, and the semi-final win at The Den underlined how far the club has travelled since. John Egan made 45 appearances at the back; Belloumi’s 22 was modest but his SF2 goal and assist were season-defining contributions.

Middlesbrough’s Road

Middlesbrough finished fifth with 80 points and a league-best +27.94 xG difference among play-off teams. Their semi-final against Southampton began with a 0-0 draw at the Riverside, despite Boro dominating the xG battle 1.82 to 0.53. The second leg at St Mary’s ended 2-1 to Southampton after extra time, with Boro’s Hayden Hackney ruled out by manager Kim Hellberg. The match was halted after Boro’s Luke Ayling accused Saints’ Taylor Harwood-Bellis of discriminatory language. On 19 May, an independent disciplinary commission expelled Southampton from the play-offs and reinstated Boro, a decision upheld on appeal the following day. Boro’s statement welcomed the ruling: ‘We believe this sends out a clear message for the future of our game regarding sporting integrity and conduct.’ Boro have not played a competitive match since 12 May — an 11-day gap before the final. Hellberg, the Swedish head coach, has had to recalibrate a squad whose on-pitch elimination was reversed by ruling rather than by performance. The potential return of Hayden Hackney, missed for the second leg, is a key fitness question — his 7.24 average rating across 38 appearances is the highest in the Boro squad and his five goals plus seven assists from midfield make him the side’s chief creative outlet. The Ayling-Harwood-Bellis incident in the second leg added a layer of tension; referee Andy Madley filed a report. Boro’s reinstatement came four days before the final, leaving limited preparation time for an opponent shape they had not been preparing for.

What Happened — Spygate

On Thursday 7 May 2026, Southampton analyst intern William Salt is reported to have parked at a golf club near Boro’s Rockliffe Park training base, walked to a raised vantage point, and pointed a mobile phone at Boro’s training session while wearing in-ear headphones. Boro staff believe he may have been live-streaming the session. Approached by a member of Boro staff, he refused to identify himself, deleted content from his phone, changed clothes in golf-club toilets and fled. Boro’s photographer matched him to a photo on Southampton’s website. The incident occurred 48 hours before the first leg of the semi-final, a match that ended 0-0. Boro’s head coach Kim Hellberg described the emotional impact: ‘It breaks my heart.’ He added: ‘If we hadn’t caught that man that they sent up five hours to drive, you would sit there and say well done in the tactical aspect of the game and I would go home and feel like I’ve failed.’ The spying was not an isolated event: Southampton later admitted to similar breaches before matches against Oxford United in December 2025 and Ipswich Town in April 2026. The EFL charged Saints with breaching Regulation 3.4 (utmost good faith) and Regulation 127 (no observing another club’s training in the 72 hours before a fixture). The use of an intern, the attempt to conceal identity, and the premeditated nature of the operation all featured in the disciplinary proceedings.

The Ruling and the Appeal

Southampton admitted breaching EFL Regulation 3.4 (‘utmost good faith’) and Regulation 127 (no observing another club’s training in the 72 hours before a fixture). The charges covered three separate incidents: Oxford United (December 2025), Ipswich Town (April 2026), and Middlesbrough (May 2026). On 19 May, an independent disciplinary commission expelled Saints from the play-offs, applied a four-point deduction for the 2026-27 season, and issued a reprimand. The EFL arbitration panel rejected Saints’ appeal on 20 May. ‘A league arbitration panel has tonight dismissed Southampton Football Club’s appeal against the independent disciplinary commission’s sanction following the admittance of multiple breaches of EFL regulations.’ No CAS route is available. The panel’s decision was final and binding. The expulsion meant that Boro, who had lost the semi-final on the pitch, were reinstated to the final. The four-point deduction will apply to Southampton’s 2026-27 Championship campaign, effectively starting them on -4 points. The reprimand covers all three charges. The speed of the process—from the spying incident on 7 May to the final appeal rejection on 20 May—reflected the EFL’s determination to resolve the matter before the play-off final. Saints’ CEO Phil Parsons described the punishment as ‘manifestly disproportionate to every previous sanction in the history of the English game’, but the panel disagreed.

