The tithi is the easy part. Ekadashi is the eleventh tithi of each fortnight: twice a lunar month, twenty-four in an ordinary year and twenty-six when a month is intercalated. There is nothing to argue about there.
Everything difficult is about which day.
And there are three situations, every year, that a table printing one date per row simply cannot represent. Every listing I checked prints one anyway.
When the tithi spans two sunrises
A tithi is twelve degrees of the Moon's separation from the Sun. Near apogee the Moon crosses that arc in over twenty-six hours, and sunrises are twenty-four hours apart. So the Ekadashi tithi can be running at two consecutive sunrises, and two days genuinely carry it.
Smartas keep the fast on the first. Vaishnavas keep it on the second.
Nirjala Ekadashi in 2024 was exactly this. The tithi ran from 04:44 on 17 June to 06:25 on the 18th, and sunrise at Varanasi is 05:07 on both mornings. Twenty-five point seven hours, both sunrises inside it. Rama Ekadashi that October was the same, at twenty-six point four hours.
A table printing a single date for either of those has silently chosen a tradition on your behalf and not mentioned it. This page shows both days and says which is which.
When the tithi touches no sunrise at all
The opposite case is less known and happens once or twice a year.
Near perigee the Moon crosses twelve degrees in under twenty-two hours. Such a tithi can begin after one sunrise and end before the next, touching neither. No day has Ekadashi at sunrise.
Vijaya Ekadashi in 2024 ran 06:31 on 6 March to 04:14 on the 7th, and Varanasi's sunrises those mornings were 06:16 and 06:15. Twenty-one point seven hours, and it missed both.
The vrat is still kept. It falls on the day the tithi began, which is why that morning shows Dashami at sunrise rather than Ekadashi. This is the dashami-viddha situation the nirnaya literature spends so much time on, arriving not as a doctrinal puzzle but as a straightforward consequence of the Moon moving fast.
Published tables list Vijaya 2024 on 6 March, and Parsva 2023 on 25 September, both of which are kshaya cases resolved this way. This page marks them rather than letting them pass as ordinary.
When an intercalary month adds two more
Roughly seven years in nineteen, an extra lunar month is inserted to stop the festivals sliding through the seasons. That month brings its own pair of Ekadashis, and they have names that exist in no other year: Padmini in the waxing fortnight and Parama in the waning one.
In 2023 they fell on 29 July and 13 July, inside Adhika Shravana. The page finds them, names them, and marks them as belonging to the intercalary month.
Why the names come from the purnimanta month
There is a naming decision here that has to be made once and made consistently, and getting it wrong shifts every Krishna-paksha Ekadashi to the wrong name.
Rama Ekadashi is Kartika Krishna Ekadashi, and it falls in late October, when the amanta month is still Ashwin. Utpanna is Margashirsha Krishna and falls in late November, in amanta Kartika.
So the traditional names are given by the purnimanta month throughout. Shukla-paksha names are unaffected, because the two reckonings agree during the waxing fortnight; only the Krishna ones move. One rule, applied everywhere, rather than twelve exceptions.
If you name them from the amanta month instead, Rama becomes Indira, which is a different Ekadashi, a fortnight earlier.
Twenty-four names, and the two that bracket Chaturmas
Each Ekadashi in the year carries its own name, and two of them do structural work in the calendar rather than merely marking a fast.
Devshayani Ekadashi, in the waxing fortnight of Ashadha, opens Chaturmas, the four months during which no marriages are performed. Dev Uthani Ekadashi, in the waxing fortnight of Kartika, closes it and reopens the marriage season.
Those two dates are the reason the wedding calendar has the shape it does. Between them, and the periods when Guru or Shukra is too close to the Sun to be seen, most of the year is unavailable for a marriage before any daily rule is applied. The muhurat page shows that arithmetic in full.
The remaining names are attached to the month and paksha rather than to anything computed, so the page simply reports them. What it does compute is which ones exist in a given year: twenty-four ordinarily, twenty-six when a month is intercalated, and between twenty-four and twenty-six falling inside any particular Gregorian year depending on where the boundaries land.
How often the awkward cases actually happen
It is worth putting numbers on this, because the impression a listing gives is that Ekadashi is a simple fortnightly event with occasional footnotes.
Across 2023, 2024 and 2025 there are four kshaya Ekadashis (tithis touching no sunrise at all) which is between one and two a year. In 2024 alone, two of the twenty-five Ekadashis spanned two sunrises and split smarta from vaishnava observance.
So roughly one Ekadashi in six is not the straightforward case. That is not a rare footnote; it is a regular feature of the calendar, and it is the reason two people following the same tradition in different cities can end up fasting on different days without either being mistaken.
Parana, which is arithmetic
Breaking the fast has three constraints, and all three are computable.
It happens on Dvadashi, after sunrise, and before the Dvadashi tithi ends. Hari Vasara, the first quarter of Dvadashi, is avoided, so the window opens at whichever is later of sunrise and the end of Hari Vasara, and closes when Dvadashi does.
That is why the window is sometimes a few hours and sometimes a few minutes. It is the overlap of a solar event and a lunar one, and neither is anchored to the other. Six of 2024's parana windows opened late because Hari Vasara was still running at sunrise.
What this page will not tell you
The nirnaya literature (the Dharmasindhu, the Nirnayasindhu) carries further conditions about when an Ekadashi is viddha, pierced by the tithi before or after it, and different sampradayas read those conditions differently. Those are not implemented here.
That is a deliberate omission rather than an oversight. The page computes what is unambiguous: when the tithi runs, which sunrises it touches, and what that implies for each tradition. Where the sampradayas genuinely differ, it shows the candidate days and leaves the nirnaya to the tradition you follow, rather than issuing a verdict it cannot source.
Why the city matters more here than elsewhere
Ekadashi dates are more place-sensitive than most panchang figures, and the reason is worth understanding.
The tithi boundary is a single instant worldwide. Sunrise is not. A tithi ending twenty minutes after sunrise in one city ends twenty minutes before it in another, and that moves the vrat by a whole day.
So an Ekadashi date copied from a national listing may not be the date where you are. The city selector on this page changes the answer, not just the display.
How the numbers check out
Of the twenty-five Ekadashis in 2024, twenty-two match published tables exactly.
The three that do not are precisely the ambiguous ones. Nirjala and Rama are the two-sunrise cases above, where a single published date has picked a tradition. Papankusha is not a split at all: Ekadashi began at 09:09 on 13 October, well after that morning's 05:55 sunrise, so the 13th was Dashami, and 13 October 2024 was Dussehra. The 14th is the only candidate, even though the tithi expired forty-five minutes after sunrise the next day.
That kind of disagreement is worth more than an agreement would be, because it is traceable. You can check the tithi times and the sunrise times and see exactly where the difference comes from.
What the fast itself involves, and what this page does not decide
The observance varies more than the date does. Some keep a complete fast without water, which is what gives Nirjala Ekadashi its name; more commonly grains and pulses are set aside while fruit, milk and certain flours are permitted. Some keep a night vigil. Practice differs by sampradaya, by region and by family, and it differs for the elderly, the unwell, children and pregnant women.
None of that is computed here, and it is worth being explicit about the boundary. This page tells you when the tithi runs, which sunrises it touches, which days that leaves as candidates for each tradition, and when the fast may be broken. What the fast consists of is a question for the practice you follow, and a calculator that answered it would be overstepping what arithmetic can establish.
The same applies to the choice between the two days when they differ. The page will tell you that smarta observance takes the first and vaishnava the second, and it will show you both. It will not tell you which one is yours.