Perya
The Filipino fairground, priced properly — what every betting stall in a perya actually costs, worked out from the dice and the deck rather than guessed.
The Filipino fairground, priced properly — what every betting stall in a perya actually costs, worked out from the dice and the deck rather than guessed.
A perya is a travelling fairground. It arrives for a town fiesta or a Christmas season, sets up on a plaza or an empty lot for a week or two, and leaves. There are rides, a stage, food stalls, and along one side a row of small booths where you can put money on the outcome of something. Those booths are what this page is about.
The word comes straight from the Spanish feria, meaning a fair or a market. In English a perya is a funfair, a carnival, or a country fair — there is no closer single word. Filipinos also say peryahan for the fair itself and peryante for the people who work it.
What you will not find anywhere is the price. Every betting stall in a perya has odds, and none of them is written on the board. This guide works them out. Where we have already published a full analysis of a game, it is linked; where the game has no fixed paytable, we say so plainly rather than inventing a number.
The line-up is remarkably stable across the country. A perya of any size will run some combination of these:
Those divide cleanly into three groups, and the division matters more than the names. Some of these games have a fixed, knowable price. Some have a price that depends entirely on a payout the stall has not told you. And some have no odds at all in the mathematical sense, because nothing about them is random.

Three dice, six colours, one colour per face. You put money on a colour, the dealer rolls, and you are paid for each die showing your colour. As a Philippine news feature described it, if your colour comes up “your bet is automatically doubled” — that is, one match pays even money, two matches pay two to one, three pay three to one.
Almost everyone at the table believes they win about half the time. They do not. Over all 216 equally likely rolls, your colour fails to appear on 125 of them:
| Dice showing your colour | Ways out of 216 | Chance | You get back |
|---|---|---|---|
| None | 125 | 57.87% | Nothing |
| One | 75 | 34.72% | Stake × 2 |
| Two | 15 | 6.94% | Stake × 3 |
| Three | 1 | 0.46% | Stake × 4 |
| Any win at all | 91 | 42.13% |
Add those up and the stall keeps exactly 17 pesos in every 216 staked, or 7.8704%. That figure does not move if you back two colours, or three, or all six — the arithmetic is identical to four decimal places, which is the single most useful thing to know about the game. Our full Color Game guide works through every case, including the one that surprises people most: backing all six colours has no winning outcome at all.
Beto-beto is the other dice stall, and it is the one worth reading carefully. Three ordinary dice go into a cup or a covered dish; a board in front of you carries the numbers 3 to 18; you back a total. Whoever has the total that comes up is paid.
The board looks even-handed. Sixteen numbers, take your pick. The dice are not even-handed at all. There is exactly one way to roll a total of 3 and exactly one way to roll 18, but there are twenty-seven ways each to roll 10 and 11. The most likely total on the board is twenty-seven times more likely than the least likely one, and the board gives you no hint of that.
| Total | Ways out of 216 | Chance | Payout that would be fair |
|---|---|---|---|
| 3 or 18 | 1 | 0.46% | 215 to 1 |
| 4 or 17 | 3 | 1.39% | 71 to 1 |
| 5 or 16 | 6 | 2.78% | 35 to 1 |
| 6 or 15 | 10 | 4.63% | 20.6 to 1 |
| 7 or 14 | 15 | 6.94% | 13.4 to 1 |
| 8 or 13 | 21 | 9.72% | 9.29 to 1 |
| 9 or 12 | 25 | 11.57% | 7.64 to 1 |
| 10 or 11 | 27 | 12.50% | 7 to 1 |
Read the right-hand column again. A stall would have to pay 215 to 1 on a total of 3 merely to break even with you, and 7 to 1 on a total of 10. Those are not close. They differ by a factor of thirty.
Suppose a stall paid the same multiple on every number, which is what the board’s appearance suggests. Take a flat 8 to 1. Back a total of 3 and you lose 95.83% of your money on average. Back a total of 10 and the stall loses 12.5% of its money to you, every single round.
No stall survives that, so no stall does it. The conclusion is worth stating directly, because it is the practical lesson of the whole page: a beto-beto board must be priced number by number, and if nobody has told you what your number pays, you do not know the price of your bet. Ask before the dice go in the cup, not after.