Context and Consequences

The only prior Championship spying case was Leeds in 2019, fined £200,000 for sending staff to watch Derby train. Regulation 127 was introduced as a direct result. The closest precedent for sporting sanction is the 2024 Olympics women’s tournament, where FIFA deducted six points from Canada and banned three staff for drone-spying on New Zealand. The FA may bring separate charges against individuals. Wrexham, who finished seventh, have an open compensation thread; forward Josh Windass questioned: ‘But why isn’t the play-offs starting again with the four other teams? Boro v Hull would have been the semi! Confused.’ Saints’ head coach Tonda Eckert’s position is effectively untenable. The club’s owner, Sport Republic, has invested heavily, and the loss of promotion revenue is a significant blow. The EFL’s decision has been widely debated: some pundits, like Paul Robinson, argued it was justified: ‘The integrity of the game is of the utmost importance.’ Others, like Jo Tessem, acknowledged the severity: ‘We have rules and we need to follow them. We have been punished hard for not following very simple rules.’ The case has also raised questions about the adequacy of Regulation 127 and whether further deterrents are needed. For Hull, the late change of opponent has been a source of frustration, with owner Acun Ilicali expressing displeasure and hinting at legal action if his side loses.

Form & Expected Goals

Hull City’s season is defined by overperformance: they scored 12.24 goals more than expected, finishing with a -18.16 xG difference but a +4 goal difference. Middlesbrough, by contrast, posted a +27.94 xG difference, the best among play-off teams, and a +25 goal difference. Boro have had an 11-day gap since their last competitive match on 12 May, while Hull played their semi-final second leg on 11 May and have had a full week to prepare—though they originally planned for Southampton. Hull assistant Dean Holden acknowledged the late change: ‘If anything changes in the next few days then we’ll have to adapt to that.’ Hull’s +12.24 goals-minus-xG represents a substantial season-long overperformance — a profile that points to clinical finishing and over-performing defence rather than dominant chance creation. Boro’s xG dominance, meanwhile, indicates a side that consistently creates high-quality chances and limits opponents. The 11-day break for Boro could be a double-edged sword: it allows rest and recovery, but may also disrupt rhythm. Hull’s preparation has been complicated by the opponent switch, but Holden’s comment reflects a squad that has adapted all season. The underlying numbers suggest Boro are the better side over 46 games, but Hull’s ability to win matches they ‘shouldn’t’ has been a hallmark of their campaign.

Form compare — last six

Rolling xG — 10-match window

Home vs away split — season-long

Tactical Preview

Under Sergej Jakirović, Hull have shown a willingness to make multiple attacking substitutions, as seen in the semi-final first leg when four attackers were introduced after 69 minutes. Middlesbrough boast the best home defence in the play-off field, conceding only 18 goals in 23 home league games, though Wembley is neutral. Set pieces are a weapon for Boro via Matt Targett. The potential return of Hayden Hackney, Boro’s highest-rated midfielder who missed the semi-final second leg, could be pivotal. Hull’s John Egan provides defensive leadership, while Mohamed Belloumi emerged as the semi-final hero with a goal and assist in the second leg. Jakirović’s tactical lever — visible in SF1, when Millar, Belloumi and Gelhardt were all introduced inside six minutes from the 69th — is bench-based aggression: change the game from off the pitch when chances are not arriving on it. The midfield battle will be a chess match: Hull’s Slater, Crooks and Belloumi against a Boro engine room whose form-line runs through Hackney (if fit), Browne and Morris. Defensively the records diverge sharply: Hull conceded 66 goals to Boro’s 47, but the xG-against gulf is wider still — Hull 75.92 vs Boro 41.07. By that measure Hull were comfortably the most over-performing defence in the play-off field; Boro the most genuinely solid. At Wembley a single set piece (Matt Targett carries Boro’s deliveries; John Egan is Hull’s aerial threat from defence) could decide the £110m question.

xG scatter — fixture-by-fixture

Each dot is a league fixture: xG produced (x-axis) against xG conceded (y-axis). Below the diagonal = outperformed underlying numbers in that game.

