Wednesday 1 July 2026 FIFA World Cup 2026 · Round of 32 · Day 4 Bulletin № 021

MM·26 Prediction Agent

The Expected
Points Bulletin

Scoreline picks that maximise contest points — not headlines. [ market 60% · form 25% · xG 15% · ranking 15% · last-match v2 ]

The Ledger

R32 · 28 Jun

40 pts

SA 0–1 CAN · exact ★

R32 · 29 Jun

10 pts

3 games · 2 pen KOs

R32 · 30 Jun

20 pts

FRA & MEX won · NOR upset

R32 · 1 Jul

20 pts

ENG & USA won · BEL 2–2 (90')

760/3240points
thru R32 · day 4

Folio I — FT: England 2–1 DR Congo Mercedes-Benz, Atlanta · 12:00 PM ET

England v DR Congo

FIFA № 4 (~1830 pts) — vs — FIFA № 46 (~1472 pts)

Official Pick

2–0

E[pts] 11.0 · exact 14.0%

England are the heavy favourite (1.29) — group-L winners with the field's meanest defence (0.57 xGA/game) and Kane, Bellingham & Saka up top against DR Congo, a low-block side in their first-ever knockout. Missing Reece James & Quansah at the back but that hardly matters here. A controlled England win — the Under is live given DR Congo sit deep. FT 2–1 — anything but controlled: Cipenga stunned Atlanta on the counter (7') and the low block held until Kane headed the equaliser (75'), then lashed the winner from just inside the box (86'). The banker survived, but DR Congo's goal killed our 2–0 exact — winner only, +10. England meet Mexico in the R16.

Win · Draw · Winblended probabilities

ENG 68%

21%

COD 11%

England windraw (90')DR Congo win

Signal Blend — expected goals

Closing market, de-viggedw 60%2.17 – 0.55

Twelve-month form & stylew 25%1.85 – 1.00

In-tournament xG model 4thw 15%1.65 – 0.67

FIFA ranking (We 80%)w 15%2.05 – 0.65

Last-match vs expectation v2Δ softened-0.02 / -0.02

Final modelρ -0.0752.00 – 0.66

Scoreline Probability Grid — P(exact) %

COD →
ENG ↓

7.70–0

3.90–1

1.50–2

0.30–3

0.10–4

13.41–0

9.91–1

3.01–2

0.71–3

0.11–4

14.02–0

9.22–1

3.02–2

0.72–3

0.12–4

9.43–0

6.13–1

2.03–2

0.43–3

0.13–4

4.74–0

3.14–1

1.04–2

0.24–3

0.04–4

Grid covers 94.7% of outcomes · over 2.5 goals: 49.6% · 90-minute scoreline

Pick Ledger — ranked by expected points

Score	P(exact)	E[points]
2–0◂ PLAY	14.0%	11.05
1–0	13.4%	10.84
3–0	9.4%	9.65
2–1	9.2%	9.60
3–1	6.1%	8.68
4–0	4.7%	8.24

England — form & style

Group L winner, 7 pts (GF6 GA2) — beat Croatia 4–2 & Panama 2–0, drew Ghana 0–0.
Elite defence (0.57 xGA/game) and a deep attack — Kane, Bellingham, Saka, Rashford. Opta has them ~74% here.
Out: Reece James (hamstring), Quansah (ankle); Rice returns. None of it threatens the favourite tag.

DR Congo — form & style

Group K third (best-third), 4 pts (GF4 GA3) — drew Portugal 1–1, beat Uzbekistan 3–1, lost Colombia 0–1.
A deep, low-block side (38% possession) — Wissa carried the attack (3 of 4 group goals).
First-ever WC knockout; needs a near-perfect rearguard and the counters to land to trouble England.

