Hindu Festival Calendar 2027: Every Date and Its Rule

How to Read This Calendar (यह कैलेंडर कैसे पढ़ें)

This page is a rendering of a table, not a transcription of one. The 45 observances below, their weekdays, their nakshatras, the classical rule beside each one and every clock figure further down are read at build time out of a single generated file, which is computed from the JPL DE421 planetary ephemeris. Nothing here was keyed in by hand, and that is the only interesting thing about the method: a year calendar is the one page shape where a wrong row looks exactly like a right one, so the defence has to be structural rather than careful.

The reason the dates move at all is that the calendar is lunisolar. A tithi is the span in which the Moon gains twelve degrees of longitude on the Sun, and it is the lunar day a festival is actually tied to; the month, meanwhile, is named from the Sun's position. Because the Moon's speed is not uniform, a tithi runs anywhere from about nineteen to twenty-six hours, so it almost never lines up with a sunrise-to-sunrise civil day. That mismatch is the source of nearly everything a reader finds confusing about two calendars disagreeing, and it is what the Rule column exists to resolve.

One consequence is worth stating before the table: you cannot get 2027 from the previous year by subtracting eleven days. The drift is roughly eleven days per lunar year, but it is not constant, and an inserted lunar month resets it by about a month in the years that have one. Each date has to be recomputed. Times on this page are worked for New Delhi; tithi start and end instants hold wherever you stand, because they describe an angle between two bodies and not a horizon, while sunrise, sunset and anything measured from them move by anything from a few minutes to the better part of an hour elsewhere in the country. For a day-by-day reading of the same quantities, see the daily panchang.

The Full 2027 Calendar (पूरे वर्ष की तिथि सूची)

The table runs in date order from Makar Sankranti on 14 January 2027 to Mokshada Ekadashi & Geeta Jayanti on 9 December 2027. Every row carries a name, a date and a weekday. The Nakshatra column gives the lunar mansion the generated table records against that date, and a dash there means the table holds none for the row rather than that one is unknown — 2 rows are in that position, Makar Sankranti and Holi (Rangwali / Dhulandi), for reasons the next section but one explains. A dash in the Rule column means the default applies.

Counted by month, the year is heavily back-loaded. October alone holds 10 observances — the closing days of the Sharad Navratri sequence, the Purnima that ends that fortnight, Karva Chauth, and then the five consecutive days of the Diwali cluster, all inside one calendar month. The index below is built from the same array as the table, so the counts cannot drift apart from it:

• January (जनवरी) — 3 entries: Makar Sankranti (14 January); Paush Purnima (22 January); Sakat Chauth (Sankashti Chaturthi) (25 January).
• February (फरवरी) — 2 entries: Vasant Panchami (11 February); Ratha Saptami (13 February).
• March (मार्च) — 3 entries: Maha Shivratri (6 March); Holika Dahan (21 March); Holi (Rangwali / Dhulandi) (22 March).
• April (अप्रैल) — 3 entries: Gudi Padwa, Ugadi & Chaitra Navratri (7 April); Ram Navami (15 April); Hanuman Jayanti (20 April).
• May (मई) — 3 entries: Akshaya Tritiya (8 May); Narasimha Jayanti (18 May); Buddha Purnima (20 May).
• June (जून) — 3 entries: Shani Jayanti (4 June); Ganga Dussehra (13 June); Nirjala Ekadashi (14 June).
• July (जुलाई) — 3 entries: Jagannath Rath Yatra (5 July); Devshayani Ekadashi (Chaturmas begins) (14 July); Guru Purnima (18 July).
• August (अगस्त) — 4 entries: Hariyali Teej (4 August); Nag Panchami (6 August); Raksha Bandhan (17 August); Krishna Janmashtami (24 August).
• September (सितंबर) — 6 entries: Hartalika Teej (3 September); Ganesh Chaturthi (4 September); Anant Chaturdashi (14 September); Pitru Paksha begins (15 September); Sarva Pitru Amavasya (30 September); Sharad Navratri begins (30 September).
• October (अक्टूबर) — 10 entries: Durga Ashtami (7 October); Maha Navami (9 October); Dussehra / Vijayadashami (9 October); Sharad Purnima (14 October); Karva Chauth (19 October); Dhanteras (27 October); Narak Chaturdashi / Choti Diwali (28 October); Diwali / Lakshmi Puja (29 October); Govardhan Puja / Annakut (30 October); Bhai Dooj (31 October).
• November (नवंबर) — 3 entries: Chhath Puja (4 November); Dev Uthani (Prabodhini) Ekadashi (10 November); Kartik Purnima (14 November).
• December (दिसंबर) — 2 entries: Vivah Panchami (3 December); Mokshada Ekadashi & Geeta Jayanti (9 December).