Key Players

For Hull, top scorer Oliver McBurnie (15 goals) leads the line, supported by Joe Gelhardt (14 goals, including the winner in the 1-0 H2H at Riverside and Hull’s second in the semi-final). Mohamed Belloumi, the semi-final matchwinner, and Lewis Koumas, a Liverpool loanee, add creativity. For Middlesbrough, Morgan Whittaker (14 goals, 7 assists) is the chief threat, with Tommy Conway (11 goals) and David Strelec (7 goals) providing support. Dael Fry (7.27 average rating) anchors the defence, while Hayden Hackney’s return would bolster midfield. Riley McGree, who scored in the 4-1 H2H win, is another danger. Hull’s midfield also features Kyle Joseph, who assisted the only goal in the 1-0 H2H win at Riverside. John Egan, with 45 appearances and 3 goals from defence, provides experience at the back. For Boro, Luke Ayling’s presence at right-back adds veteran nous, while Solomon Brynn has been reliable in goal. The individual battles could decide the match: McBurnie vs Fry, Gelhardt vs Targett, and Belloumi vs Ayling. Hull’s ability to create chances from set pieces, with Egan a threat, will test Boro’s organised defence. Boro’s Whittaker, who has two goal involvements in the 4-1 H2H win, will look to exploit any space behind Hull’s full-backs.

Joe Gelhardt — recent form (Hull, 14 league goals)

Morgan Whittaker — recent form (Boro, top scorer)

BTP Watch — Lewis Koumas

Hull’s tracked player. Three Championship goals in only 619 minutes since his Liverpool loan arrival — the kind of bench cameo Jakirović has used decisively in this campaign.

Head to Head

The two sides met twice this season, each winning away. Middlesbrough thrashed Hull 4-1 at the MKM Stadium in December, with Morgan Whittaker scoring and assisting, while Hull won 1-0 at the Riverside in December, Darko Gyabi scoring the only goal despite Boro dominating xG 1.47 to 0.29. Over the last five seasons, the head-to-head is perfectly balanced: five wins each and two draws in 12 league meetings. They have never met at Wembley or in the play-offs before. The December 4-1 win was Hull’s worst home defeat of the season, with Boro 4-0 up by half-time. The return fixture at the Riverside was a classic smash-and-grab: Gyabi’s 12th-minute goal at the Riverside held up despite Boro’s xG of 1.47 to Hull’s 0.29 — a textbook smash-and-grab that epitomised Hull’s season: outperforming xG to win games the underlying numbers said they should not have. The H2H xG aggregate this season is Boro 2.76 to Hull 1.33, reflecting Boro’s dominance in open play. Hull have shown they can win even when outplayed. The historical balance suggests no clear psychological edge either way, though Boro’s 4-1 December win at the MKM is the sharpest single-match reference point in living memory. The two draws in 12 meetings since 2019 indicate that draws are relatively rare in this fixture — a footnote consistent with the Poisson model’s 25.7% draw probability sitting above the historical 16.7% H2H rate.

H2H — last five meetings

The Wembley Factor

Wembley is a neutral venue; the fixture’s ‘home’ designation for Hull is a scheduling convention only. There is no second-leg cushion — one game, then extra time, then penalties if required. The 90,000-capacity stadium will be split roughly evenly between the two travelling supports. The occasion is the biggest single match in the English football calendar in terms of financial reward — the £110m PL revenue prize is the largest in club football globally. For players, the scale is a known unknown; neither set of fans has seen their side at a Championship play-off final this decade. The neutral venue removes any meaningful home advantage; the model treats Wembley as ‘home’ for Hull purely as a scheduling convention, which slightly inflates Hull’s LR cal probability. In a meaningful sense both sides arrive on equal terms, with one critical asymmetry: Hull have played within the last fortnight, Boro have not played competitively for 11 days. Whether that lengthens Boro’s recovery or shortens their rhythm is a question the data cannot answer cleanly. The two sides have never met at Wembley before; Saturday writes the first chapter of that head-to-head.