Odds (DK): 1.29 / 5.25 / 13.0 · O2.5 1.91 · U2.5 2.00 xG ENG 2.03 · COD 0.95 · final 2.00–0.66 · ρ -0.07 · FT 2–1 (Cipenga 7', Kane 75' & 86') → pick 2–0, winner +10

Folio II — FT 90': Belgium 2–2 Senegal · BEL 3–2 in ET Lumen Field, Seattle · 4:00 PM ET

Belgium v Senegal

FIFA № 9 (~1742 pts) — vs — FIFA № 15 (~1684 pts)

Official Pick

2–1

E[pts] 7.1 · exact 9.2%

The tightest tie of the day (Belgium 2.10 / Senegal 3.65): De Bruyne, Doku & Lukaku against a Senegal side that creates plenty but is without keeper Édouard Mendy (knee). Both thumped weak MD3 opponents 5–0/5–1. The model reads a narrow, open game — the Over is live and a one-goal Belgium edge the likeliest single scoreline. FT 90' 2–2 — Senegal led 2–0 through Diarra (24') and Sarr (51') until Lukaku (86') and a Tielemans header (89') completed the latest two-goal comeback in World Cup history; Tielemans then won it 3–2 from the spot at 120+5', the latest winning goal ever. Belgium advance to face the USA — but the contest grades the 90-minute result, a draw: our 2–1 scores 0.

Win · Draw · Winblended probabilities

BEL 43%

27%

SEN 30%

Belgium windraw (90')Senegal win

Signal Blend — expected goals

Closing market, de-viggedw 60%1.51 – 1.10

Twelve-month form & stylew 25%1.60 – 1.45

In-tournament xG model 4thw 15%2.25 – 1.40

FIFA ranking (We 56%)w 15%1.56 – 1.31

Last-match vs expectation v2Δ softened+0.00 / +0.09

Final modelρ -0.1271.63 – 1.33

Scoreline Probability Grid — P(exact) %

SEN →
BEL ↓

6.60–0

5.40–1

4.60–2

2.00–3

0.70–4

7.01–0

12.61–1

7.51–2

3.31–3

1.11–4

6.92–0

9.22–1

6.12–2

2.72–3

0.92–4

3.73–0

5.03–1

3.33–2

1.53–3

0.53–4

1.54–0

2.04–1

1.44–2

0.64–3

0.24–4

Grid covers 96.3% of outcomes · over 2.5 goals: 56.9% · 90-minute scoreline

Pick Ledger — ranked by expected points

Score	P(exact)	E[points]
2–1◂ PLAY	9.2%	7.05
1–1	12.6%	6.49
1–0	7.0%	6.40
2–0	6.9%	6.36
3–1	5.0%	5.80
3–0	3.7%	5.42

Belgium — form & style

Group G winner, 5 pts (GF6 GA2) — drew Egypt 1–1 & Iran 0–0, hammered New Zealand 5–1.
De Bruyne, Doku, Lukaku, Trossard — quality throughout, but two draws show they can be held.
Debast a minor doubt; otherwise full-strength and the nominal favourite.

Senegal — form & style

Group I third (best-third), 3 pts (GF8 GA6) — lost France 1–3 & Norway 2–3, beat Iraq 5–0.
Sarr & the front line create big chances (28 shots v Iraq), but the defence leaks (6 conceded).
Keeper Mendy out (knee) — Diaw deputises, a real downgrade against Belgium's attack.

Odds (DK): 2.10 / 3.30 / 3.65 · O2.5 1.95 · U2.5 1.83 xG BEL 2.27 · SEN 1.83 · final 1.63–1.33 · ρ -0.13 · FT 90' 2–2 (Diarra 24', Sarr 51', Lukaku 86', Tielemans 89'; BEL 3–2 AET) → pick 2–1, 0 pts

Folio III — FT: USA 2–0 Bosnia-Herzegovina Levi's, SF Bay · 8:00 PM ET

USA v Bosnia-Herzegovina

FIFA № 17 (~1671 pts) — vs — FIFA № 64 (~1387 pts)

Official Pick

2–1

E[pts] 8.6 · exact 9.9%

The co-hosts are strong favourites (1.40) with the crowd behind them — Pulisic (fit), Balogun & McKennie against a Bosnia side riding Džeko but posting the R32's lowest xG (0.62/game). The USA back line leaked in a 2–3 loss to Türkiye, so this isn't a shutout lock — a two-goal USA win with one at the other end is very much in range. FT 2–0 — Balogun struck on the stroke of half-time (45', his 3rd of the tournament) but was sent off for serious foul play (64'); ten-man USA held and Tillman buried a 20-yard free kick (82'). The clean sheet we didn't trust cost the exact — winner, +10. The USA meet Belgium in the R16, without Balogun.

