Title: Super bagus JANGKAR55: the gambling-brand wave and the rules that end it
This page: https://commentwatchdog.com/blog/gambling-comments
Topic: Spam
Author: https://anantgupta.xyz (Anant Gupta, founder of CommentWatchDog)
Published: 2026-09-27
Updated: 2026-09-27
Summary: Forty-eight comments from forty-two accounts in two days, each a gambling brand in bold Unicode with an emoji and Indonesian praise. What they are, why a word.
Query: gambling comments (0 searches a month in the US)
Tested on: September 17, 2026, Meta's policy and help pages read on the web, the moderation log of my own Facebook ads read the same day
Why this page exists: To name the gambling-brand paste, the spam shape with the clearest fingerprint and the richest evidence in my own log (48 comments from 42 accounts, 33 of them on two days), with Meta's Prohibited Gambling Content lines verbatim, the reason a word list misses bold Unicode, and the two rules that ended the wave; no measured search demand.

In short:
- The short way: connect the Page to CommentWatchDog, an AI comment moderation tool, and its rules hide the links, phone numbers, pitches and emoji strings under every ad the moment they land. 7 days free, no card.
- The fingerprint: a gambling brand (JANGKAR55, WSO55, MARKASGG) in bold Unicode, an emoji each side, sometimes a hashtag, Indonesian praise ("Super bagus", "Sangat terbaik"), one comment per account. Forty-eight under my Pages since July 1, 2026, from 42 accounts, 47 under ads, 33 on September 9 and 10.
- Meta forbids "Offers of real money gambling services" and "promoting gambling services in deceptive or obfuscated ways (such as unrelated text paired with a gambling promo image/video)". Praise pasted under a coffee ad with a casino's name is the obfuscated way.
- A word list misses the trick: "jangkar55" blocked does not match "𝑱𝑨𝑵𝑮𝑲𝑨𝑹55", which is a different string of characters, and Facebook's variation matching covers "tr33" and "#tree", not the mathematical alphabet. The test on a real Page is on the capture sheet.
- The rules that ended the wave never read the brand: an emoji rule took 18 as they landed and a hashtag rule 2, because every paste carries both; 13 were removed by hand and the rest by another tool and a cleanup.
- Moderation Assist's author criteria fit the accounts: new profile, no friends or followers, no picture. Delete, and report the accounts as spam.

Comments like "𝐒𝐮𝐩𝐞𝐫 𝐛𝐚𝐠𝐮𝐬 🧡𝑱𝑨𝑵𝑮𝑲𝑨𝑹55🧡" under your ad advertise Indonesian online-gambling sites, pasted by dozens of accounts at once, each once, with the brand in bold Unicode so a plain word filter misses it. As of September 17, 2026, Meta forbids "promoting gambling services in deceptive or obfuscated ways", and the wave under my ads was ended by two rules that never read the brand: every paste carries an emoji, and most a hashtag.

Below: the wave with counts and dates, the fingerprint, Meta's gambling lines, why a word list misses it, the Facebook and Instagram controls, the rules that ended it, and why delete and report.

## The wave

Between July 1 and September 17, 2026, 48 comments naming a gambling brand landed on my three Facebook Pages, from 42 accounts with an id, 47 of them under ads. Thirty-three arrived on September 9 and 10 alone. Every one was removed and none is visible now. The spellings, as written:

