Every matka guessing trick makes one of two claims, and sorting them by which one is the whole of honest analysis. Either it is arithmetic about a number that has already been declared — in which case it is usually correct, and useless, because it describes a record rather than a future draw — or it is a claim about a number that has not been declared yet, which is the only kind that would be worth anything. This page takes the five formula genres that circulate, reads each one against what this platform's own code stores and computes, and shows where each is secretly resting on a definition, a symmetry, or a coverage trick dressed as an insight. The arithmetic here is not opinion: it is enumerated from the one thousand possible three-digit panas and from the settlement and chart code, so every figure below is reproducible on your own machine.
The three-question test, before you read any formula. (1) Which quantity does it claim to move — the Ank, the Jodi, or the Pana? (2) Is the input it uses available before the declaration you are predicting, or only after? (3) What count would you have to see, over how many declarations, to accept it? A trick that cannot answer question three is not a formula. It is a habit.
What The Software Actually Stores
Start from the record, because it settles more disputes than any argument does. A declared result is three digits plus a timestamp, and the single digit — the Ank — is computed from it as (digit1 + digit2 + digit3) % 10. That is the entire arithmetic in the entity. There is no field anywhere in the schema for a cut number, a guess, a formula, a prediction or a confidence: the words exist in tip-sheet prose and in search queries, not in a column.
Two deprecated methods in that same class are the most revealing code in the whole betting engine, because they are the platform deleting a popular "formula" from its own source. getPatti() was renamed because returning a single digit under the name "patti" was misleading. And getJodi() is annotated, with unusual directness, as WRONG LOGIC: it concatenated the first two digits of one pana, when a jodi is the open session's Ank followed by the close session's Ank — two different declarations. The most common guessing formula in circulation ("add digits from yesterday's number and read the pair") is therefore not a disputed theory here. It is a computation this codebase performed once, labelled incorrect, and removed. How a declared number is matched against a ticket is set out in the results-reading walkthrough.
Genre 1: Digit-Sum Formulas Are True By Definition
"Add the three digits and you get the number" is the most widespread matka formula, and it is exactly correct — as a definition. It is how the Ank is computed, so it cannot fail and cannot predict: it converts a pana that has already been declared into the digit that the chart will print beside it.
The interesting part is what the conversion throws away. The unreduced digit sum of a three-digit pana runs from 0 to 27 — twenty-eight distinct totals — and the Ank is that total modulo ten. So the sum determines the Ank, but the Ank recovers only part of the sum: Ank 3 arrives from totals 3, 13 and 23; Ank 0 from 0, 10 and 20. Enumerating all 1,000 ordered panas, the totals are triangular (the peak totals, 13 and 14, carry 75 panas each; total 27 carries just one, 999), and the chart displays none of them — the unreduced total is not a column on any page here.
| Quantity | Where it lives | Range over 1,000 ordered panas | Can a formula use it? |
|---|---|---|---|
| Pana (3 digits) | Stored, displayed | 1,000 ordered strings; 220 unordered patti | It is the outcome itself — nothing predicts it here. |
| Digit sum (unreduced) | Not stored, not displayed | 0 to 27; peak 13 and 14 at 75 panas each | A "panel sum" rule reads a hidden quantity. |
| Ank (sum mod 10) | Computed on read, displayed | Exactly 100 panas per digit, all ten digits | Definition, not forecast. |
That last row is the one to remember: every Ank digit arises from exactly 100 of the 1,000 possible panas, so the ten digits are equally likely by construction of the map, and the 100 two-digit jodis are equally likely across the full set. The combinatorics of the pana themselves are derived properly in where the 220 patti count comes from.
Genre 2: Cut-Digit Rules Describe A Symmetry, Not A Direction
The "cut" of a digit is that digit plus five, modulo ten. It is real arithmetic and it is used here — not as a prediction, but as a display rule. The chart marks a jodi red when its two digits are equal or differ by exactly five, which is 20 of the 100 jodis: ten doubles such as 11 and ten cross-fives such as 27 and 72. So "cut digit" is a name for a relation the platform already draws on screen.