What the Rule Column Means (नियम स्तंभ का अर्थ)

Here is the fact that most published calendars leave out, and it is the reason this page is worth more than a list of dates. Because a tithi rarely fits inside one sunrise-to-sunrise day, knowing when a tithi begins does not tell you which day the festival is on. Each observance carries its own test: the festival lands on whichever day holds the tithi at the moment the rule names. Change the moment and you change the date, on the same astronomy, with nobody having made an arithmetic error.

16 of the 45 rows name such a rule. The other 29 carry a dash, and that absence is itself a fact about the table rather than a gap in it: where no rule is recorded, the default applies, which is that the tithi current at sunrise names the day. Most vrats and Ekadashis are fixed that way, and so are the rows for Guru Purnima and Sharad Navratri begins; reading a dash there as “unknown” would be reading it backwards. The rules the table does use are these:

• Purnima at pradosh, Bhadra excluded — The Purnima has to be running at pradosh, and the Bhadra portion of it is set aside for the fire itself. (1 row)
• Sun enters sidereal Capricorn — A solar event rather than a tithi. The date is the day the Sun crosses a sign boundary, so it barely moves between years. (1 row)
• day after Holika Dahan — No tithi test of its own. The day is fixed by the night before it, so it moves only when that night moves. (1 row)
• tithi at madhyahna — The tithi has to be running at madhyahna, the middle fifth of daylight. This can land a day earlier than a sunrise test would. (4 rows)
• tithi at nishita kaal — The tithi has to be running at nishita kaal, the watch that brackets local midnight. A night observance belongs to the day whose midnight it covers. (3 rows)
• tithi at pradosh — The tithi has to be running at pradosh, the window that opens at sunset. This is what fixes a date when a tithi straddles two evenings. (4 rows)
• tithi in aparahna — The tithi has to be running in aparahna, the fourth fifth of daylight. (1 row)
• tithi through sayahna kaal — The tithi has to cover the whole of sayahna kaal, the last fifth of daylight, so a tithi that opens mid-evening defers to the next day. (1 row)

Dussehra / Vijayadashami shows the mechanism at its clearest, because here the rule and the default give different answers. Its Dashami opens 9 October at 09:02 and closes 10 October at 11:41, so it has not started at the first of the two sunrises it touches and is still running at the second. A sunrise test would name the later day. But the rule on the row is tithi in aparahna, and aparahna is the fourth fifth of daylight, which against a New Delhi sunrise of 06:18 and sunset of 17:58 falls well into the afternoon — hours after the Dashami has opened on 9 October 2027, and hours after it has shut on the day following. So the calendar prints the earlier date. Both arithmetics are sound; only one is the test the classical texts assign to this observance, which is why the column is printed rather than assumed.

Akshaya Tritiya is the same story with a different moment. Its Tritiya runs 8 May at 11:43 to 9 May at 09:03, so the sunrise default points at the later of the two days and the recorded tithi at madhyahna at the earlier, and the table prints 8 May 2027. Diwali / Lakshmi Puja works from the evening end instead. Its Amavasya runs 28 October at 20:48 to 29 October at 19:06 against the rule tithi at pradosh, and pradosh for New Delhi on 29 October 2027 is 17:38-20:02. Read the two together and the fit is tight: the tithi opens after the previous evening's window has already shut and closes partway through this one, so only one evening of the pair holds it and there is nothing to arbitrate. In a year where an Amavasya covers two pradosh windows the same rule has to break the tie, and that is where published calendars split. Narasimha Jayanti is the strictest condition in the table — tithi through sayahna kaal asks the tithi to span the whole of the last fifth of daylight rather than merely to touch it, so its window of 18 May at 16:04 to 19 May at 16:03 has to be read against a sunset of 19:06 and not against sunrise at all.