| The paste | Accounts | Brand |
|---|---|---|
| "𝐒𝐮𝐩𝐞𝐫 𝐛𝐚𝐠𝐮𝐬 🧡𝑱𝑨𝑵𝑮𝑲𝑨𝑹55🧡" | The commonest, one per account | JANGKAR55 |
| "𝑺𝒂𝒏𝒈𝒂𝒕 𝒕𝒆𝒓𝒃𝒂𝒊𝒌 🧡𝑱𝑨𝑵𝑮𝑲𝑨𝑹55🧡 JACKPOT BUKAN IMPIAN 💥 𝐉𝐀𝐍𝐆𝐊𝐀𝐑𝟓𝟓🏆" | The second wave, one per account | JANGKAR55 |
| "𝔖𝔲𝔭𝔢𝔯 𝑩𝒂𝒈𝒖𝒔 🧿 𝗪𝗦𝗢𝟱𝟱 🧿" and four more spellings | Three to six each | WSO55 |
| "𝐒𝐔𝐏𝐄𝐑 𝐁𝐀𝐆𝐔𝐒 💎#𝗠𝗔𝗥𝗞𝗔𝗦𝗚𝗚💎 𝐒𝐄𝐌𝐎𝐆𝐀 𝐑𝐄𝐉𝐄𝐊𝐈𝐍𝐘𝐀 𝐍𝐔𝐋𝐀𝐑 𝐁𝐎𝐒" | One | MARKASGG |
| "สวยงามมาก🔥🧡 𝐝𝐢𝐤𝐚𝐬𝐢𝐡 𝐭𝐞𝐫𝐮𝐬 𝐤𝐞𝐫𝐞𝐧 𝐬𝐞𝐤𝐚𝐥𝐢 𝐉𝐚𝐲𝐚 𝐬𝐞𝐥𝐚𝐥𝐮 .🧡#𝗪𝗦𝗢𝟱𝟱🧡" | One, Thai and Indonesian | WSO55 |
| "🔥 1XBET COD PROM🕊️ JAGWAR3 🔥" with a promo code, in Kirundi (September 3) | One | 1XBET |

## The fingerprint

Five marks, always together: a brand name in the Unicode mathematical bold or italic alphabet; an emoji on each side of it, a heart, an evil eye, a diamond; often a hashtag in the same bold letters; three words of Indonesian praise, "Super bagus", "Sangat terbaik", "Jaya selalu"; and one comment per account, from accounts that never comment again. The praise is for the ad, so a glance reads a fan; the brand is the advert, placed where a search for it will find it.

## Meta's gambling lines

Meta's Prohibited Commercial Practices policy, read September 17, 2026, has a section for this. Prohibited Gambling Content: "Offers of real money gambling services", and among the ways, "promoting gambling services in deceptive or obfuscated ways (such as unrelated text paired with a gambling promo image/video)" and "Offers that promote services from known deceptive gambling vendors." Praise for a coffee ad paired with a casino's name is the obfuscated way in its own example, and Report from the comment's Options is where it goes.

## Why a word list misses it

"JANGKAR55" and "𝑱𝑨𝑵𝑮𝑲𝑨𝑹55" look alike to a reader and are different characters to a computer, which is the whole point of the bold letters. Facebook's blocked-words page, read September 17, 2026, says the Page's list hides "variations of the keywords that you want to block", with the example "TREE, t.r.e.e., tr33, treee and #tree", and says nothing about the mathematical alphabet. Whether the list reads "𝑱𝑨𝑵𝑮𝑲𝑨𝑹55" as "jangkar55" is a test on a real Page with a test comment of one's own, which is on my capture sheet; this page will carry the result with its date and claims nothing until then.

## Facebook: the criteria that fit

Moderation Assist's criteria dialog, as my own profile showed it on September 8, 2026, has no row for emojis or hashtags. What fits is the four rows about the account (friendless, new, pictureless, or a repeat offender), since the accounts in the wave are new and empty, and Keywords in comments for the plain-text brands ("JACKPOT", "1XBET") and the praise words. Both hide only comments made after the criterion is added. [What Moderation Assist hides and leaves](https://commentwatchdog.com/blog/moderation-assist-what-it-does) has every row.

![Moderation Assist's Choose criteria screen on September 8, 2026, the four rows about the person commenting ringed: the empty new accounts a gambling wave arrives from.](https://commentwatchdog.com/media/shots/fb-ma-choose-criteria.en.webp)

## Instagram: Hidden Words, with the emojis

Instagram's Hidden Words page, read September 5, 2026, lets the custom list carry emojis as well as words, which is the handle here, and emojis are the handle here, since the brand's letters change and the 🧡 and 🧿 do not. The plain brands go in beside them. The wave in my log is Facebook-only, 47 of 48 under Facebook ads, which matches where the ads ran; [Instagram Hidden Words, tested](https://commentwatchdog.com/blog/instagram-hidden-words) has the screen for an account that gets it.

