Short Strangle
A short strangle writes an out-of-the-money put and an out-of-the-money call on the same expiry: here the FTSE 100 October 10,200 put and 11,250 call, sold for 90.5 points, £905.00 on one £10-a-point contract. The credit is the ceiling on profit, kept in full on any settlement from 10,200 to 11,250. Below 10,200 the loss grows by £10 a point until the index reaches zero; above 11,250 it has no limit. The writer gives up protection: no option is bought, so nothing caps a large move. The structure is built for a view that the index will move less than the options imply. The index is a model input here; nothing on this page is a view on it.
It is written for readers who already know the iron condor (this strangle with wings bought) and the two credit spreads inside it. The short straddle is its mirror with both strikes at the money; this page quotes the straddle's figures rather than re-deriving the shared mechanics, and the straddle page covers the hedged, realised-volatility view of the same trade. Every number is modelled on the library's model sheet.
Where 16 delta lands on a skewed chain
Strangle writers often choose strikes by delta rather than by distance: a put and a call of about 0.16 each, so that each sits roughly one standard deviation from the index. On a chain where every strike has the same implied volatility that produces two roughly symmetrical strikes. The FTSE 100 chain is not like that. Lower strikes carry higher implied volatility, because protection against a fall is in demand (skew); the library's surface gives the 10,200 put 16.10% and the 11,250 call 12.18%. Higher volatility raises an out-of-the-money put's delta, so the 16-delta put moves further from the index; lower volatility shrinks a call's delta, so the 16-delta call moves closer.
The credit barely changes, 89.08 points against 90.49, but where it comes from does: 60% of it now pays for the put, against 49%. The last row shows what happens to strikes chosen on a flat model once the chain's own volatilities are used: the 10,250 put has a delta of −0.184 and the 11,300 call +0.126, so the pair is no longer balanced and leans towards the downside before the index has moved. Skew also reshapes the odds the chain implies, though not in the way the higher put price might suggest. Read from the surface (risk-neutral), the index settles below 10,250 with probability 16.1%, slightly less than the 16.8% of a flat 14%, below this page's 10,200 put with 14.3%, and above 11,250 with 17.5% against 17.6%: moderate moves are priced much as before. The difference is the far left tail. The surface's 1-in-100 settlement is 9,210, against 9,585 on a flat 14%. That tail is what the put's extra premium pays for, and it is why the stop and cycle sections below weigh the downside more heavily.
The 10,200 put and 11,250 call on Tuesday 1 September
Both legs are ICE FTSE 100 options (ESX): £10 a point, European, cash-settled on the exchange delivery settlement price, so there is no early assignment, no delivery and no SDRT (FTSE 100 contracts; contract sizes). The library fixes the index at 10,750 for all its FTSE examples; the real close on 1 September 2026 was 10,789.3 (price data: Yahoo Finance). ESX strikes are listed at 25 to 200-point intervals depending on expiry and distance from the money, so the live chain decides which strikes exist. A strangle, like a straddle, sits in IBKR's Options Level 4, which needs a margin account (permissions and account types).
Open this worked example in the strategy builder (it opens with this page's two fills over 45 days, backs out each option's volatility and, unlike the surface figures here, works out probabilities at a single 14%).
Payoff: a plateau from 10,200 to 11,250, then no floor
Everywhere on the plateau, 1,050 points wide, the result is the same £905.00, and the profit zone between the breakevens is 1,231 points wide, 11.5% of the index. The model's one-standard-deviation range for 16 October, 10,234.3 to 11,291.6, lies entirely between the breakevens: the credit is 0.17 of the 528.4-point one-SD move, so the writer is paid a little for a lot of room. Read from the surface, the model probability of keeping the whole credit is 68.2% and of any profit 76.1%. The price of that room shows at the edges, where each point costs £10 with nothing underneath. The curves also show how the plateau is earned: with the index unchanged on Friday 25 September the position shows £661.76, and only £243.24, 27% of the credit, is still to come.
Stress and requirement: seven instant moves on 1 September
Method as on the straddle page: each shock hits on 1 September, every strike keeps its surface volatility, and the stated shift is then added to all of them. Margin follows the published Cboe rules for uncovered broad-index options, shown only to give a number to reason with (how the formula works); an ICE broker sets its own figure.