Two Rules That Are Not Tithi Tests (दो नियम जो तिथि पर आधारित नहीं हैं)

Two rows in the table name a rule and carry no tithi window at all, and they are the exceptions that make the column readable. Makar Sankranti and Holi (Rangwali / Dhulandi) are fixed without reference to a lunar day.

Makar Sankranti is solar. The rule on its row is Sun enters sidereal Capricorn, and the crossing this year is at 21:05 IST on 14 January 2027, a Thursday. That is why the row has no tithi and no nakshatra: neither is what decides it. It is also why this is the one date in the calendar that barely moves between years, and why an error in it can sit unnoticed for a long time. The hour matters more than the day here, and the reasoning is set out on our Makar Sankranti 2027 page rather than repeated.

Holi (Rangwali / Dhulandi) is relative. Its rule is day after Holika Dahan, so it inherits whatever date the bonfire night takes and needs no test of its own. Holika Dahan does carry one — Purnima at pradosh, Bhadra excluded — with a Purnima running 21 March at 18:22 to 22 March at 16:13. Bhadra is the Vishti karana, and the rule sets aside whatever part of the Purnima it covers; the shastras are not of one mind about how much of that portion disqualifies the observance. That is a disagreement about jurisprudence rather than about astronomy, and it is the commonest reason a published Holika Dahan date sits an evening away from this one.

The Tithi Windows Behind Each Rule (तिथि का आरंभ और अंत)

The remaining 14 rule rows can be checked rather than taken on trust. Each line below gives the instant the tithi opens, the instant it closes and the test applied between them; put the rule against the window and the date in the calendar follows. Read the pairs carefully and a pattern appears: most of these tithis open in the afternoon or evening of one day and close in the afternoon of the next, which is exactly the configuration that makes a single sunrise rule inadequate for all of them.

Maha Shivratri is the case worth walking through, because it is the one people most often expect to find on the wrong day. Its Chaturdashi opens 6 March at 12:04 and closes 7 March at 13:46, which means it is current at the sunrise of the later date as well. The rule is tithi at nishita kaal, and nishita kaal is the watch bracketing local midnight, so that is where the tithi has to be tested: it covers the midnight belonging to 6 March 2027 and not the one after it. The night-long vigil therefore belongs to the earlier of the two days the tithi touches, which is the opposite of what a sunrise reading suggests.

Where Two Almanacs May Honestly Differ (वृद्धि तिथि और पंचांगों का अंतर)

A tithi long enough to hold the deciding moment on two consecutive days is called vriddhi, a swollen tithi. When that happens the rule stops discriminating, because both days satisfy it, and the calendar has to fall back on a convention. The convention applied here is that the first of the two days takes the observance, and the reason for choosing it was empirical rather than doctrinal: taking the second day moves two recent Diwali dates off the days on which they were actually observed. That is a good reason, but it is still a convention rather than a derivation, and an almanac that takes the later day is following a defensible reading instead of making a mistake.

Exactly one row in the calendar carries the flag, marked (vriddhi) in the Rule column: Durga Ashtami on 7 October 2027. Its tithi runs 7 October at 04:13 to 8 October at 06:28 — longer than a full day, which is what carries it across two sunrises. Notice how narrowly, though: the sunrise on that same row is 06:17, so the tithi clears the second sunrise by a matter of minutes, and a panchang worked for a city far enough west, where the Sun comes up later, might not find it holding that second sunrise at all. If your family almanac prints the following day here, there is nothing to reconcile. Every other row in the calendar is unambiguous on the test it carries, which is a considerably more useful thing to be told than a blanket warning about disagreement would be.

Two further causes of legitimate disagreement have nothing to do with vriddhi. The first is sampradaya: Smarta and Vaishnava practice observe Janmashtami a day apart in many years, so two dates in circulation is the normal state of affairs and not evidence that one source is careless. The second is longitude. A tithi boundary that falls within a few minutes of sunrise or sunset puts two cities on different sides of the same test, because their horizons differ even though the tithi does not, and India is wide enough for that to bite. Where the dates differ by a whole month rather than a day, though, something is genuinely wrong — usually a purnimanta month label attached to an amanta calculation.