And when you are told, check it against the table above rather than against your instinct, because instinct is exactly what the game is built on. The outside numbers advertise themselves — a big payout on 3 or 18 sounds like the generous corner of the board. It is the opposite. Here is what a stake is worth at various quoted payouts:
| Total backed | at 5 to 1 | at 8 to 1 | at 10 to 1 | at 20 to 1 | at 50 to 1 | at 100 to 1 |
|---|---|---|---|---|---|---|
| 3 or 18 | −97.22% | −95.83% | −94.91% | −90.28% | −76.39% | −53.24% |
| 4 or 17 | −91.67% | −87.50% | −84.72% | −70.83% | −29.17% | +40.28% |
| 5 or 16 | −83.33% | −75.00% | −69.44% | −41.67% | +41.67% | +180.56% |
| 6 or 15 | −72.22% | −58.33% | −49.07% | −2.78% | +136.11% | +367.59% |
| 7 or 14 | −58.33% | −37.50% | −23.61% | +45.83% | +254.17% | +601.39% |
| 8 or 13 | −41.67% | −12.50% | +6.94% | +104.17% | +395.83% | +881.94% |
| 9 or 12 | −30.56% | +4.17% | +27.31% | +143.06% | +490.28% | +1068.98% |
| 10 or 11 | −25.00% | +12.50% | +37.50% | +162.50% | +537.50% | +1162.50% |
A minus sign is what the bet costs you; a plus sign would mean the stall is paying above true odds, which is a thing you should assume does not happen and which the table is showing only to make the spread visible. Note the corner: a total of 3 paid at a hundred to one still keeps 53% of everything staked on it. The biggest number on the board is the worst bet on the board at every payout short of 215 to 1.

Sakla uses a forty-card Spanish deck laid out as twenty pairs on a cloth. You put money on a cell, the dealer turns cards face up one at a time, and the first pair to complete wins. The standard quoted payout is 18.
Because the forty cards divide into exactly twenty pairs and nothing distinguishes one pair from another, every cell on the board wins exactly one round in twenty. That is a proof, not an estimate, and it kills the belief that certain pairings run hot. Against true odds of 19 to 1, a payout of 18 to 1 prices the game at exactly 5.00% — the cheapest bet in the perya, and better than plenty of things sold in casinos.
One question changes that answer completely, and no other guide asks it: does “18” mean eighteen pesos of profit, or eighteen pesos handed back including your stake? The second reading is 17 to 1, and it doubles the price to 10.00%. Our full Sakla guide sets both out, along with the finding that covering all twenty cells returns 19 pesos on 20 staked every single round — a guaranteed 5% charge with no gamble in it at all.
Lucky 9 turns up in peryas, in homes at Christmas, and on licensed casino floors, and the three versions are not the same game. The home and fairground version has no fixed rule for the third card, so the banker’s advantage comes from the tie rule rather than from the cards. The licensed casino table game has published rules, and against them the correct line is to hit on four and stand on five, for a house edge of 1.93% — the cheapest number on this page by some distance. Our Lucky 9 guide separates the two properly.
The other card games you may see being played near a perya rather than in one — Tongits, Pusoy and Pusoy Dos — are player-against-player games with no house at all. The money moves between the people at the table. That makes them a completely different proposition from anything with a stall behind it, and the skill component is real.
Perya bingo works the way bingo works everywhere: cards are sold for the round, numbers are called, the first card to complete a line takes the prize. Nobody is betting against a paytable, so there is no house edge in the dice-game sense.
The price is simply the difference between what the cards brought in and what the prize pays out. If a hundred cards sell at ten pesos and the prize is six hundred, the operator has kept forty per cent of the money in the room. That number is entirely at the operator’s discretion and it is never announced. It is worth knowing that this makes bingo’s price structurally invisible in a way the dice games’ is not: with beto-beto you can at least ask what your number pays and then work it out. With a pot game you would have to count the cards sold.
These are played on the ground, not at a booth, and they usually run alongside a perya rather than inside it. Cara y cruz is two coins tossed and a bet on heads or tails. Hantak, described in accounts of the game as three coins of the same value tossed onto a stone slab with a bet on which side they land, adds one rule that matters: coins that miss the slab are not counted.
The arithmetic of three fair coins is worth spelling out because it explains why the game is organised the way it is. With three coins, one side is always in the majority — there are eight outcomes and not one of them is a tie. An even-money bet on the majority is therefore an exact coin flip, with a house edge of zero. A game like that cannot support anyone, which is why the person running it takes a cut of the pot, called the tong. The tong is the entire price.
The slab rule pushes in the same direction. If one coin misses and only two count, half of all tosses are a one-one tie, and whatever the local convention does with a tie is where the money goes. Before you play, that is the rule to ask about — not the coins.
The ring toss, the dart-and-balloon board, the pellet gun, the basketball hoop: these are not games of chance and it is a mistake to think about them as though they were. There is no paytable to analyse and no probability to compute, because the outcome is not random. It is determined by how the equipment has been set up.
A ring that is a few millimetres narrower than the block it must drop over, a hoop with an oval rim, a balloon inflated soft enough to bounce a dart, an air rifle whose sights are off: none of these is a matter of luck, and none is visible from the customer’s side of the counter. The honest thing to say about the skill stalls is that their price is set by physical calibration rather than by arithmetic, so no guide — including this one — can tell you what it is. Treat the fee as the cost of the attempt and the prize as a bonus, and you will be about right.
