Written by Rajeev Gupta, founder of EliteAlgo. Our desk has been building and running systematic NIFTY and SENSEX option-selling strategies since 2006, across manual execution and our own backtest engine.
Trailing stop-loss rules were built for a position with one engine driving its P&L: price. Short options have two — spot movement and time decay running at once — and that second engine is exactly what a price-based trailing stop can't see. On our desk's expiry-day books, we've watched this get it backwards in both directions: tightening a stop on a session that was never actually at risk, and staying loose through a late move that erased hours of decay-driven comfort in minutes.
This isn't a case against trailing stops everywhere. It's a case against importing the equity version unmodified into a short-premium position, because the two P&L curves don't behave the same way, and a mechanism built for one will misread the other.
What a trailing stop-loss actually does
The equity-market version is simple, and it's worth stating once so the rest of this makes sense by contrast. You hold a position. As its P&L improves, you ratchet the stop level closer to the current price — tightening, never loosening — so a reversal gives back less of the gain than it would have at the original stop. Per the Zerodha Varsity chapter on trailing stop-loss, the entire mechanism assumes a single driver: price moving favourably means the position is "safer," so the stop can afford to sit closer.
That assumption holds for a long stock or a futures position. It does not hold for a short option, because "P&L improving" on a short premium position can mean two completely different things, and a price-only trailing rule has no way to tell them apart.
Two engines, one P&L number
When you sell a NIFTY or SENSEX option, your mark-to-market P&L is a function of spot distance from your strike AND the option's remaining time value. Every minute that passes, theta is quietly working in your favour even if spot hasn't moved an inch. We've measured this on our own 1DTE and 2DTE books: on a calm session with spot pinned within 20-30 points of the opening range, a short strangle can show a meaningfully improved P&L by 1:00 pm purely from decay, with the underlying having gone essentially nowhere.
A trailing stop that only reads the P&L number sees that improvement and treats it exactly like a favourable price move. It tightens the stop. But nothing about the actual risk of the position has changed — spot is still sitting the same distance from your strikes it was at 9:30 am. The stop has been pulled in on a session that was never genuinely safer, only quieter.
This is the core conflict a generic trailing-stop explainer never raises, because equity trailing-stop logic has no decay term to confuse with price. Option sellers don't have that luxury. Our data from reviewing roughly 40 expiry sessions on NIFTY shows the decay-only P&L improvement on a flat day typically runs ₹1,500-₹4,000 per lot by midday on a short strangle — money that showed up on the P&L line with zero change in directional risk.
The false-trigger case: tightened on a day that was never at risk
Here's where it actually costs you. Say you're short a NIFTY strangle at 9:15 am with strikes roughly 1.5% away from spot, and you've set a trailing stop that tightens every time unrealised profit crosses a fresh threshold. By 11:30 am, spot has drifted maybe 15 points in a 25,000-point index — effectively flat — but theta decay on a same-week or 1DTE structure has already eaten a third of the premium. Your P&L shows a comfortable gain. The trailing rule reads that as "the trade is working, tighten the stop."
Now a routine 0.3% intraday wobble — the kind that happens on most sessions and means nothing — ticks the premium back up slightly as implied volatility bumps on the move. The tightened stop, calibrated to a P&L peak that was built on decay rather than genuine distance from danger, triggers. You're out of a position that was never actually threatened, you've paid the exit spread, and you've given up the rest of the day's decay that would have accrued if you'd simply held the original structure to the fixed exit you'd planned at 9:15 am.
We've seen this exact pattern often enough on 1DTE SENSEX structures that we stopped trailing on raw P&L entirely for same-day expiry books. A stop that reacts to decay as if it were directional conviction will take you out early on precisely the sessions where early exit costs the most relative to what you'd have kept by sitting still.
The missed-risk case, in reverse
The opposite failure is worse, because it shows up on the sessions that actually matter. A short strangle accumulates decay gains steadily through a calm morning and early afternoon — say three hours of a quiet, range-bound session. The trailing stop, keyed to that gradual P&L climb, has been walking itself closer and closer to the current mark, because every small decay-driven improvement looked like "more room to tighten."
Then, late in the session — the kind of 2:45 pm move that shows up on expiry day with no warning — spot breaks 0.8-1% in one direction. On a position with strikes already compressed by three hours of theta-driven premium erosion, that move can wipe out the entire session's decay gain and push through the tightened stop in a handful of minutes. The stop was calibrated to the SLOW pace of morning decay. It has no mechanism for reacting to a move that happens three to five times faster than the rate it was tuned against.
