1. Home
2. Official PAGCOR data, reproduced with its own dates

Official PAGCOR data, reproduced with its own dates

By Eric, Founder, RedClaw · Published 2026-08-31 · Last updated 2026-09-07

PAGCOR publishes quarterly industry figures reaching back to 2014 and quarterly player-exclusion figures reaching back to 2022. Both series are released as PDFs, both carry as-of dates set by the regulator rather than by whoever downloads them, and neither is widely reproduced anywhere in readable form. This section republishes them as tables, keeps the original files, and shows what changed between releases.

Two quarterly series sit on PAGCOR’s site. Both public. Both in PDF, and both largely unread outside the regulator and a handful of analysts who already know where to look. They are the most useful material the agency publishes and by some distance the least reproduced anywhere else.

What series does PAGCOR actually publish?

Two of consequence. Philippine Gaming Industry Data, quarterly, reaching back to calendar 2014. Data on Player Exclusion, quarterly, reaching back to the second quarter of 2022.

Those files are not linked from PAGCOR’s licensees and foundations page, which is where a reader would reasonably look for them. We retrieved that page on 7 September 2026 and it carries fifty-one PDF links, none of which is a licensee register. The registers are served from three separate directories and have to be addressed directly: the venue register, the accredited affiliates register under an “App Kits” path, and the internet gaming licensees register under “offshore”. All three returned HTTP 200 on 7 September 2026. We record the direct paths because a register nobody can navigate to is, for verification purposes, close to unpublished.

What do the two most recent quarters actually show?

Total gross gaming revenue of ₱88.14 billion in the second quarter of 2026, against ₱87.60 billion in the first. Growth of 0.61 per cent across the whole regulated market.

Figures parsed from Industry-Data-1Q-2026.pdf and Industry-Data-2Q-2026.pdf. Percentage changes are ours.

The Electronic Games line is the one to watch, and the arithmetic around it is more interesting than the raw figure. Electronic Games is the online segment, and at ₱36.25 billion it accounts for 41.13 per cent of all Philippine gaming revenue in the second quarter, down from 41.47 per cent in the first. It fell 0.22 per cent while the total market grew 0.61 per cent. Set aside PAGCOR’s own directly operated venues, which are a rounding error at 3.3 per cent of the market, and Electronic Games is the only major segment that shrank in a growing quarter. Land-based casinos in Entertainment City grew 3.82 per cent over the same period.

The obvious reading of that divergence is wrong, and checking the dates is what shows it. The central bank order removing e-wallet links to gambling sites was issued on 14 August 2025, not during this window — a full year before the quarters in the table. Its impact was severe and it was recorded at the time: PAGCOR’s own monthly income fell from ₱5.7 billion in May 2025 to ₱2.9 billion in September 2025, a 49 per cent decline, according to testimony its assistant vice-president Jessa Mariz Fernandez gave to a House committee on 22 October 2025 and reported by Yogonet. So the 2026 figures do not show the order biting. They show the segment having stopped falling, roughly flat at a level reached after the shock, while PAGCOR lobbies the central bank to reverse the measure. That is a materially different conclusion from the one the quarter-on-quarter number invites, and it is only visible if you check when the policy actually landed.

We published the wrong version of that paragraph first, which is worth recording rather than quietly fixing. The original draft asserted that the wallet order “bit during this window” and treated the −0.22 per cent as evidence of it. The claim came from a summary that named the order without dating it, and it survived because it fitted the argument being made. Verifying the primary reporting moved the order back a year and inverted the conclusion. A causal claim that rests on two figures being contemporaneous should never be published before the dates are checked against a source that states them, and an analysis that happens to support the writer’s existing thesis deserves more scrutiny than one that does not, not less.

Why is the annual file missing from that table?

Because our parse of it failed an arithmetic check, so it is excluded rather than published. The calendar-year 2025 file is stored but not used.

The failure is instructive and it very nearly got through. Every line in the annual file parsed to a plausible peso figure, no exception was raised, and nothing looked wrong on inspection. Summing the components against the printed section totals is what caught it: PAGCOR Operated came to ₱12.52 billion against a printed total read as ₱2.52 billion, and Other Licensees to ₱101.12 billion against ₱1.12 billion. The leading digit of each total was being lost, because the annual file marks totals with footnote superscripts and the extraction was splitting the number at that point. A parser that only checked for well-formed numbers would have published figures off by an order of magnitude. Our loader now rejects any period whose components do not reconcile with its stated totals, and the quarterly files reconcile to the peso.

