In the first quarter of 2026 the International Betting Integrity Association recorded seventy suspicious betting alerts worldwide. Six of them related to African sporting events. Across the five years from 2021 to 2025 the continental total was 117, an average of roughly twenty-three a year.
Those numbers are worth stating carefully, because they are routinely asked to carry more than they can. An IBIA alert is a referral generated when betting on an event moves in a way that a member operator judges unusual. IBIA’s network covers more than ninety operators and two hundred brands, monitoring in excess of US$300bn of turnover a year. That is substantial coverage and it is not complete coverage. Alerts describe the markets IBIA members book and report on. They are not an estimate of manipulation, and they were never designed to be one.
Set against them, GCI’s Winning Ground report estimates unregulated online gambling revenue across Africa at US$17.8bn for 2025, against US$5.2bn regulated, with more than four thousand unlicensed operators targeting the continent. Those are GCI’s figures on GCI’s methodology, and the unregulated component is by its nature modelled rather than reported.
Even allowing generously for both, the two sets of numbers do not describe the same market. Twenty-three referrals a year, across fifty-four countries, tells you about the reach of the reporting network. It tells you almost nothing about what is happening inside African competitions.
Alerts measure money. They do not measure people.
Suspicious betting alerts are generated by operator-side market monitoring. They fire when money behaves oddly: when liquidity appears where it should not, when a price moves against the run of information, when patterns across books converge. That work is valuable, it is well run, and none of what follows is a criticism of it.
What it does not do is measure a separate and more basic condition. Registered players, match officials, coaches and club staff who hold betting accounts and use them.
A participant holding a betting account has not fixed a match, and nothing here should be read as suggesting otherwise. What that account represents is a prohibited or conflicted position under most sporting codes, and a compliance signal available years before any market ever moves. Conventional monitoring cannot produce that signal, because there is no unusual money to detect.
What Turkey actually discovered
On 27 October 2025 the Turkish Football Federation published the results of an audit covering all 571 professional referees in the country.
- 371 of 571 referees, 65 per cent, held registered betting accounts.
- 152 had actively placed bets on football.
- One had placed 18,227 bets. Ten had placed more than ten thousand each.
- 152 were banned. 144 had their refereeing rights terminated.
- On 10 November, 1,024 footballers were referred to disciplinary proceedings. 638 of them were third tier or below.
The reporting that followed treated these numbers as a verdict on Turkish football. That reading is almost certainly wrong, and the error matters, because it allows every other federation to conclude that the problem belongs to somebody else.
How the numbers were produced matters more than the numbers. No investigation ran. No market alert fired. Turkish legal betting operates through a system in which accounts are tied to a national identity number, and the federation holds a list of 571 licensed referees. Comparing the two is a database join. It takes an afternoon.
Why the question is hard to ask in Africa
Most regulated African markets hold two relevant datasets.
- Federations hold registers of licensed players, match officials, coaches and club personnel.
- Licensed betting operators hold customer identity records produced under the anti-money-laundering and know-your-customer obligations their regulators impose.
Neither is perfect. Verification standards vary by market, by operator, by product and by payment rail, and are often stronger at withdrawal than at registration. Federation records go out of date. Names appear in several forms across languages and transliterations. Some jurisdictions have no universal identity number to match on at all. Any honest proposal has to start by admitting that matching will be partial.
The binding constraint is the absence of any instrument compelling a comparison. Licensing is national or, in South Africa, provincial. Operators are numerous. No consolidated account register exists anywhere on the continent, and building one is not necessary for this purpose, whatever the arguments for it elsewhere.
The result is that the prevalence of participant betting in African sport is unmeasured. That is a different statement from saying it is low. One is a finding. The other is the absence of one.
The tooling is not the missing part
It would be convenient to conclude that somebody should build a system for this. Somebody did, three years ago.
Screening a sportsbook’s account base against a register of restricted individuals is an established product category, in commercial use in the United States since 2023, where college sport uses it to screen designated officials. Identity fields are matched as cryptographic hashes, so neither party hands over a readable list, though identifying an individual is necessarily part of resolving a match. The category remains thin, IC360’s ProhiBet being the principal instance of it.
A licence condition should specify the capability and the outcome, never the supplier. Any provider able to meet a published technical standard should be able to serve it, and a regulator that names a vendor has created a monopoly it will later have to defend. A thin supplier market is an argument for writing the standard carefully and letting the market answer it. It is also an opening for anyone minded to compete.
What does not transfer automatically is the surrounding architecture, and that is where the work sits.
- The obligation. In the United States the pressure came through state regulators and league rules. In an African market it would have to arrive as a condition of operating licence, because nothing else reaches an operator directly.
- The register. A federation has to hold a list of restricted participants that is current, complete and structured on identity fields the jurisdiction can actually support. Several cannot support a national identity number at all, and matching will be partial wherever that is true.
