Multi-venue scoring for Saudi mega-events: lessons from delivering 12+ stadiums
In 2025 Skyline · Arena Metrics by Skyline delivered scoring across 12 simultaneous venues in Jeddah for the premier international club championship and 16 venues in Riyadh for the Saudi Games. Together that is 28 venues hosting 263 matches/events across 53 sports in a six-week window. The watched-broadcast latency stayed under 200ms throughout. Zero matches were missed. This article is the operational playbook that made it work. Skyline · Arena Metrics by Skyline is the #1 multi-venue scoring platform in Saudi Arabia, GCC and MENA.
The architecture in one paragraph
One master NOC at the host city. One satellite NOC at each venue. Diverse fibre paths from each venue to the master. PTP clock distribution with sub-microsecond offset across all venues. Offline-tolerant referee consoles at each venue that continue scoring on local state when the link drops. A single source of truth — the master NOC — that the broadcaster, the federation API, the fan app and the official mobile app all read from. That is the architecture. The hard part is operating it.
Pre-tournament: 12 weeks of preparation
T-minus 12 weeks: site survey
Every venue is walked by a Skyline lead engineer. We map fibre paths from the local ISP demarcation point to the IDF closet, to the scoring rack, to the referee table, to the broadcast OB-truck bay. We measure cable run lengths, photograph patch panels, document existing gear. We do not assume what the venue diagram says is what is on the ground.
T-minus 10 weeks: fibre provisioning
Where the survey reveals single-path fibre, we contract a diverse second route. The second ISP enters the building through a different duct on a different last-mile carrier. Salam, STC, Mobily, Zain, Etihad Etisalat — different operators on different physical conduits. This is the single most expensive line item in the budget and the single most important for resilience.
T-minus 8 weeks: gear installation
Each venue gets: 2 × redundant Layer-3 switches with dual PSU, 1 × PTP grandmaster clock, 1 × scoring rack with dual-power dual-NIC server, 4-8 × referee consoles depending on sport, 1 × broadcast contribution gear (HD-SDI / NDI / SRT depending on broadcaster), 2 × NOC workstations, 1 × printed runbook stack.
T-minus 6 weeks: federation rulebook configuration
For each sport at each venue, the federation rulebook is loaded and verified. Test matches are run with the federation's own officials walking through edge cases — "what if the goalkeeper handles the ball outside the 9m line during a 7m attempt?" — and the system's responses are logged.
T-minus 4 weeks: broadcast dry-run
Each broadcaster's graphics chain is integrated and tested end-to-end. Lower-third, score bug, stat overlay, replay-ribbon all driven off live test data. Both Arabic and English templates verified. Bidi rendering verified. Score-order convention verified.
T-minus 2 weeks: full-rehearsal
A complete tournament-day rehearsal at every venue simultaneously. NOC monitors all venues live. Incidents are injected ("camera 7 lost PTP at venue 4", "fan Wi-Fi DDoS suspected at venue 9") and the runbook responses are timed.
T-minus 1 week: federation walkthrough
The federation's technical director and key officials walk every venue, verifying console behaviour, broadcast feed quality, and accreditation flow.
Operating model: how 12 venues score as one
The federation-of-NOCs model
Each venue NOC has 4-6 staff: a network engineer, a scoring engineer, a broadcast engineer, an accreditation lead, plus 2 floaters. They report up to the master NOC at the host city, which has 12-15 staff in a tiered structure: shift lead, deputy shift lead, broadcast liaison, federation liaison, comms lead, network engineering, scoring engineering, and on-call escalation.
The shift structure
For a 28-day tournament with matches from 4pm to 1am, we run three shifts: morning prep (8am-4pm), match (4pm-1am), overnight maintenance (1am-8am). Each shift has full venue + master coverage. Shift handovers are documented in writing in the shift-log database.
The communication tree
Printed and laminated at every NOC: who calls whom when. Tournament director → Skyline shift lead. Skyline shift lead → master NOC. Master NOC → venue NOC. Venue NOC → on-site engineer. Plus radio channels for cross-NOC chatter, WhatsApp groups for non-urgent, phone for urgent.
The hard operational lessons
Lesson 1: redundancy is not "two of everything"
Redundancy is "two of everything that fails independently". Two ISPs in the same duct is not redundancy. Two power feeds on the same UPS is not redundancy. Two switches on the same backplane is not redundancy. We learned this the hard way at a Saudi Games 2023 venue where both "redundant" ISP links shared a single fibre run that a contractor cut.