At entry the call leg sets the requirement: its 36.0 points plus 15% of the index less its 500-point out-of-the-money amount (1,112.5) is larger than the put's 54.5 plus 1,062.5 (above its floor of 1,020.0, 10% of the strike), so the requirement is £12,030.00, of which the account funds £11,125.00 beyond the credit. In a fall the put takes over and the requirement climbs to £29,279.40 after a 20% gap, while the mark shows −£15,474.40, 17.1 times the credit; an account that began below £44,753.80 would then be under its requirement with one contract. The broker's response and the index's gap record are on the Level 3 page (margin spiral; gap history; no negative-balance protection).
The row a strangle writer meets most often is the fourth. One standard deviation down, at 10,234.3, the index is still above the 10,200 put, so settling there would keep the whole £905.00; yet the position is marked at −£1,526.59, 1.69 times the credit, because the put has 45 days left and two more points of volatility. The mark and the payoff disagree most exactly where a stop is usually placed.
Where a stop fires, and what it trades away
A stop on a short strangle is usually set as a multiple of the credit: close when the marked loss reaches one, two or three times what was collected. The table solves the model for the index level at which each fires, instantly on 1 September, with volatility unchanged and with volatility four points higher, and gives the model probability (lognormal, IV 14%, zero drift) that the index touches that level at some point before expiry. For comparison, the breakevens sit at 10,109.5 (−1.25 SD) and 11,340.5 (+1.09 SD), and the probability of the index finishing below 10,109.5 on the same model is 11.0%.
A one-times stop fires above the put strike, where settlement would still pay the full £905.00, and with a four-point rise in volatility it fires at 10,595, −0.30 SD, with the index barely moved. Because every model path that ends below the breakeven must pass 10,306 on the way, the ratio of the two probabilities is the share of stopped paths that would have lost at expiry anyway: 28%. Most one-times stops close trades that would have paid. A two-times stop, the one this page's worked plan uses, sits almost exactly on the breakeven when volatility is unchanged (10,110 against 10,109.5), and £283.41 beyond the one-SD row's marked loss; on the same reasoning about half (51%) of the paths that reach it go on to finish below the breakeven. With volatility four points higher it fires at 10,250, close to one standard deviation down, so a sharp rise in volatility alone can bring it forward. A three-times stop fires beyond the breakeven, where a path that touches it is already losing at the settlement price, and in exchange it lets the loss run to £2,715.00. None of these is a correct setting: each trades being shaken out of paths that recover against the size of the loss taken on paths that do not. A gap passes every level at once, as Branch C shows. The straddle page finds that on an at-the-money straddle a one-times stop fires beyond the breakeven instead; the difference is the strangle's small credit against its wide range.
Worked example: the strangle from 1 September to 16 October
Branch A: half the credit, Friday 18 September. With the index and volatility unchanged, the model value of the strangle falls to half the credit, 45.25 points, during Thursday 17 September; on the 0.5-point tick the two legs can first be bought back at or below that on Friday 18 September, with 28 days left: the put at 26.5 (model 26.68) and the call at 15.5 (model 15.75), 42.0 points in all. The worked plan closes: £478.20 after four commissions, £418.20 after the spread. For tax that is two computations dated 1 September, each grant less its buy-back: £276.60 on the put and £201.60 on the call, both in 2026/27. Holding to expiry instead would have kept £901.60 with any settlement between the strikes, and ended at −£2,098.40 with a settlement at 9,900.
Branch B: the put is tested, Wednesday 16 September. Fifteen days in, the FTSE 100 is sitting on the put strike, 10,200, and every volatility is two points up. The 10,200 put now costs 207.64 to buy back and the 11,250 call 1.20; the position is marked at −£1,183.35 (1.31 times the credit), short of the two-times stop, and its delta is +£4.75 a point. The worked plan's move is to roll the untested side toward the money: buy back the 11,250 call at 1.0 and write the 10,700 call at 39.0 (model 38.85, delta +0.16), a further 38.0 points, £380.00. Total credit collected rises to 128.5 points, the delta falls to +£3.25 and the breakevens become 10,071.5 and 10,828.5. The strikes are still 500 points apart, so the position is still a strangle.