Win · Draw · Winblended probabilities

USA 56%

24%

BIH 19%

USA windraw (90')Bosnia-Herzegovina win

Signal Blend — expected goals

Closing market, de-viggedw 60%2.05 – 0.70

Twelve-month form & stylew 25%1.55 – 1.05

In-tournament xG model 4thw 15%1.91 – 0.93

FIFA ranking (We 75%)w 15%1.93 – 0.79

Last-match vs expectation v2Δ softened-0.02 / +0.22

Final modelρ -0.0901.89 – 1.03

Scoreline Probability Grid — P(exact) %

BIH →
USA ↓

6.30–0

4.60–1

2.90–2

1.00–3

0.30–4

9.21–0

11.51–1

5.41–2

1.91–3

0.51–4

9.62–0

9.92–1

5.12–2

1.82–3

0.52–4

6.03–0

6.23–1

3.23–2

1.13–3

0.33–4

2.84–0

2.94–1

1.54–2

0.54–3

0.14–4

Grid covers 95.3% of outcomes · over 2.5 goals: 55.9% · 90-minute scoreline

Pick Ledger — ranked by expected points

Score	P(exact)	E[points]
2–1◂ PLAY	9.9%	8.61
2–0	9.6%	8.51
1–0	9.2%	8.40
3–1	6.2%	7.51
3–0	6.0%	7.45
3–2	3.2%	6.60

USA — form & style

Group D winner, 6 pts (GF8 GA4) — beat Paraguay 4–1 & Australia 2–0, lost Türkiye 2–3.
Pulisic (fit again), Balogun, McKennie, Adams — the crowd and quality edge, but the defence can be got at (GA4).
No suspensions; Richards rested vs Türkiye and available. Opta ~68% USA.

Bosnia-Herzegovina — form & style

Group B third (best-third), 4 pts (GF5 GA6) — drew Canada 1–1, lost Switzerland 1–4, beat Qatar 3–1.
Džeko (40) leads the line, but the tournament's lowest xG (0.62/game) — clinical over-performance beat Qatar.
First-ever WC knockout; a leaky defence (6 conceded) against the host crowd is the concern.

Odds (DK): 1.40 / 4.50 / 8.75 · O2.5 1.77 · U2.5 1.90 xG USA 1.54 · BIH 0.62 · final 1.89–1.03 · ρ -0.09 · FT 2–0 (Balogun 45', Tillman 82'; Balogun red 64') → pick 2–1, winner +10

Appendix A

How the picks are made

STEP 01

Blend four signals

Start from the de-vigged closing market (60%), then blend twelve-month form & style (25%), an in-tournament xG model (15%) rating each team on the expected goals it created & conceded across the group stage, and the FIFA ranking's Elo expectancy (15%). The xG signal is shrunk to the league mean so it nudges rather than dictates.

STEP 02

Last-match read, softened v2

Each team's xG is nudged by how its most recent game beat/missed expectation; the attack term (finishing) is halved because it regresses. Note: some sides thrashed weak final-group opponents (Belgium 5–1, Senegal 5–0), which flatters that recency read.

STEP 03

Fit Dixon-Coles & maximise points

A Poisson model with the Dixon-Coles low-score correction ρ gives a probability for every scoreline. Each is ranked by 30·P(exact) + 10·P(1X2) — the contest's scoring, which stack (an exact hit is worth 40). Knockouts are scored on the 90-minute result, so the model still picks within 90 minutes.

Round of 32, Day 4 — graded: 20 points from two winners. Both favourites advanced but neither kept the scoreline we wrote: England 2–1 DR Congo (Cipenga's 7' shock, Kane 75' & 86') — the comeback rescued the result, not our 2–0 exact, +10. USA 2–0 Bosnia (Balogun 45' then red-carded 64', Tillman 82' free kick) — the ten-man clean sheet beat our 2–1 by one goal, +10. And the coin-flip warning landed exactly as written: Belgium 2–2 Senegal at 90' — from 2–0 down (Diarra 24', Sarr 51') Lukaku (86') and Tielemans (89') forced extra time, where Tielemans' 120+5' penalty (the latest winner in World Cup history) sent Belgium through 3–2. The contest grades the 90-minute result: a draw, our 2–1 scores 0. Day total 20/120; tournament 760/3240. Scoring: 30 exact · 10 winner (max 40).