Option leg premium only, as a percentage of spot — says nothing about whether writing calls beat owning the stock
Equity curve cumulative sum of monthly P&L, in percentage points
Covered-call position stock plus calls, in rupees, against simply holding the shares
How to read this
Excess is strategy minus buy-and-hold, as a difference rather than a ratio — both can be negative, and 93% of a loss is an outperformance that a ratio would report backwards. A covered call cushions a decline and caps a rally, so expect it to lead when the stock falls and lag when it runs. Cycles never overlap: one lot of stock covers one contract, so a longer hold writes fewer calls. A revised lot size changes how many contracts the holding covers, not how many shares are owned.
| Entry | Settles | Shares | Lot | Contracts | Entry spot | Strike | Exit spot | Premium | Assigned | P&L |
|---|
What losing months had in common entry conditions only — what was knowable before the trade
Bars show Cliff's delta — the chance a random losing month scores above a random winning one, minus the reverse. The shaded band is the threshold below which an effect is considered negligible; a bar that stays inside it separates nothing.
Why these features and not others
Only conditions observable before the trade appear above, because
the question is what could have warned you. Measurements taken over the cycle
itself — how far the underlying moved, where it finished against the strike
— separate winners from losers perfectly and predict nothing, since they are
simply restatements of the result. They are listed below for understanding, never
as a signal.
No p-values are shown. With dozens of cycles and seven features, testing them all
at once inflates any confidence claim, so effect size and sample count are reported
instead and the result is flagged when the groups are too small to support a
conclusion.
| Entry condition | Losing median | Winning median | Delta | Effect | Direction | n (lose/win) |
|---|
Measured over the cycle describes the loss, cannot predict it
| Measurement | Losing median | Winning median | Delta |
|---|
Monthly outcomes one row per entry month
How to read this table
Each row sells the contract expiring in the check month and closes it at that contract's own expiry — entry and exit are always the same instrument. Strike held is judged on the underlying's close on expiry day, and means the underlying stayed below the strike — not that the trade made money. A breach can still profit when the premium collected exceeds how far the option went in-the-money. Greyed rows had no listed strike at or above the target for that expiry, so no trade could be priced; they are excluded from every statistic above. Entry volume is how many contracts changed hands in the sold strike on the entry day. A zero is highlighted: that premium is a carried settlement mark, not a price anyone could have sold at, and far-dated strikes are often quoted for months without ever trading. Raise min entry volume to make the engine climb to the next strike that actually traded, or leave it at 0 to price those months and judge them yourself.
| Entry month | Entry | Target strike | Traded strike | Entry premium | Yield % | Check month | Check price | Exit premium | Entry volume | Strike held | P&L % |
|---|
Cumulative P&L percentage points — rows are OTM buffer, columns months held
How to read this grid
Cells are shaded by cumulative P&L; hover any cell for its full breakdown. A cell with a high strike held rate but low coverage traded fewer months than its neighbours and is not directly comparable to them. Check traded before trusting a cell: a high return assembled from strikes that never traded was never actually available.
All combinations sorted by cumulative P&L
| OTM % | Offset | Cumulative % | Max DD % | Strike held % | Coverage | Traded | Avg win % | Avg loss % | Worst % |
|---|