Two days later, on Friday 18 September, the index opens at 9,950 with volatility four points above entry. Buying both legs back would cost 383.5 points (put 366.0, model 366.01; 10,700 call 17.5, model 17.70), a result of −£2,560.20 after all six commissions, 2.83 times the original credit: the loss is beyond the two-times stop, and the worked plan closes. The alternative a writer sometimes reaches for is to invert: roll the call down again, below the put, to 10,000 at 208.0 (model 208.07), a further 190.5 points. Total credit would reach 319.0 points (£3,190.00) against an inversion width of 200 points (£2,000.00), and because the credit is larger than the width, any settlement from 10,000 to 10,200 would keep £1,190.00, with breakevens at 9,881.0 and 10,319.0. The price is a narrow zone and full exposure either side of it: settled at 9,800 the inverted position ends at −£810.00, at 9,500 at −£3,810.00 and at 10,500 at −£1,810.00 (at 10,100, £1,190.00). Had total credit been smaller than the width, no settlement at all could have made it profitable. Roll mechanics are on the rolling page, and the choice between closing and rolling on its close-or-roll table.
Branch C: the 9,675 opening, Tuesday 8 September. A week in, the FTSE 100 opens 10% down at 9,675, 38 days before expiry, and every strike's volatility is ten points up. The strangle is marked at −£5,665.75, 6.3 times the credit, and the requirement is £21,083.25. Every stop in the menu was passed overnight; closing at the opening prices, put 652.0 (model 652.23) and call 5.0 (model 4.85), realises −£5,671.80 after commissions. The iron condor on the same short strikes would be marked at −£1,108.23 that morning.
Branch D: held to settlement at 10,900 on 16 October. Both legs expire worthless and the result is the credit less the opening commission, £901.60: two grant gains dated 1 September, £543.30 and £358.30. Cash settlement leaves nothing to deliver and no SDRT to pay. It sets aside the 21-day convention so that settlement itself can be seen.
Greeks, and the gamma at a tested strike
With the index sitting at 10,750 the strangle's Greeks fade as expiry approaches: gamma from −£0.092 to −£0.016 and theta from £28.66 to £5.51 a day, because both options are drifting towards worthless. That is the misleading half of the picture. The other half is the index sitting on a strike.
At a tested strike, gamma rises 4.3 times between 45 and 2 days left at one unchanged volatility. A further 3% fall from 10,200 to 9,894, at the same volatility, costs £1,775.44 with 45 days left and £2,234.22 with 7. Inside the range the strangle is a slow position; on a strike in the last fortnight it is a fast, directional one that the writer never chose (position Greeks). That, not a belief that time decay slows down, is the reason for the plan's 21-day time stop.
Strangle or iron condor over repeated cycles
The iron condor is this strangle with a 9,950 put and an 11,500 call bought, 250 points beyond each short strike. The wings cost 28.0 and 8.5 points (model 27.96 and 8.48), £365.00, 40% of the strangle's credit, leaving the condor 54.0 points. What they buy is a floor. The table compares the two over one 45-day cycle and over eight, about a year of cycles, under two models that are labelled as models and nothing more: the surface's own risk-neutral probabilities, which price every option fairly, and a lognormal index with 12% volatility and no drift, two points below the 14% sold at the money, which is the kind of gap between implied and realised volatility the implied volatility page documents in the FTSE 100 record.
On the surface's own probabilities neither structure has an edge: the credit is the fair price of what the options are expected to pay, and costs make the expectation negative. Any edge has to come from the index proving calmer than the options priced. In the 12% model the strangle earns more per cycle than the condor, £315.42 against £82.60 after costs, because it keeps the £365.00 the wings would cost. But that model has no fat left tail, which is exactly what skew prices. On the surface's probabilities a settlement at or below 9,788 happens one cycle in 20 and leaves the strangle at −£3,215.00 and the condor at −£1,960.00; over eight cycles the chance of at least one such cycle is 34% (1 − 0.958). The 1-in-100 fall leaves the strangle at −£8,995.00, about 29 cycles of its model profit. Per pound of requirement the condor collects 2.9 times as much as the strangle. The iron condor page works a condor on the same chain with 10,400/11,100 short strikes and 250-point wings (£910.00 credit, £1,590.00 maximum loss); the version here keeps the strangle's own short strikes, and the sizing framework sets how much of either an account can carry.