- The rulebook. The screen should return what is prohibited for that role. For many match officials that is betting on any sport, not merely their own.
- The recipient. Matches go to a designated integrity function. Not to commercial teams, not to clubs.
- The legal basis. The regulatory obligation, not participant consent. Consent obtained from someone whose registration depends on giving it is unlikely to be treated as freely given, and a programme resting on it is built on the weakest available foundation. Hashing limits what each side handles, and it is a data-minimisation control rather than anonymisation. A match identifies a named person and the output is personal data, governed accordingly.
- The process. A match opens a confidential integrity review. It is not a finding. A disciplinary matter follows only once identity, account ownership, the applicable rule and the apparent breach have been established, with a route for identity disputes and false positives.
Around that sits the ordinary furniture a regulator will expect: a written data-sharing framework, purpose limitation to integrity and rule compliance, defined retention and deletion, access controls and audit logging, independent oversight, and an impact assessment where the jurisdiction requires one. None of it is novel. All of it has to be written down before the first cycle runs.
Ghana is about to improve its numbers
On 20 August 2026, at its 32nd Ordinary Congress, the Ghana Football Association restated that no betting company is licensed to offer odds on Ghanaian domestic competitions. Ghana is further ahead than most of the continent in treating manipulation as a live problem. This particular instrument will make it worse.
In the first quarter of 2026, Ghana accounted for two of the six African alerts, both in football. Those alerts exist because licensed operators booked Ghanaian football and reported what they saw. Withdraw licensed odds from domestic competitions and the reporting that produced them goes with the markets.
What follows is inference rather than established fact and should be read as such. Some demand will fall away, because friction and advertising both matter. The rest will find books outside Ghanaian authorisation, and what happens to visibility then depends on which books those are.
The distinction that matters is whether an operator sits inside a reporting network, and a domestic licence is a poor proxy for it. An operator licensed in a European jurisdiction while serving a market that prohibits it will often belong to one and will file what it finds. An operator registered somewhere that asks no questions will not, and no African regulator has purchase on it. Displacement therefore produces a range of outcomes rather than one, and the further down that range the activity travels, the less anybody sees. Ghana’s alert count would become a worse indicator of risk than it is today, and it would look better.
It is worth being precise about what the announcement is. That is an association position concerning the use of its competitions and its data. It is not a licensing condition imposed by the Gaming Commission and it is not a statutory prohibition. Those three instruments impose very different duties, and only the second reaches an operator through its licence. A federation acting alone can announce a policy it has no power to enforce against the operators that matter.
The origin of the position deserves a fair hearing. The Association’s 2024 complaint to the Gaming Commission concerned the unauthorised use of its match data by named operators, which is a legitimate commercial rights argument on its own terms. The difficulty is that a data-rights position is now doing service as integrity policy, and in that role it is self-defeating. A second rationale is sometimes advanced for restricting domestic markets, that lower liquidity leaves less to be made from manipulation. That has limited force where offshore and Asian-facing operators continue to price lower-tier African football beyond the reach of Ghanaian licensing policy.
Suppressing licensed supply without building detection removes the regulator’s remaining visibility and leaves the conduct untouched. Ghana has the harder half of this problem in hand already, which is the willingness to act. What it needs is an instrument that reaches operators through their licence, and that is a conversation between the Association and the Commission rather than either acting alone.
The window that Turkey never had
Turkey sanctioned more than 1,150 individuals in four tranches across three weeks. Once the audit became public, the federation faced substantial pressure to act quickly and at scale. The audit had followed a criminal investigation into a single second-tier match. The federation was managing a crisis rather than running a programme.
A federation acting deliberately has an option Turkey did not. Announce the screening programme with a defined closure window of sixty to ninety days, during which any restricted participant may close a betting account and self-declare without sanction. Screening begins when the window closes. The window should confer no immunity for manipulation, the misuse of inside information, betting on one’s own event, proxy betting, coercion or any other evidence of corrupt conduct.
A screening programme will surface conduct ranging from technical rule breaches to matters requiring serious integrity investigation, and an account match is not by itself evidence of manipulation. A disclosure window matters most where lower-tier participants have had inconsistent guidance on what their role actually prohibits. It converts a purge into a compliance programme, it is easier to defend publicly, and it concentrates disciplinary resource on non-disclosure and on the conduct that warrants it.
Start with the officials
Begin with match officials alone. It is the smallest register, it carries the greatest integrity leverage, and it is the fastest to run. Turkey’s entire disclosure began with 571 people.
What comes out is a list of the officials on the register who may hold betting accounts. It will be imperfect, bounded by the quality of the matching and the reach of the licensed market. It would still be more than any African jurisdiction knows today. The capability exists. The obligation does not