One more thing in the calendar reads like an error and is not. 2 dates carries two observances: Sarva Pitru Amavasya and Sharad Navratri begins both fall on 30 September 2027, while Maha Navami and Dussehra / Vijayadashami both fall on 9 October 2027. In each case one tithi closes and the next opens partway through a single civil day, so two successive lunar days share one Gregorian one, and each pair is printed in that order, earlier lunar day first, with the handover falling at 08:06 on 30 September 2027, and at 09:02 on 9 October 2027. That handover is visible in the tithi columns above rather than something you have to take on faith, and it is the reason a well-made calendar sometimes looks like it has double-booked a day.

The Solar Calendar Beside the Lunar One (सूर्य की बारह संक्रांतियाँ)

Running underneath the lunar calendar is a solar one, and almost nobody publishes it beside the festival list even though half the calendar depends on it. A sankranti is the instant the Sun crosses from one sidereal sign into the next. There are twelve in the year, one roughly a month apart, and they are given below in the Lahiri sidereal reckoning the rest of this site uses. They matter for three separate reasons: several regional new years are fixed by one, the closed periods further down open and shut on them, and each lunar month takes its name from the ingress that falls inside it — which is the machinery behind the amanta and purnimanta month labels.

The Capricorn entry at 21:05 on 14 January 2027 is Makar Sankranti, and the festival row and this table are held to the same instant by a guard, because a page printing two different times for one crossing is worse than a page printing none. The Aries entry on 14 April 2027 is the solar new year kept as Baisakhi in Punjab, Vishu in Kerala and Puthandu in Tamil Nadu, all of them fixed by this crossing rather than by any tithi. Notice also how little the intervals vary and yet how much the clock times do: the crossings drift across the day from year to year even as the dates hold, which is why an hour copied forward is a likelier error than a date copied forward.

The Year Has No Intercalary Month (इस वर्ष अधिक मास नहीं है)

An adhika maas is an extra lunar month, inserted when a whole lunation passes without the Sun changing sign. Twelve lunar months come to about 354 days against a solar year of about 365, and the insertion is what keeps the two from drifting apart; it happens roughly every thirty-two months. This year has none. The table records that explicitly, and the absence is worth saying out loud because it is what makes the calendar above read normally: nothing in the second half is displaced a month later, which is not something you could assume from the neighbouring years.

What the year does have is the ordinary set of closed periods, during which the tradition defers weddings, griha pravesh and other formal beginnings. Kharmas, also called Malamas, is the Sun's passage through Sagittarius and then through Pisces. Because it is solar it recurs within a day or so annually, and every edge below is a computed sankranti rather than a template date:

The first of those opened before the year began and closes on 14 January 2027, the Makar Sankranti date itself, which is why the calendar has nothing in it before that. Chaturmas is the other long closure: it runs from Devshayani Ekadashi (Chaturmas begins) on 14 July 2027 to Dev Uthani (Prabodhini) Ekadashi on 10 November 2027, about four months in which the tradition holds that formal beginnings wait. Between the two Kharmas windows and Chaturmas, a substantial part of the year is unavailable for muhurat work, which is worth knowing before planning around it. Our wedding muhurat guide works through how those windows interact.

The Timings, and the Place They Are Computed For (समय: नई दिल्ली के लिए)

Every figure in the table below is computed for New Delhi at latitude 28.6139 and longitude 77.209, and none of it transfers unchanged to another city. That needs saying plainly because it is a defect this site has had to repair: one city's sunrise once appeared on a page as though it were a national figure. The pradosh figures are measured as the two hours and twenty-four minutes following sunset, which is why they walk down the clock through the autumn rows along with sunset itself. Rahu Kaal is one of the eight equal parts into which daylight is divided, and which part it is depends on the weekday. Abhijit is the eighth of fifteen equal parts, so it straddles local noon by construction. None of the three is a clock time in its own right — each is a fraction of an interval the day itself defines, which is why any source quoting one window for the whole country is guessing.