The claim that tipsters attach to it — if yesterday's open was 2, today's 7 is statistically more likely — fails on the structure of the relation itself. The pairs are 0/5, 1/6, 2/7, 3/8, 4/9, and the relation is mutual: the cut of 2 is 7 and the cut of 7 is 2. A rule with no direction cannot point forward in time; whatever it predicts for 2 it equally predicts for 7, and the two together are 200 of the 1,000 panas, which is 20% of the space and no more. Each digit inside those pairs is exactly 100 panas. Nothing about yesterday changes a single one of those counts.
Two corollaries worth having written down. A "cut family" bet is a coverage decision — you are buying two of ten equally likely digits, which is a stake-and-payout question, not an insight; the arithmetic of what a covered position returns per unit staked is in the rate, stake and net return guide. And because a red jodi needs two real digits, a cell showing a placeholder can never be red, so any red-count taken off the Jodi grid silently under-counts — one of several reasons the counting discipline in the Kalyan chart guide starts by fixing the window before it starts counting.
Genre 3: Family And Coverage Tricks Buy Breadth
The third genre conditions on something true and then draws a false conclusion: "play the double pana on Ank 5", "avoid triples, they rarely come", "7 repeats in a number more often than you think". Enumerating the space settles all of them in one table.
| Claim to test | What the 1,000 panas show | Verdict |
|---|---|---|
| Pana class mix depends on the Ank | Every one of the ten Anks has 100 panas split 72 single / 27 double / 1 triple — identical for all ten. | Conditioning on an Ank changes nothing about class. |
| Triples are "due" after a long absence | 10 of 1,000 panas are triples (1%), and each Ank owns exactly one of them. | Base rate is 1%; absence changes nothing. |
| A named digit "must appear" often | A given digit appears somewhere in 271 of 1,000 panas; exactly once in 243. | True, and it is a coverage fact about three positions. |
| One position is "hotter" than another | Each of the three positions carries every digit exactly 100 times. | Uniform by construction. |
Notice the pattern: every one of these is a statement about the shape of the space, and the space is fixed. Coverage tricks exploit it honestly — buying more combinations does raise your hit rate, in exact proportion to how much of the space you buy, which is why the cost per unit staked is the only interesting number. What coverage never does is change the odds attached to any single combination.
Genre 4: Calendar And "Reversal" Rules
The fourth genre is the one stated most confidently and supported least: if odd Anks run on Monday and Tuesday, bet even on Wednesday "to capture the reversal". This is the gambler's fallacy with a weekday attached, and it has a specific failure mode worth naming — it assumes the generator keeps a memory that the record shows does not exist.
What the data model actually carries per market is one dated row with an open result and a close result. Nothing in it feeds a previous declaration into the next one, and no column stores a streak, a parity history or a bias. The only weekday-aware structure in the schema is a settings table keyed on market plus weekday with a single open flag: it exists to close a market for a recurring day off, it never reaches a digit, and because the key includes the market, one leg's closed Tuesday says nothing about the other leg's. A parity run over two days is therefore a description of two rows that already exist. The arithmetic of how long an absence should last before anything is concluded — including why a number is never "due" — is worked through in the gap arithmetic section of the Kalyan guide.
Genre 5: Chart-Shape Rules
The final genre reads the grid rather than the digits: a missing jodi is "about to return", a leading zero "means" something, a column of red cells is a signal. These are the only tricks with any legitimate content, because the grid really is assembled by rules you can read — but the rules are about presentation. A jodi cell is built from two placeholders and one rule: if the open Ank is missing the cell prints **, and if only the close is missing it prints the open digit with a star, so a "gap" on the board can be a session that has not been declared yet rather than a draw that failed to appear. The sharper consequence is that a missing open blanks the whole pair even when the close digit was declared normally — the record holds it, the cell does not show it. A pair like 07 keeps its leading zero because the cell is a two-character string, not a number. And red styling is applied by the equality-or-five rule above, so a run of red cells is a property of a display function, not of a streak. None of that makes the grid useless — it makes the denominator mandatory.