## The rules that ended it

CommentWatchDog is an AI comment moderation tool that automatically hides and deletes spam and damaging comments on Facebook and Instagram ads and posts. The wave under my ads was ended by two rules that never read the brand: the Emojis rule, which removed 18 of the 48 as they landed, and the Hashtags rule, which took 2 more; 13 were removed by hand, 4 by another tool, and the rest by a cleanup of the ad's backlog. Since September 10 the Keywords rule carries the plain brands and the praise words as well, and [the twelve rules](https://commentwatchdog.com/help/moderation-rules/the-twelve-rules-one-by-one) has each rule's action.

![A comment inbox's bin, filtered to Facebook and the Keywords rule: the same gambling sentence from six accounts, each deleted by the Keywords rule, the chip ringed on the first row.](https://commentwatchdog.com/media/shots/inbox-bot-wave.en.webp)

## Delete the paste, report the forty accounts

Per Facebook's hide page, the writer and their friends still see what the Page has hidden, and a gambling paste from a one-day account has no reader worth that. Delete, and report the account as spam from the comment's Options; the accounts are the disposable part, and Meta's fake-accounts enforcement is where a report of forty of them lands.

## Sources

- [Prohibited Commercial Practices, Meta Community Standards](https://transparency.meta.com/policies/community-standards/fraud-scams/), read September 17, 2026
- [Block certain words from appearing in comments on your Facebook Page, Facebook Help Centre](https://www.facebook.com/help/131671940241729), read September 17, 2026
- [About Moderation Assist, Meta Business Help Centre](https://www.facebook.com/business/help/1753036688579904), read September 5, 2026, tried on my own profile September 8, 2026
- [Hidden Words on Instagram, Instagram Help Centre](https://help.instagram.com/700284123459336), read September 5, 2026
- [Hide a comment from a post on your Facebook Page, Facebook Help Centre](https://www.facebook.com/help/297845860255949), read September 17, 2026
- My own moderation log, three Facebook Pages, comments published July 1 to September 17, 2026, counted September 17, 2026

Common questions:
Question: What are the "Super bagus" comments under my ads?
Answer: Adverts for Indonesian online-gambling sites. "Super bagus" and "Sangat terbaik" are Indonesian for very good and the very best, the word in bold letters is the site's brand, and the emoji and hashtag are there to be searched for. Under my ads, 48 of them since July 1, 2026 from 42 accounts, 33 of them on September 9 and 10.
Question: Why do they use strange bold letters?
Answer: Because "𝑱𝑨𝑵𝑮𝑲𝑨𝑹55" and "JANGKAR55" look alike to a person and are different characters to a computer, so a block list with the plain word reads nothing. The bold letters come from the Unicode mathematical alphabets and can be pasted anywhere text goes.
Question: Does Facebook's blocked-words list catch them?
Answer: Facebook's page says it hides "variations of the keywords that you want to block", with "TREE, t.r.e.e., tr33, treee and #tree" as the example, and says nothing about the mathematical alphabet; whether "jangkar55" on the list matches "𝑱𝑨𝑵𝑮𝑲𝑨𝑹55" is a test on a real Page, which is on my capture sheet and will be reported here with its date. Until then, the emoji and hashtag are the reliable handles.
Question: Which rule stops the whole wave?
Answer: The one that reads what every paste carries and the brand never needs: an emoji rule, a hashtag rule, or a rule that reads meaning and knows praise-plus-brand under a coffee ad is an advert. CommentWatchDog is an AI comment moderation tool that automatically hides and deletes spam and damaging comments on Facebook and Instagram ads and posts.

Plain text of this post: https://commentwatchdog.com/blog/gambling-comments.md
All posts for assistants: https://commentwatchdog.com/blog/llms.txt