Why does the exclusion series matter more than it looks?

Player exclusion counts how many people have formally removed themselves from Philippine gaming. It is the regulator’s own measure of harm, published on the regulator’s own schedule.

Read that series next to the advertising restrictions and a pattern emerges that neither document states on its own. PAGCOR spent 2026 tightening advertising rules, strengthening its responsible-gaming code, and prohibiting the 17:30 to 20:30 window. Across the same period it continued publishing quarterly exclusion figures. A regulator that publishes harm data on a fixed cadence while narrowing promotional freedom is assembling a public record that supports further narrowing. Anyone forecasting whether the full television ban arrives should weight that record above the press commentary, which is what the advertising rules section does.

What does the “as of” date on these files actually mean?

It is PAGCOR’s date, not the download date. The venue register we parse is dated 31 August 2026 even though it was retrieved on 7 September.

That distinction sounds pedantic until a licence check turns on it. A file retrieved today can describe a position several weeks old, and an operator whose status changed in the interval will not appear to have changed. Every figure on this site carries the regulator’s date rather than ours for that reason.

Extracting these files is harder than it should be, which is part of why so few sites reproduce them. PAGCOR overlays a light-blue watermark on the register PDFs, and the watermark characters interleave with the body text at the character level, so a naive text extraction returns strings like “GamingO Corp.” instead of “Gaming Corp.” The venue register additionally uses a multi-column layout with a caution notice down the left margin, which a linear text extractor splices into the operator names. Our parser filters the watermark by fill colour and crops the notice column by x-coordinate. Both fixes, and the reason a page-count check is not a sufficient test of a parser, are written up in the parsing notes.

How is the archive kept?

Every source PDF is stored at the version we parsed, alongside the parsed output. Nothing is overwritten in place.

Keeping the originals is what makes the monthly comparison possible. The diff between two months of a licence register is a fact about the market; a single month is only a snapshot. The registers themselves and how to read each one are covered in the PAGCOR licence section.

Why is a diff more useful than a snapshot?

Because a register only becomes evidence when you can see what moved. One month of names tells you the current position. Two months tell you the direction.

The first comparison is now available and it makes the point better than the argument for it did. We parsed the venue register at its 31 July 2026 revision and again at its 31 August 2026 revision. Both parse to 258 entries. The count is identical, and the membership is not.

One operator in, one operator out, and a total that does not move. Anybody monitoring this register by its headline number recorded nothing happening in August. The alphabetical ordering confirms the change is real rather than an artefact of our parsing: Play Park Entertainment Inc. appears at position 188, correctly sorted between PJ E-Games OPC and Playmate Leisure Solutions Corp., and every entry after the removed one shifts up exactly one position. A parse error does not reorder a list consistently.

We are deliberately not characterising why Pokerstars Club Inc. is absent. Removal from this register is consistent with a licence lapsing, with a licence being surrendered, with a licence being revoked, and with a corporate name change we have not matched. The register records the fact of absence and nothing about its cause, and the difference between those four is exactly the sort of thing that gets invented by sites reporting register movements as enforcement news.

The affiliates register moved its own as-of date from 12 August to 20 August 2026 with no change to its 54 entries at all. That is worth recording too: a new revision date is not evidence of a change, and treating a re-dated file as a market event is the mirror image of the error above.

Which PAGCOR register covers internet gaming, and what state is it in?

One exists. It is offshore-only, it was last revised on 7 October 2024, and it is published as a scan that no software can read.

The file is titled List of Internet Gaming Licensees and carries the sub-heading Offshore Gaming Licensing Department. Unlike every other register PAGCOR publishes, it contains no text at all: two pages, each a single full-page image of 3,507 by 2,480 pixels, with zero extractable characters. A text extractor returns an empty string, which is why it is absent from every dataset we have seen reproduce the other registers, and why our own parser skipped it for a fortnight before we transcribed it by reading the images. It lists thirty-four regular licensees and four provisional ones, thirty-eight in total, with a single asterisk against Gambol Leisure and Amusement, Inc. footnoted “with Approved Suspension of License”.