Two readings of that table are useful. Compare the Rahu Kaal column across rows and you can watch the window walk through the day as the weekday changes — it is not a fixed slot, and a page that prints one per weekday for the whole year is wrong for most instances of that weekday. Compare Dhanteras and Bhai Dooj, four days apart, and the pradosh window has already shifted as the days shorten. For the same quantities on any date, worked for your own coordinates, use the panchang tool instead of adjusting these by hand.

Frequently Asked Questions (अक्सर पूछे जाने वाले प्रश्न)

Q1: Two calendars give me different dates. Which one do I follow?

First work out which kind of disagreement it is. If the gap is one day and the row is flagged vriddhi, both are defensible and you should follow your family or regional practice; nothing on this page settles it. If the gap is one day on Janmashtami, that is a difference of sampradaya and has been for centuries. If the gap is one day on a festival whose tithi boundary falls near sunrise or sunset, the two sources may simply be computing for different cities. If the gap is a month, one of them is wrong. And if the two sources disagree about the rule rather than the astronomy, the rule is the thing to ask about — which is why it is printed in the calendar here.

Q2: Can I get these dates by shifting last year's by eleven days?

No, and it is the single most common way a wrong festival date gets into circulation. The lunar year is about eleven days shorter than the solar one, so the drift is in that neighbourhood, but it is not a constant: a tithi varies in length, the rule each festival applies interacts with where its boundary lands, and an inserted lunar month resets the whole offset by roughly a month in the years that carry one. A shifted date will often be close enough to look plausible and wrong often enough to be useless. The only honest method is to recompute, which is what produced the table above.

Q3: Are the times on this page valid for my city?

The dates are, and so are the tithi start and end instants, because a tithi is an angular relationship between the Sun and the Moon and does not care where the observer stands. The sunrise, sunset, pradosh, Rahu Kaal and Abhijit figures are not: all five are computed for New Delhi, and a city a thousand kilometres away will find them displaced by a quarter of an hour or more. That is usually a detail, and occasionally it is the whole answer — when a tithi boundary sits within a few minutes of your local sunrise, your city can legitimately take a different day from New Delhi. Our Vedic astrologers can compute the windows for your exact coordinates.

Q4: Why do some rows share a date, and why is one festival missing a nakshatra?

Both are properties of the table rather than oversights. Two observances share a date when one tithi ends and the next begins inside the same civil day, so the calendar prints two names against one date; the tithi windows above show it happening. A row without a nakshatra is a row whose date is not fixed by the Moon at all — a solar crossing, or a day defined as the one following another day — so there is no lunar moment at which a mansion could be recorded for it. The general principle this page tries to hold to is that an empty cell should mean something specific and be explained, not filled in to look complete.

Q5: Why is Onam not in the calendar?

Because we cannot derive it reliably. Onam is fixed by Thiruvonam nakshatra inside the Malayalam solar month of Chingam, not by a tithi, and Chingam occasionally contains two Thiruvonams, in which case the observed date is not always the first. No rule we tried reproduced the recent observed dates all together, so the row is absent rather than guessed at, and a Kerala panchangam is the right source for it. The sankranti-based regional new years are in the solar table above for the mirror-image reason: the ingress instants settle those exactly, so there is nothing to guess. A calendar that marks the edge of what it knows is worth more than one with every row filled.

Conclusion (निष्कर्ष)

45 observances sit in the table above, opening with Makar Sankranti on 14 January 2027 and closing with Mokshada Ekadashi & Geeta Jayanti on 9 December 2027. No intercalary month displaces the second half of the year, the twelve solar ingresses are set out beside the lunar dates, and exactly 1 row is flagged as a place where two almanacs may honestly print different days. A list of dates is not what this page is for — lists are everywhere, accurate or not. What it offers is the working: put the tithi window next to the rule and the ingress instant next to the sign, and the date stops being something you accept and becomes something you can verify. Where the working was not available to us, as with Onam, the row is missing and the reason is stated. That is a deliberate trade, and it is the trade a reader is better served by.

What a shared calendar cannot do is tell you which of these 45 dates has any bearing on you. Whether a particular observance is one your own chart makes significant, and which part of a permitted window suits a specific undertaking, are not questions about the sky at large; they are questions about a single horoscope, and no year calendar contains the input needed to answer them. That reading needs a kundli in front of it as well as a panchang.