The spread in pounds
A strangle's credit is small against its legs' prices, so the cost of crossing the market takes a large share of it. The table assumes the same quoted spread on every leg, crossed by half each way, and closing before expiry; the straddle column uses the straddle's £4,200.00 credit.
The same quote costs the straddle a sliver of its credit and the condor a quarter of its credit at 3 points. The strangle sits between the two, and the spread is the single cost a writer can see before the order is placed: on a thin chain a 6-point quote on each leg would take 14.0% of the credit before the index moved at all. The spread quoted on a live screen, not the model's mid-price, is what a fill is measured against (cost conventions).
A smaller strangle: the Mini FTSE 100 daily options
One ESX contract is £10 a point, about £107,500 of index at 10,750, and cannot be split. ICE also lists Mini FTSE 100 Index Daily Options (8LX): £1 a point, European, cash-settled on the closing auction, with the front five daily expiries plus the third Friday (ICE product page, checked 26 September 2026). A tenth-size strangle therefore exists, but not in a 45-day expiry: on 1 September the longest listed is the 18 September third-Friday series, 17 days away. Whether a UK broker offers 8LX, and quotes it two ways, we could not confirm (checked 26 September 2026).
On the same surface the 16-delta strangle for 18 September is the 10,400 put (−0.15) and the 11,050 call (+0.17), worth 28.50 and 26.36 and filled at 28.5 and 26.5: 55.0 points, £55.00 a contract, with breakevens at 10,345.0 and 11,105.0. Three things change with the size. Ten of them collect £550.00 against the 45-day ESX strangle's £905.00: 17 days carry less time value than 45, though more of it per day. Per pound of credit the short-dated strangle carries 2.7 times the gamma of the 45-day one (gamma −£0.0150, theta £4.66 a day per contract), so the same shock costs more of what was collected: the instant 10% fall marks one 8LX strangle at −£690.22, 12.5 times its credit. And a fixed commission weighs more: at IBKR's £1.70 index-option rate, assumed here to apply to 8LX, opening and closing costs 12% of the credit before any spread.
UK tax: no bought legs, and one way into the next tax year
Each written leg is one disposal on the day it is written (TCGA 1992 s144(1)); a buy-back is folded into it (s148); a leg settled in cash against the writer becomes one transaction with its grant, dated at settlement (s144A(2)). A strangle has no bought options, so the tax year of its result is almost always the year it was written. Figures assume the £3,000 annual exempt amount is used elsewhere and show 18% and 24%.
The strangle's own trap is the settled leg: a put written in March and settled against the writer after 5 April moves its whole result, here −£1,456.70, into 2027/28, where it cannot reduce the 2026/27 gain on the call. The condor adds a second route into the new year: its wings' loss, −£368.40, arrives in 2027/28 even when everything goes to plan, while the £901.60 of grants is taxed in 2026/27 (£162.29 at 18%, £216.38 at 24%); the relief is deferred, not lost (positions across 5 April; counting computations). Options cannot go into an ISA, and we found no SIPP administrator that allows them: this is a general-account position (wrappers, checked 26 September 2026). SA108 boxes are listed on the SA108 page.
Commission, permissions and a single-stock strangle
Commission at IBKR UK's fixed index-option rate is £1.70 a contract: £3.40 to open, and nothing more if the strangle is held to cash settlement. No SDRT arises on the index. The spread is the larger cost, as the table above shows. Brokers that offer ICE options are mapped on the broker page, and index options offered as spread bets or CFDs are compared with listed ones on the three-routes page.
On a single share the strangle becomes small and physically settled. On BP, at the library's illustrative 530p and 26% volatility for the 60 days from 17 August to 16 October 2026 (no ex-dividend date falls in that life, so no dividend is modelled), the October 490 put (−0.20) and 570 call (+0.28) are worth 6.46p and 9.24p; filled at 6.50p and 9.25p on the standard 1,000-share contract they collect £157.50. Half of an illustrative 1.00p quote each way on both legs plus £1.40 commission per leg each way would take 16% of that credit. An assigned 490 put means buying 1,000 shares at 490p and, as the buyer, paying £24.50 of SDRT. Early assignment of a short put comes when it is deep in the money and the interest on the strike outweighs its remaining time value, which on BP is more likely just after an ex-date than before one; a December strangle would also face the 12 November ex-date on its call (worked on the straddle page; early put assignment). BP is a model underlying; this is not a view on BP.