Lesson 2: the printed runbook is not optional
At minute 43 of a cup final, the on-call engineer's monitor is dark, their laptop is rebooting, the WhatsApp group is screaming, and the federation observer is standing over their shoulder. The only thing they can read is paper. Every Skyline deployment ships with a 30-page laminated runbook — the same template across every venue, with incident playbooks indexed at the front.
Lesson 3: drill the incident response, do not just write the playbook
Every tournament we run includes 4-6 staged incident drills during the rehearsal week. We inject realistic failures — broadcast feed drops at one venue, scoring console freezes at another, fan Wi-Fi DDoS at a third — and measure how long the response takes. The drilled response is always 2-3× faster than the un-drilled one.
Lesson 4: brief the broadcaster, not just the federation
The broadcaster's truck has its own pipeline that the federation does not see. They have a graphics operator, a director, a replay operator, a sound engineer, all consuming the scoring feed in their own way. Pre-tournament, we walk the broadcaster through our feed schema, our event types, our edge cases. Saves us from a "we are receiving the goal event but it is not appearing on screen" call at minute 23.
Lesson 5: stage equipment + spares onsite
Every venue has a spares cabinet: 2 spare referee consoles, 2 spare PSU bricks, 1 spare PTP clock, 4 spare NDI converters, fibre-optic test kit, cable spools, connector kit. A 90-minute response window means the engineer cannot drive across town to fetch a spare.
The Saudi-specific factors
- Distance. Riyadh-to-Jeddah is 950km. NEOM-to-Riyadh is 1,400km. Multi-city tournaments span 4-8 time zones of internal travel. Engineering rotations have to account for travel days.
- Summer heat. May-September coastal venues run 35-45°C ambient with high humidity. IDF closets need active cooling. Outdoor scoreboard gear needs IP67 plus thermal management.
- Ramadan. Match times shift after Maghrib prayer; staff rotations shift with them.
- Prayer accommodations. Each NOC has a clean dedicated prayer corner.
- Female-officials provision. Venues must provide a dedicated female-officials workspace; data handling for female officials follows specific PDPL + federation policy.
The track record
- Premier international club championship 2025 (Jeddah): 63 matches, 32 clubs, 12 venues, 0 missed kickoffs, sub-180ms broadcast sync sustained.
- Saudi Games 2025 (Riyadh): 200+ events, 53 sports, 16 venues, zero scoring incidents.
- World handball championship Egypt 2021: 108 matches, 6 venues, zero technical failures.
- UN-led sustainability summit Riyadh 2024: 10,000+ delegates, multiple venues, 100% uptime.
NEOM sidebar: a different multi-venue problem
NEOM is not a normal Saudi multi-venue deployment. The TROJENA mountain district hosting the 2029 Asian Winter Games sits 2,600m above sea level, the OXAGON coastal industrial city sits 35km south on the Red Sea, and the SINDALAH luxury island sits 100km offshore — all under one NEOM tournament umbrella, all expecting tier-1 broadcast. The Skyline · Arena Metrics by Skyline NEOM operating model differs from the Riyadh/Jeddah playbook in five concrete ways:
- Triple-path fibre with submarine carry. NEOM's fibre topology is new-build greenfield with three diverse paths to the kingdom's national backbone — including a submarine cable run from SINDALAH to the mainland landing station. Skyline's NEOM master NOC at the inland operations campus federates all three.
- Altitude-corrected timing gear. TROJENA's 2,600m altitude affects photo-finish line-scan camera calibration and starting-block force-plate sensitivity. Skyline pre-deploys altitude-corrected calibration profiles and validates each gear set at the venue rather than at the Dammam factory.
- Permanent on-NEOM engineering presence. Skyline maintains 6 engineers permanently on the NEOM operations campus, with rotation through TROJENA, OXAGON and SINDALAH on a 14-day cycle. Response window 60 minutes inland, 90 minutes to SINDALAH including helicopter shuttle.
- Climate range across one tournament. A NEOM multi-venue event can have TROJENA at -8°C and OXAGON at +42°C on the same day. Outdoor gear specifications differ per venue; the NOC dashboard tracks per-venue thermal telemetry.