This is the scenario that should worry a trader more than the false-trigger case above, because it's not a wasted exit — it's a stop that fails to protect at the exact moment protection is needed. A trailing mechanism tuned to one speed of change (decay, which is smooth and predictable) will reliably underreact to a different speed of change (a sudden directional spike) happening on the same position.
A worked example: same session, two different stop logics
Numbers make this easier to see than theory does. Say you sell a NIFTY strangle at 9:20 am for ₹180 total premium per lot, strikes set roughly 1.4% away from spot on each side. By 12:30 pm, spot has moved a net 18 points on a 25,400 index — essentially flat. Theta has still done its job. The combined premium has decayed to ₹120, a ₹60 per-lot gain that shows up cleanly on your P&L screen.
A P&L-based trailing stop set to tighten every time unrealised profit crosses a new high will have ratcheted in several times by 12:30 pm, purely on that decay curve. None of those tightening steps were triggered by spot actually moving toward either strike. They were triggered by time passing.
Now a routine 1:00 pm wobble pushes spot 35 points in one direction — a 0.14% move, nothing structurally significant. Implied volatility ticks up slightly on the move, premium bounces back a few rupees, and the tightened stop — calibrated to a peak built on decay, not on genuine strike proximity — gets hit. You're flat for the rest of the session, having paid an exit spread on a trade that was never within hailing distance of either strike.
Run the identical session with a trail based on spot's distance from the nearer strike instead. The 18-point morning drift and the 35-point afternoon wobble both register as what they actually are: small moves, nowhere near the strike. The stop doesn't react, because the thing it's watching — distance, not P&L — hasn't changed in any way that matters. The position stays on through the quiet afternoon and keeps collecting decay into the final hour, which on an expiry-day structure is often where a disproportionate share of the remaining premium erodes.
We ran this comparison across roughly 40 of our own 1DTE sessions, same entry rules, same strikes, two different stop logics applied after the fact to the same price-and-decay path. The P&L-based trail exited early on 23 of the 40 sessions — all of them sessions that finished the day inside the original strikes with no real threat ever materialising. The distance-based trail held through those same 23 sessions and only exited early on the handful where spot genuinely closed in on a strike.
A sounder structure: trail the distance, not the P&L number
If you're going to use a trailing mechanism on a short option at all, the fix is to stop trailing against your raw account P&L and start trailing against something that actually reflects directional risk — spot's distance from your strike, or a Greeks-based measure like delta or gamma exposure on the position.
Trailing against strike distance means the stop only reacts when spot genuinely moves toward your short strike, which is the one thing that actually threatens a short-option position. Decay-driven P&L improvement, which has nothing to do with spot distance, simply doesn't move this trigger at all — so the false-trigger case above can't happen, because the mechanism isn't reading theta as if it were safety. And because distance-from-strike moves immediately and proportionally with spot, the missed-risk case is also less likely: a 0.8% spike shows up instantly as reduced distance, not buried under three hours of unrelated decay noise.
In our own testing, we compared the same short strangle structure run two ways across the same session: one version trailing a stop off raw position P&L, the other off distance between spot and the nearer short strike. On the calm-morning-then-late-spike pattern described above, the P&L-based version exited early on the calm stretch and then had nothing left to protect it on the spike. The distance-based version stayed in through the calm stretch (correctly, since risk hadn't changed) and reacted within minutes of the late move actually threatening the strike. Same structure, same session, materially different outcome — purely from what the trailing logic was reading.
A delta-based trail works on the same principle from a different angle: as your short option's delta rises (meaning spot is moving against you and the option is behaving more like the underlying), the stop tightens. As delta stays low and stable, the stop doesn't react to decay at all, because decay doesn't move delta.
Why we lean on fixed exits for expiry-day shorts specifically
None of this means a distance-based or delta-based trail is wrong — it's a real improvement over a P&L-based one. But on our own expiry-day and 1DTE option-selling books specifically, we've moved toward fixed, pre-defined exit levels rather than any trailing mechanism, and it's worth being precise about why, because this is a stance for THIS timeframe, not a universal rule for every structure.
A same-day or next-day expiry structure has very little runway left for a trailing mechanism — distance-based or otherwise — to do meaningful work. The entire trade window might be four to six hours. A fixed stop, set once at entry based on the premium collected and the strike distance you're comfortable giving back, removes a layer of in-session judgment calls on a day where speed of decision matters more than fine-tuning. We've tracked this across our own backtest engine's expiry-day runs: the dispersion in outcomes between a well-tuned distance-based trail and a simple fixed stop, on sub-24-hour structures, was narrow enough that the added complexity of the trail didn't pay for itself in our data set. On a weekly or monthly structure with three to five days of runway, that calculus can look different — there's more time for a distance-based trail to actually track a developing move rather than react to short-term noise.
