← AlgoRishi Blog · algorishi.com

Expiry Thursday cost us ₹26,200: the full SENSEX trade log for 10 September

10 September 2026 · AlgoRishi
010304080910 Month-to-date cumulative P&L (signals book, 900 qty virtual) +₹33,252
Tracked on real market prices · virtual capital · past performance is not indicative of future results.

Expiry Thursday on the SENSEX is the day most retail traders remember for the wrong reasons. Today we are one of them. The machine took three trades on 10 September 2026 and finished the session down ₹26,200.

We publish the bad days the same way we publish the good ones. Here is exactly what happened, minute by minute.

3trades taken
1 / 2wins / losses
−₹26,200day P&L
+₹33,252month to date

The session, in order

TimeTradeEntryExitHeldResult
11:15 – 11:1774700 CE242.70220.702 min−₹19,800
11:51 – 12:0674700 CE203.75213.7515 min+₹13,400
13:07 – 13:0974800 PE161.90139.902 min−₹19,800

11:15 — the first call, stopped in two minutes

The machine marked a resistance level, waited for a confirmed break above it, and bought the 74700 call at 242.70. The break did not hold. Price came straight back through the level and the fixed stop-loss did its job at 220.70.

That is exactly 22.00 points of loss. Not 30, not 45. The stop distance is fixed before the trade is placed, so the size of the damage is decided while we are still calm — not while we are watching the screen go red.

11:51 — the same level, the second attempt

Thirty-four minutes later the machine bought the same 74700 call again, this time at 203.75. Cheaper, because the first attempt had already bled the premium.

This one worked. The scale-out rule banked part of the position on the way up for ₹8,800, and the trailing floor carried the rest until it was taken out at 213.75 for another ₹4,600. Total: +₹13,400 in fifteen minutes.

Same strike. Same direction. Same rulebook. One attempt lost ₹19,800 and the next made ₹13,400. The difference was not skill or a better opinion — it was that the second break actually held. No system knows which one that will be in advance.

13:07 — the put, and the same 22 points

After lunch the machine flipped side and bought the 74800 put at 161.90 on a downside break. Two minutes later the stop filled at 139.90. Again 22.00 points. Again −₹19,800.

The running total, drawn out

Running P&L through the session ₹0 11:17 stop 12:06 trail exit 13:09 stop −₹26,200 11:15 12:06 15:30
Two fixed-size losses, one managed winner. The shape of the day was set by the size of the stop, not by the number of losses.

The cricket bit

Think of a T20 innings. A batsman who edges to slip for 4 has not necessarily played worse than one who edges to slip for 4 in the next over. Both got out. What decides the scoreboard is how many runs the partnership in between managed to put on before the next wicket fell.

Today our partnership put on ₹13,400 and we lost two wickets at ₹19,800 each. The arithmetic did not work. On a different day the same two wickets fall and the partnership runs to ₹40,000, and the day is green. The wicket size is the part we control; the partnership size is the part the market decides.

Both losing trades were held for exactly two minutes. That is what a fixed stop looks like when a level break fails immediately — it is fast, it is ugly, and it is the cheapest version of being wrong. A trader without a pre-set stop on the same two trades would still have been holding those options at 3:30 PM on expiry day.

Where the month stands

September so far, day by day, as recorded: −₹19,350 · +₹38,723 · +₹3,600 · +₹11,192 · +₹25,287 · −₹26,200. Month to date: +₹33,252.

Today was the single worst day of the month and it removed roughly 44% of what the month had built. That is not a footnote we want to bury. Six recorded sessions, two of them red, and a month that is still positive only because the green days were larger — not because the red days were rare.

If you are reading this because you lost money on an expiry Thursday: the useful question is not "what should I have bought instead". It is "how much was I risking per attempt, and did I decide that before or after I entered". Today cost us 22.00 points per failed attempt because that number was decided in advance.

What we are not going to tell you

We are not going to tell you the machine will make it back tomorrow. We do not know that. We are not going to tell you a filter would have avoided both losses — the second trade came from the exact same level as the first, so any filter that blocked the loss would very likely have blocked the winner too.

What we will keep doing is publishing the log. Three trades, two stops, one trail exit, −₹26,200. That is the day.

Disclosure: AlgoRishi publishes signal tracking on real market prices using virtual capital, for information and education only. Nothing here is investment advice or a recommendation to trade. Futures & options trading involves substantial risk of loss. Past performance is not indicative of future results. AlgoRishi is not a SEBI-registered investment adviser or research analyst.