Goal-scoring trend — xG vs actual (season-long)

The Model — LR Calibration

The logistic regression calibration gives Hull a 44.7% win probability, Boro 33.3%, and the draw 22.0%. Hull is narrowly favoured, with an edge of 11.4 percentage points over Boro. The model uses lineup ratings, rolling form, and venue; Wembley is treated as ‘home’ for Hull by convention, which slightly inflates their probability. A known flaw is that the LR model has not predicted a draw in any Championship fixture this season, so the draw probability is likely understated—the true chance is probably closer to the Poisson estimate of 25.7%. The LR calibration is based on a three-class softmax output, and the model’s tendency to avoid predicting draws is a known pathology that will be addressed in an off-season retrain. Despite this, the model’s relative probabilities between Hull and Boro are informative. The features that drive Hull’s edge include their semi-final momentum (a 2-0 aggregate win over third-seeded Millwall), the quality of their starting lineup as rated by the model, and the venue effect. Boro’s longer gap since competitive football may also be captured in the rolling form feature. The LR model does not directly incorporate season-long xG, which explains why it favours Hull despite Boro’s superior underlying numbers.

The Model — Poisson + Dixon-Coles

The Poisson model, which conditions on rolling team scoring and conceding rates, produces near-identical attacking rates: λ_H = 1.33, λ_A = 1.31. The probabilities are Hull 37.6%, draw 25.7%, Boro 36.6% — no clear favourite at result level, the two sides level on the night with an unusually low draw share. The conditional-modal scoreline, restricting to Hull-win outcomes, is Hull City 1-0 Middlesbrough (25.4% probability within that restriction). The next most likely Hull wins are 2-1 (8.3%) and 2-0 (6.3%). The Poisson model’s near-identical lambdas reflect the fact that both teams have similar rolling scoring and conceding rates over recent matches, despite their divergent season-long xG profiles. The draw probability of 25.7% sits slightly above the historical H2H draw rate of 16.7% (2 in 12 meetings since 2019) — typical for a single-match Poisson on near-equal lambdas. The conditional-modal scoreline of 1-0 is the most likely Hull win; the overall raw modal across the whole grid is 1-1 at 12.4%, but the conditional-modal rule (intersect the Poisson grid with the 1X2 model’s most-likely outcome) is the published BTP scoreline standard — it avoids self-contradiction with the 1X2 bars. The Poisson model does not see lineup changes or recent injury news (such as Hackney’s potential return), so it should be read as a baseline expectation grounded in recent team-level performance, not a final word.

What the Data Disagrees On

The LR model favours Hull, while the season-long xG difference strongly favours Boro (+27.94 vs -18.16). This disagreement reflects different inputs: the LR captures recent form and current lineup quality, while xGD is a season aggregate. Which is more predictive of a single Wembley final? There is no definitive answer. Hull’s ability to overperform xG all season suggests a resilience that the numbers cannot fully capture, but Boro’s underlying dominance indicates they may be the better side over 90 minutes. The LR model’s reliance on recent form may overstate Hull’s momentum from their semi-final win, while Boro’s 11-day gap means their recent form is based on matches from early May. The Poisson model, which uses rolling rates, sits between the two, essentially calling it even. The truth likely lies somewhere in between: Hull have a tactical edge and momentum, but Boro have superior quality over the season. The model’s conditional-modal scoreline of 1-0 reflects a tight, low-scoring game, which is consistent with both teams’ defensive strengths. The draw probability of around 25% should not be ignored, as it is a significant possibility in a single-match final.

How has the Championship model been doing?

Across the 2025-26 Championship season, the LR calibration’s outcome accuracy reads 234 correct from 544 predictions (43%). Home / draw / away split of where the calls have and haven’t landed:

The Verdict

Models: goals_logreg_v1_cal (Platt-calibrated LR + player-rating features) for the 1X2 call; goals_poisson_v1 (Poisson + Dixon-Coles) for the scoreline grid. Predicted scoreline is the conditional modal — argmax over the Poisson grid restricted to cells consistent with the LR’s most-likely outcome. Probability framing: ≥45% in a three-way market = clear favourite; 35-45% with ≥8 pp gap = narrowly favoured / edge; 33-37% all three = near coin-flip.