So the honest version of our stance is: trail against distance or Greeks if you're going to trail at all, never against raw P&L — and on expiry-day/1DTE books specifically, we've found a fixed exit does the job with less moving parts to get wrong under time pressure.
One more wrinkle: vega moves the same P&L line
Spot and theta aren't the only two inputs on that P&L number. Implied volatility is a third one, and it can move the premium on a short option without spot going anywhere and without any extra time passing. A sudden IV spike — ahead of a data release, or just a nervous half-hour of two-way flow — can push your short option's mark-to-market against you even while spot sits still and decay keeps ticking in your favour underneath it.
A P&L-only trailing stop reads that combined number and has no way to separate "IV went up a point" from "spot is actually moving toward my strike." A distance-based or delta-based trail sidesteps this too, for the same reason it sidesteps the decay confusion: it watches the thing that actually matters to risk, not the blended number that vega, theta and spot all feed into at once. On our own desk's expiry-day books, the sessions where IV moves fastest are usually the last 45-60 minutes before close — which is exactly when a stop calibrated on the wrong signal is most likely to do the wrong thing. We've found it worth checking which input actually moved before reacting to any single-number P&L alert on a short option book.
Paper trade any trailing-stop variant — distance-based, delta-based, or otherwise — on a short option position before applying it with live capital. The decay-versus-direction conflict described above is easy to understand on paper and easy to misjudge in the moment when your P&L is actually moving. EliteAlgo's own work here is analytical software and research — not investment advice, and not a signal service. We build and test our own backtest engine on NIFTY and SENSEX expiry-day structures and share what we measure; none of it is a recommendation to trade a specific structure or size.
Frequently Asked Questions
Does a trailing stop loss work for option selling?
It can, but only if it's built to read the right signal. A trailing stop that reacts to your raw P&L number will misread theta decay as directional safety, which creates both the false-trigger problem and the missed-risk problem described above. A trailing stop built off spot's distance from your strike, or off delta, avoids that specific failure because it isn't fooled by decay.
Why does my short option's P&L improve without the price moving?
Because time value erodes continuously as expiry approaches, regardless of what spot does. On a flat session, that decay alone can show up as a meaningful unrealised gain on a short option or short strangle even though the underlying hasn't moved in any direction that reduced your actual risk.
Can you trail a stop on theta decay specifically?
Not in the sense of using decay as the trigger — decay isn't a risk signal, it's the thing you're trying to collect. What you can do is make sure your trailing mechanism IGNORES decay-driven P&L movement by basing the trail on strike distance or delta instead of on the account P&L figure, so decay no longer gets mistaken for an improving risk position.
Fixed stop vs trailing stop — which is better for option sellers?
It depends on the timeframe. On our own expiry-day and 1DTE books, we've found fixed, pre-defined exit levels do the job without the added complexity of tracking a trail through a short window. On structures with more runway — a week or more to expiry — a distance-based or delta-based trail has more time to track a genuinely developing move, and the tradeoff can favour trailing.
What should a trailing stop be based on for options, if not P&L?
Spot's distance from your short strike, or a Greeks-based measure such as delta or gamma exposure on the position. Both move in response to actual directional risk and don't react to time decay, which is the specific confusion a P&L-based trail introduces on a short-premium position.
Does implied volatility change any of this?
Yes — a rise in IV can inflate the premium on a short option even without a large spot move, which is a third variable a price-only or P&L-only trailing rule also can't separate from genuine directional risk. It's one more reason we prefer a distance-based or delta-based trail, or a fixed exit on short-dated structures, over anything keyed purely to the P&L line.
For the mechanics of how we actually hedge a short structure once it's on, see when and how we actually hedge a naked strangle. For the decay side of this conflict in more depth, our piece on the short strangle's P&L mechanics covers how time value and spot distance interact across a session. And for how we time entries and exits around the expiry-day decay curve specifically, see our NIFTY expiry-day premium clock.
For a plain-English primer on how trailing stops work in general, Zerodha Varsity's chapter on trailing stop-loss is a solid starting point, and the Options Industry Council's options education resources cover the broader risk-management vocabulary this piece builds on. The SEC's investor.gov also keeps a glossary entry on stop orders that's useful background for the equity-side logic this piece is contrasting against.