The revision date is the part that matters, and it needs a second date to mean anything. The file’s own footer reads “REVISED AS OCTOBER 7, 2024”. Executive Order No. 74, section 2(4), states that “All POGOs/IGLs and other offshore gaming operations and other offshore gaming-related/auxiliary/ancillary services with issued licenses, permits or authorizations shall completely cease operations, including the winding up of their affairs, on 31 December 2024 or earlier.” The order defines Internet Gaming Licensees as business-to-consumer licensees catering exclusively to offshore-based or foreign players — the exact category this register enumerates. So the register PAGCOR currently serves is a snapshot taken before the order, of a licence class the order abolished with effect from 31 December 2024.

We retrieved the file on 24 August 2026 and again on 7 September 2026. Both copies are 1,359,229 bytes with MD5 9496b49c76eb59c3e5bc47bca82c4b97. It is byte-identical across the fortnight and has not been revised in the twenty-three months since its printed date.

There are two readings and we are not in a position to choose between them. It may be retained deliberately as a historical record of who held these licences when they were cancelled, which is a legitimate reason to leave a file up. It may equally be an artefact nobody removed. What is not in question is the practical effect on anyone trying to check an operator: the one PAGCOR register whose title contains the words “internet gaming” describes offshore business-to-consumer licensing as it stood in October 2024, answers nothing about any site serving Philippine players today, and cannot be searched by machine. Which list does cover domestic online operators is taken up in the PAGCOR licence section.

What can these figures not be used for?

Forecasting operator revenue, or estimating any individual company’s performance. The industry series is aggregated by segment, not by licensee.

That limitation matters because the temptation runs the other way. A quarterly revenue series covering more than a decade looks like something you can model. It is, at the market level. At the operator level it says nothing, and a per-company number derived from it would be an invention presented in the regulator’s units. The same applies to the exclusion series: it counts exclusions across the regulated market and cannot be attributed to any single brand.

How should someone cite these figures?

Cite PAGCOR, and cite it with the as-of date printed on the file. Our tables are a convenience layer over the regulator’s documents and should never be the attributed source for a regulator’s number.

The tables here carry a Creative Commons attribution licence because the parsing work and the layout are ours. The figures inside them are not. Anyone quoting a count should name PAGCOR and the file’s own date, and can link here for the readable version. Journalists in particular should retrieve the source PDF from pagcor.ph rather than relying on any secondary reproduction, including this one.

There is a practical reason to insist on this beyond academic tidiness. Philippine media platforms and payment partners ask operators for evidence of licensing, and they will not accept a third-party website as that evidence. We have had this conversation on the operator side more than once. A compliance reviewer wants the regulator’s own document, dated, with the entity name matching the contracting party. Anyone assembling that pack should treat this site as an index that points at the right file, never as the file itself.

What we will not publish here

Derived rankings, projections presented as data, or any figure without its source date attached.

The temptation with a series running to 2026 is to extrapolate it. We do not, because a forecast rendered in the same table as regulator figures reads as though the regulator produced it. Analysis that draws on these numbers is written as analysis and labelled as ours, which is the standard set out on the author page.

Where do these files live?

On PAGCOR’s site, published without announcement and replaced in place. The quarterly series are linked from the licensees and foundations page. The three licence registers, as of our 7 September 2026 retrieval, are not linked from it at all and are reachable only by direct path.

Both series are parsed on the same monthly cycle as the registers, and every retrieved file is kept at the revision we read rather than overwritten in place. That last part is not housekeeping. PAGCOR replaces these files at the same address, so a copy not taken at the time cannot be recovered afterwards.

What do these figures not capture?

Regulatory intent. The filings record what happened; they say nothing about what the regulator is trying to do next, and that gap has to be filled from elsewhere.

PAGCOR is preparing a position paper asking the central bank to reverse the e-wallet delinking order, and has said publicly that it would prefer a complete broadcast advertising ban to the current prime-time window — both stated at the Senate hearing of 11 February 2026 and reported by Inside Asian Gaming and AGB. Neither position appears in any quarterly filing, and both would move these numbers materially if acted on.

That is why this section reads the filings against the hearing record rather than on their own. A revenue series tells you where a segment settled; the hearing tells you which direction the next push comes from and who is pushing. Read together they support a forecast. Read separately, the filings look like a market at equilibrium and the hearings look like noise, and both readings are wrong.

Disclosure. LegitPH is written by Eric, founder of RedClaw, a performance-marketing agency whose clients include iGaming operators. We publish no ratings and take no payment for placement in any list on this site. How we source and check every figure is set out in our methodology.