The real BARMM election starts after the count

By Francis Allan L. Angelo

The machines were never the problem. Voters across the Bangsamoro Autonomous Region in Muslim Mindanao go to the polls on Monday, September 14, marking the region’s first regular parliamentary election. Yet the two most serious independent assessments of the vote barely mentioned the counting system. Instead, they focused on the conditions surrounding it and reached the same conclusion through completely different routes.

The Asian Network for Free Elections and the National Democratic Institute put a technical mission in the field in July. They ran 105 stakeholder consultations across the region and published their findings in August. The National Citizens’ Movement for Free Elections worked the other end by synthesizing official data, agency statements, and open-source reporting into a pre-election assessment released on September 7.

One mission talked to 221 people on the ground. The other analyzed what institutions were saying. Neither found much wrong with the hardware.

A QUIET COERCION

The numbers check out. Over 2.3 million ballots were printed for 2.39 million registered voters across 5,212 precincts, with all materials reaching provincial hubs by August 30. Officials walked the public through Data Center 3 in June to demonstrate redundancy and ran mock elections with live voters. Key watchdogs signed the random manual audit agreement by late August.

ANFREL found locals felt positive about the independence and capacity of both COMELEC and the Bangsamoro Electoral Office. COMELEC Chairman George Erwin M. Garcia even shifted the focus away from technology to point out external factors, chiefly vote-buying and campaign conduct. He was surprisingly candid.

But ANFREL delivered a sharp warning. Before anyone declares Monday night a success, remember their conclusion: the number of reported security incidents is an incomplete measure of electoral security. In areas where political authority, clan structures, and armed influence overlap, coercion shapes voter behavior without generating a single reportable incident. An election can be quiet simply because nobody dares make noise.

THE INCIDENT GAP

That observation collides directly with the official record. A joint COMELEC and police report for July to September recorded exactly one validated election-related incident alongside 32,172 checkpoints, 20 gun ban violations, and 34 confiscated firearms.

Meanwhile, Climate and Conflict Action Asia tracked 611 incidents and 235 deaths across Mindanao from January to July 2026. They linked 28 directly to the election. Violence is trending up, matching the spikes seen in 2023 and 2025. The official tally excluded the August 15 ambush on the convoy of interim Chief Minister Abdulraof “Sammy Gambar” Macacua, which happened hours after he took his oath as chairman of the Bangsamoro Federalist Party. COMELEC could not confirm if the ambush was election-related, and NAMFREL blamed a definitional gap rather than bad data. Expect this exact pattern to spark post-election arguments by Tuesday morning.

The physical security preparations are massive. Police deployed roughly 7,000 officers within the region, placing 4,300 at voting centers. Outside reinforcements pushed that number above 8,300. In the final week, NAMFREL tracked the total climbing past 9,400 as army infantry stepped in. The state prepared contingencies including 2,000 Quick Reaction Force personnel on standby and 3,000 police trained to act as Special Electoral Board members.

COMELEC, police, and the military jointly flagged 108 areas of concern. Lanao del Sur holds 21 Red-category areas, the highest in the region, while Maguindanao del Sur carries 11 Orange zones. NAMFREL pulled a different count from a later COMELEC briefing, recording 92 towns and 42 barangays. The numbers conflict because they come from different dates.

The danger lies in who decides these classifications. ANFREL sources objected to partisan officials helping designate these zones. Reclassifying an area often forces voting to move to municipal centers where controlled voting is much easier to run. Others fear the tags will justify last-minute precinct transfers or delays. A security measure and a suppression tactic often wear the same uniform.

A RUPTURE INSIDE THE REBELLION

Underneath this election sits the unfinished half of the peace process. The decommissioning of 14,000 MILF combatants is suspended, and private armed groups remain intact. A May 2026 monitoring report found the normalization track still stuck in its early stages twelve years after the Comprehensive Agreement on the Bangsamoro was signed. Twelve years is a standstill, not a transition.

ANFREL sources blamed the usual drivers: rido, political rivalries, private armies, and violent extremists. Above all, they pointed to an environment flooded with loose, unregistered firearms. Several commanders with armed followers align directly with competing political camps, and some are running for office themselves.

National coverage misreads the political fracture. This is not the government against rebels. It is a rupture inside the Moro Islamic Liberation Front. MILF chairman Ahod “Al Haj Murad” Ebrahim leads the United Bangsamoro Justice Party. Facing him is Macacua, his former military chief of staff who became interim chief minister in March 2025 and now chairs the rival BFP. Multiple sources told the mission they see the BFP tying itself closely to the Marcos administration and its Mindanao allies, a connection that brings better access to political and logistical resources.

NAMFREL flagged national patronage interference as a critical threat and gave it a maximum 16 out of 16 risk score. When both camps repeat unproven charges of Malacañang interference, the loser gets a ready-made excuse for defeat. Palakasan is woven into Philippine politics, and so is the habit of complaining about it only after the votes are counted.

THE PRICE OF A VOTE

On money, both reports pull no punches. Sources described vote-buying as deep and widespread. Campaigns use cash, goods, and coupons redeemable for PHP 2,000 to PHP 3,000, using specific identifiers to track who gets paid. Locals also flagged a recent presidential visit to BARMM that came with announcements of roughly PHP 3.4 billion in projects. While the mission clarified this is not an outright abuse of state resources, the timing and optics clearly favored the incumbent regional administration.

Despite this, zero formal vote-buying complaints exist for this cycle. Every province has a task force, and the state partnered with GCash and Maya to flag suspicious transfers. Still nothing. Commissioner Ernesto Maceda Jr. admitted only two vote-buying incidents were formally recorded during the entire 2025 midterm cycle, blaming normalized underreporting. Everyone knows what is happening, but no one files a case.

Then comes a deeper problem: voter comprehension. ANFREL found severe public confusion over the parliamentary system itself. Voters do not understand the three votes, proportional representation, or sectoral seats, and many do not know how the chief minister is selected. Observers estimate a large majority of voters still lack a basic grasp of the system.

THE REAL DECISION

The government amended the legal framework as late as February 2026, lowering the bar for party representation and widening the field. This happened after the Supreme Court excluded Sulu, after the annulment of the October 2025 polls, after courts struck down two districting laws, and after four separate postponements.

The women’s quota tells a similar story. Every party met the rule requiring 30 percent female nominees, then buried those women at the bottom of the list where they cannot win. In district races without a quota, only 14 out of 107 candidates are women. Parties comply with the letter of the law while completely ignoring its purpose.

Infrastructure adds another quiet threat. Forced outages at major baseload plants pulled more than 840 megawatts from the Mindanao grid in early September. NAMFREL calculates that just one more forced outage this weekend will trigger rotational brownouts in mainland BARMM right between 5 p.m. and 10 p.m. Those are the exact hours when polls close and initial results move. A delay caused by a brownout looks exactly like a delay caused by tampering. From outside a canvassing center, you cannot tell the difference.

COMELEC expects a 75 percent turnout. One party told observers their supporters would vote even if they could die because they believe in the Bangsamoro. Contrast that with the advice officials keep giving voters in contested towns: cast the ballot and go straight home. Both realities exist at once.

Here is what most coverage will miss on Monday night. The voters do not choose the chief minister. The 80-member Parliament does, and that decision happens during a transition window running up to October 30. Sixteen parties are fighting over a fragmented map. The core rebel movement is split into two camps, and the national ruling party clearly has a favorite. Add it up, and this contest will be settled in rooms, not precincts.

The ballots will be counted correctly. The machines will do exactly what they were built to do. Yet none of that answers the only question that matters. Can a people asked to build a government out of a ceasefire be trusted with the result?