The Five Genres At A Glance
| Genre | What it claims | What it actually is |
|---|---|---|
| Digit sum | Adding pana digits yields the number | The definition of the Ank, applied after declaration. |
| Cut / ±5 | Yesterday's digit points at its cut today | A symmetric relation used to style red jodis. |
| Family / coverage | Playing a bracket or class is smarter | Breadth bought with stake; odds per combination unchanged. |
| Calendar / reversal | A run of odds must produce an even | Gambler's fallacy; no field carries a previous draw forward. |
| Chart shape | Gaps, zeros and red runs signal a return | Rules about presentation and formatting, not about draws. |
A Written Method For Testing Any Formula
The honest version of "guessing methods" is a test protocol, and it takes about six lines. Write it before you stake, because the point is to be able to say no afterwards.
- Name the target quantity — Ank, Jodi or Pana — and confirm the input you are using exists before that declaration, not after.
- State the claimed rate as a number: "this jodi should appear about 15 times per 1,000 rather than 10". A formula without a number cannot be checked.
- Fix the window and the market row before counting, and write the extraction date beside the count — the archive window slides, so a number taken last month is not reproducible.
- Compare to the base rate and its ordinary band for that sample size, not to zero. The bands for 30, 90 and 365 declarations are tabulated in the Kalyan guide.
- Count how many things you looked at. Ten jodis screened for a hot streak will produce a "finding" with no cause; that is the multiple-comparisons problem, and it is the usual explanation for a formula that "worked".
- Write the kill criterion — the result that will make you abandon the rule — and then abandon it when you see it.
Why none of this is a betting edge. Matka is a game of chance: a position wins only by matching digits that somebody else declares, and no choice of a player affects which digits those are. Every formula above is either a definition, a symmetry, a coverage decision or an untested claim, and the honest ceiling for a "system" built from them is the same expected return as a blind pick — with extra bookkeeping. Play only where it is lawful for you, treat any stake as money spent, and never increase a stake to repair a loss.
Frequently Asked Questions
Is there any mathematical formula that predicts Matka results?
No. The only arithmetic in the platform's record converts a declared pana into its Ank — (d1+d2+d3) % 10 — which is a definition applied after the fact. Nothing in the schema stores a generator, a seed or a rule that produces the next declaration, so there is nothing in the software for a formula to reverse-engineer.
What is the cut digit, and is the cut rule real?
The cut of a digit is that digit plus five, modulo ten, so the pairs are 0/5, 1/6, 2/7, 3/8 and 4/9. The relation is real and the platform uses it to style red jodis. The prediction attached to it is not: the relation is mutual, so it has no direction in time, and each of the two digits covers exactly 100 of the 1,000 possible panas.
Why is "add the digits of yesterday's number" not a formula?
Because adding the three digits of a pana is how its own Ank is computed. Applied to a single pana to produce a pair, it is a computation this codebase previously contained and then deprecated as wrong logic — a jodi is the open session's Ank followed by the close session's Ank, which needs two separate declarations.
Does a number that has not appeared for a long time become more likely?
No. Each Ank arises from 100 of 1,000 panas and each jodi from one of 100 combinations, every time, whatever the gap. Absence lengths are worth counting to check your records are complete, not to forecast a return; the expected gap and its spread are worked out in the Kalyan chart guide.
Do double and triple panas cluster on particular Anks?
They do not. Of the 1,000 ordered panas, 720 are single, 270 double and 10 triple — and every one of the ten Anks holds exactly 72 single, 27 double and 1 triple. The class mix inside an Ank is identical to the class mix overall, so conditioning on the Ank buys no information about the class.
Is betting a "family" of numbers a smarter approach?
It is a different cost, not a better odds. Buying two Anks covers 200 of the 1,000 panas, which is twice the hit rate of one and twenty per cent of the space — and the payout per unit staked shrinks in step. What changes is variance and bookkeeping, which is why the rate and net-return arithmetic matters more than the selection.
What should I do with a tipster's formula that "worked" last week?
Ask how many formulas were tried. Screening ten jodis over a few hundred declarations produces a hot one almost every time with no cause behind it. Re-run the claim on a window you did not select, against the base-rate band for that sample size, and write the number down before you look.
Where does Cut Ank fit among these genres?
In genre two. "Cut Ank" is a guesswork-community word rather than a platform field — as the terminology glossary records, no such column exists anywhere in the schema — and any technique built on it inherits the symmetry problem: the cut relation is the same in both directions, so it cannot single out a future digit.