This is the comparison the perya has never had. Every figure is the share of everything staked that the operator keeps, computed over the full outcome space rather than simulated.
| Game | What decides it | Price to the bettor |
|---|---|---|
| Lucky 9 (licensed table, correct play) | Cards, published rules | 1.93% |
| Sakla at 18 to 1 | 40 cards in 20 pairs | 5.00% (10.00% on the other reading) |
| Color Game | Three six-colour dice | 7.87%, whatever you back |
| Beto-beto, total 10 or 11 | Three dice | Depends on the payout; fair is 7 to 1 |
| Beto-beto, total 3 or 18 | Three dice | 53% even at 100 to 1; fair is 215 to 1 |
| Bingo | A pot, not a paytable | Whatever the operator keeps — not announced |
| Hantak / cara y cruz | Coins, majority always decided | 0% on the toss; the tong is the price |
| Ring toss, darts, pellet gun | Physical calibration | Not computable from the outside |
| Jueteng (not a perya game) | 1,369 outcomes | 26.88% to 70.71% |
Two things stand out. The cheapest bets in the perya — sakla and a well-played Lucky 9 — are genuinely competitive with a casino floor. And the spread across a single fairground is enormous: the same hundred pesos buys you a 5% charge at one stall and a 53% charge at the next, without anything on either board telling you so.
This is genuinely unsettled, and anyone who tells you otherwise is guessing.
On 6 February 2024 the Senate committee on public order held an inquiry into complaints that perya operators and workers were being harassed by law enforcers during anti-gambling crackdowns. The president of the Perya Industry of the Philippines Association told the committee that colour games were legal because operators hold permits issued by their local government units, and that in practice “the authority of the LGU over perya is not recognized by the law enforcement units”, with cases filed under Presidential Decree No. 1602, the general illegal-gambling law.
Senator Ronald dela Rosa, a former chief of the Philippine National Police, made the distinction that matters: perya operations may well be legal, because they are covered by LGU permits — but the permits may not necessarily cover the colour games. A permit to run a fairground is not obviously a permit to run a betting stall. He asked the Department of Justice for a legal opinion to settle it, saying: “We cannot decriminalize illegal gambling… That’s why we’re asking the DOJ to set the record straight.”
We have not found a published DOJ opinion answering that question. So the position as it stands is this: the perya itself is a licensed fairground, the betting stalls in it are a legal grey area that the police have repeatedly treated as illegal gambling, and arrests of colour-game operators and bettors have been reported in several provinces. A mayor’s permit tacked to the booth is not the same thing as a gaming licence, and the reported prosecutions were brought under PD 1602 regardless of it.
Two further points that are not in dispute. Gambling operations in the Philippines are licensed by PAGCOR, the Philippine Amusement and Gaming Corporation, and a perya stall is not a PAGCOR licensee. And where gambling is licensed, the minimum age is 21, not 18.
Search for perya today and most of what comes back is not a fairground. It is a set of Philippine online-casino brands that have taken the perya name and the perya games — colour game, pula puti, drop-ball — and rebuilt them as licensed digital products with live dealers and a mobile app.
They are a genuinely different proposition from the stall, in both directions. In their favour: the ones operating under a PAGCOR licence publish rules, settle automatically, and can be checked against the regulator’s register of licensees before you deposit. Against: the rebuilt games are not the games your grandparents played — multipliers, side bets and jackpot rounds have been added, and each of those has its own price, which is generally worse than the plain bet it sits next to. Nothing on this page’s arithmetic carries over to a version with a 60× jackpot wheel bolted on.
The one rule worth carrying across: check the licence yourself on PAGCOR’s own list rather than trusting the badge in the site’s footer, and treat any operator that will not tell you what a bet pays exactly the way you would treat a beto-beto stall that will not tell you either. If you deposit with an e-wallet, our GCash guide covers the mechanics and the limits.
The other thing that moved online and then did not stay there is cockfighting. Sabong is a bigger business than every perya stall combined, and its online form has been suspended nationwide by executive order since 2022 — our guide sets out what that order actually bans, what it deliberately leaves alone, and why the commission rather than the odds is the whole cost of a sabong bet.
A perya comes once a year and the stalls are part of the fiesta. Nothing here is an argument against putting a hundred pesos on a colour while the ferris wheel is going round. It is an argument for knowing which hundred pesos buys the most evening.
If gambling has stopped being a fiesta and started being a habit, help exists and is free. Talk to someone before the next perya comes to town, not after.

Each of these takes one perya game and works it through completely — the rules, the full outcome space, the exact price, and the beliefs the arithmetic contradicts.
After going through several successful experiences with several industry pioneers, Fadi Hayek has managed to excel in the world of content for online gaming by providing unique content for several projects that allowed him to obtain the necessary experience to launch his career. Last but not least, Fadi also holds several certificates, including but not limited to content marketing and iGaming.