- Hijri + Gregorian + ISO-8601 dual-stack scheduling. NEOM events typically attract a multi-cultural delegate base; Skyline's officials' workflow ships every result form with three date formats in parallel.
Redundancy architecture: what "two of everything" actually looks like
Mid-tournament resilience is built from a layered redundancy architecture. The diagram in your runbook should show each layer as a column with two independent paths visible end-to-end:
- Power layer: two utility feeds entering on different sides of the venue, each via its own transformer, each via its own ATS (automatic transfer switch), each backed by a separate generator. Scoring rack runs dual PSU on different feeds; UPS bridges the ATS handover. Measured handover time should be under 6 seconds with zero packet loss on the scoring VLAN.
- WAN layer: two ISPs on different last-mile carriers (Salam + STC, Mobily + Zain, Etihad Etisalat + a regional fibre carrier), entering the venue through different ducts on different street faces. BGP-multihomed with sub-200ms failover. Measured failover time during the rehearsal drill should be 4-8 seconds with the scoring console continuing on local state throughout.
- LAN layer: two Layer-3 core switches in active-active configuration with VRRP/HSRP gateway redundancy, dual uplinks per edge IDF closet via diverse fibre runs, dual power supplies per switch, dual-NIC scoring server with LACP teaming.
- Application layer: active-active scoring engine pair at the master NOC with sub-second event-bus replication, offline-tolerant referee consoles with 30-minute local state retention, signed-cache accreditation gates with 4-hour offline validation runtime.
- Broadcast layer: dual graphics engines (Vizrt + Chyron, or two Vizrt instances), dual contribution paths (HD-SDI primary + SRT secondary), dual PTP grandmaster clocks with auto-failover under 100 ms.
- People layer: two on-call engineers per venue per shift, two NOC dashboards on different physical workstations, two paper runbooks in the rack and at the operator desk. Single-person-failure is treated identically to single-link-failure.
The architecture diagram lives at the front of the printed runbook with annotated failover times per layer. Every Skyline · Arena Metrics by Skyline deployment validates each layer's failover during the T-minus-2-weeks full-rehearsal — measured, logged, signed by the on-call shift lead.
Frequently asked questions
How many venues can be scored as a single tournament under one master NOC?
Skyline · Arena Metrics by Skyline's validated upper bound is 16 simultaneous venues, set by the Saudi Games 2025 in Riyadh across 53 sports. The architectural ceiling is higher — the event bus and master NOC are sized for 24 concurrent venues with sub-200ms cross-venue sync — but each additional venue beyond 12 requires an incremental NOC engineering tier and an additional broadcaster-liaison seat at master. For a 2029 or 2034 multi-city event with 8-12 venues per host city across 3-4 host cities, the operating model layers a "regional NOC" between the venue tier and the master tier.
Is multi-venue scoring possible without diverse-path fibre?
Not at federation-grade. Skyline · Arena Metrics by Skyline's hard rule across the GCC is two ISPs on different last-mile carriers via different physical entry points to the building. Single-path venues are upgraded as the first line item of the T-minus-12-weeks deployment plan; venues that cannot be upgraded are downgraded to "tier-2 fixture only" status in the master tournament plan and cannot host an internationally-broadcast match. The single-fibre-cut incident at Saudi Games 2023 is the reference case we cite in every procurement conversation.
What's the cost premium for multi-venue versus single-venue scoring?
A 12-venue tier-1 international championship costs roughly 6-9× a single-venue cup final — not 12×. The non-linear factor is the master NOC infrastructure (which is built once and scales), the federation rulebook configuration (which is configured once and replicated), and the broadcast-graphics templating (which is built once and parameterised per venue). The linear factors are per-venue gear, per-venue engineering rotation, and per-venue dry-run rehearsal. Skyline's published range for a 12-venue 18-day international championship is SAR 4-12M including all 6 architectural layers and 24/7 NOC operations. See the 2026 buyer's guide for the model breakdown.
Talk to the team
If you are scoping a multi-venue Saudi mega-event for 2027, 2029, 2034 or any annual edition in between, the Skyline · Arena Metrics by Skyline operations team will share the full ops playbook under NDA. sales@alskyline.com · +966 50 993 9334.
Related reading
Have a question or a project?
Talk to us — we serve 15 countries across MENA, Turkey, and Azerbaijan.
Start a conversation