Auction Drift
You are shown the opening indicative prices of a real NSE closing auction, with the rest hidden. Call where it finished. Then the hidden ticks unfold over your call and the settled price is marked.
Leaderboard
All gamesLoading the board.
XP is pooled across every game, so two rows may have earned their points in different games. Rank is total XP, which rewards playing often as well as calling well.
What a closing auction actually is
Since 3 August 2026, the closing price of an NSE stock with listed derivatives is no longer the last half hour’s volume-weighted average. It is set by a call auction held between 15:20 and 15:30, followed by matching that finishes by 15:35. During collection the exchange publishes an indicative equilibrium price: the price at which the most shares would trade if the auction closed at that instant.
That indicative price moves as orders arrive. It is the closest thing to watching a price being negotiated in public, and it exists for about ten minutes a day.
Why you cannot look up the answer
NSE archives each auction’s outcome, but its official historical file leaves the indicative price and imbalance columns blank. The path is gone from every public source within minutes. We capture it live every trading day, which is the only reason these rounds can exist, and it is why knowing the ticker would not help you: the settled close is public, but the walk that got there is not.
The ticker is withheld until you have made your call, for the same reason.
Why calling it early is worth more
Because it is measurably harder, not as a gimmick. Across the 1,040 symbol-sessions in our archive as of 20 August 2026, this is how far an auction still travelled after a given number of ticks:
| Ticks shown | Median move left | Moved over 0.2% |
|---|---|---|
| 5 | 0.147% | 41% |
| 8 | 0.121% | 37% |
| 12 | 0.099% | 34% |
| 20 | 0.057% | 20% |
Measured over the sessions we hold, which is a small and recent sample. It describes this archive, not closing auctions in general, and it will shift as the archive grows.
Two things the same measurement showed, both of which changed the game before it shipped. Predicting the settle from the LAST visible tick is not a game at all: the final price equals it exactly 67% of the time. So the question is asked from the opening ticks instead.
And a quarter of the possible rounds were dropped. In those, the indicative price had not moved at all across the ticks you would have seen, which left a flat line with nothing to read, and the answer was flat 52% of the time. Guessing flat would have beaten a quarter of the deck. Every round you are dealt now moved at least once while you were watching.
How the answers fall depends on how many ticks you take, so it is worth showing per tier rather than as one average. The last column is what you would score by ignoring the chart and always giving the same answer, which is the floor any read has to beat:
| Ticks shown | Rounds | Higher | Lower | Flat | Best blind guess |
|---|---|---|---|---|---|
| 5 | 640 | 40% | 40.9% | 19.1% | 40.9% |
| 8 | 755 | 37.4% | 38.4% | 24.2% | 38.4% |
| 12 | 809 | 36.6% | 36.1% | 27.3% | 36.6% |
| 20 | 873 | 32% | 26.2% | 41.8% | 41.8% |
Counted from the 3,077 rounds currently in the deck, recomputed on every request rather than typed in once. It moves as the archive grows.
This is a game, not investment advice
Downstox is not a SEBI-registered investment adviser or research analyst, and nothing on this page is investment advice or a recommendation to buy or sell anything. Every round replays an auction that has already settled, so this game expresses no view about what any price will do next, and being right in it says nothing about any future trade. There is no money, no wager, no entry fee and no prize.