The Wheel: puts until the shares arrive, calls until they go, and the tax of every turn

The Wheel is the cash-secured put and the covered call run one after the other, so this page quotes their mechanics rather than repeating them. Tesco is used as a model underlying throughout; this is not a view on Tesco.

The Wheel on one Tesco contract

The Wheel alternates two written options on the same share. With cash set aside, the writer sells a put; while the puts lapse, each premium is kept. When one is assigned, the writer owns the shares and starts selling calls against them; when a call is assigned, the shares go and the cycle starts again. The most that can be lost on a turn is the strike value less the premiums collected, plus costs: £4,472.80 on the turn below if Tesco went to zero. What is given up is the shares' rise above each call strike, and the chance to stop buying on the way down. It is designed for a range-bound share, where owning it at the put strike and selling it at the call strike are both intended outcomes.

Both phases are covered, so both can sit in a cash account: at IBKR the call is Options Level 1 and the put Options Level 3, and the cash account allows each with the cash or shares restricted. What a cash account cannot do is enter a roll as one combination order, or hold any spread: those need a margin account. The Level 1 page compares the account types.

The worst case is the strike value, plus SDRT and commissions, less every premium received, less any dividend received while the shares were held (only if they were held on the ex-dividend date): here £4,500.00 + £22.50 + £2.80 − £52.50 = £4,472.80 once the November put is assigned, falling by each later call premium. With a model probability of assignment of about 24% a cycle for a put like the first one, the Wheel spends on average about 4 put cycles per assignment, if nothing changes; real paths are lumpier. Open the November put in the strategy builder.

Four states, and the tax event on each move

The cycle is easiest to follow as four states. The diagram gives the UK tax consequence of every move between them, which is where a UK Wheel differs most from the US version it is borrowed from: there is stamp duty on the way in, none for the writer on the way out, and far fewer tax computations than the number of trades suggests.

Counting follows one rule (counting computations): a put or call that lapses or is bought back is one computation, dated when it was written; a put that is assigned is not a computation at all, because its premium comes off the cost of the shares; and a call that is assigned folds into the sale of the shares. A full turn, from an assigned put to an assigned call, is therefore one share computation, plus one for each option along the way that lapsed or was bought back (a Wheel turn on the tax page).

What one turn pays, and what it gives up

The turn in the middle of the year runs from Monday 19 October 2026 to Friday 15 January 2027: the November 450 put assigned, the December 450 call lapsed, the January 450 call assigned. The three premiums add to £195.00 before costs. With both strikes at 450p, the turn's best result is exactly those premiums: the shares are bought and sold at the same price, so all of the income is time value, none of it intrinsic. The chart and table show the turn against the Tesco price on 15 January; the dotted line values the November put alone on 19 October, the day it is written, on the binomial tree, so the gap between it and the dashed expiry line is the time value still to be earned.

Below 450p the turn is £144.20 better than having bought the shares at 450p in August, the premiums less the 5.08p interim dividend that the Wheel, still in its put phase on the 15 October ex-date, did not receive. Above about 464p the holder is ahead, and the gap grows £10.00 a penny. That is the Wheel's trade in one line: income that is capped at the premiums, for a downside that is the shares'.

One year of the Wheel, taxed

Over the year the Wheel collected £722.50 of premiums, paid £10.00 to buy back the October put, £23.80 of commission, £65.00 of spread and £44.50 of SDRT (£22.50 and £22.00, both on put assignments; the call-aways cost the writer none), and received one dividend of £97.00. Both turns bought and sold at the same strikes, so the shares added nothing. The total is +£676.20 on £4,500 of cash, before tax. Someone who bought 1,000 Tesco on 17 August for £4,525.50 with SDRT and commission, kept them to 20 August 2027 at 449p and received both dividends would have +£112.30. On this path, which dips below each put strike and recovers to each call strike, the Wheel does well; the next section shows paths on which it does not.

Three lines in the ledger carry the UK lessons. The October put (cycle 2) was bought back at 1.00p on Friday 2 October 2026: under s148 the buy-back is a cost of that put's grant, so the grant and the buy-back are one computation, a gain of £37.20, dated 21 September. The April put (cycle 8) was written on Monday 22 March 2027, in 2026/27, and assigned on Friday 16 April 2027, in 2027/28: its £67.50 premium never appears as a 2026/27 gain, because on assignment it comes off the cost of the shares acquired in the new tax year (£4,357.30); the 2026/27 return, filed later, simply leaves it out. And the May call (cycle 9) lived across the 13 May ex-date. On Wednesday 12 May 2027 Tesco was 438p, below the 440p strike, so the call had no intrinsic value and exercise for the dividend could not pay: the writer kept the shares over the ex-date and received the £97.00 dividend, taxed as dividend income in 2027/28, the year it is paid (when a call is exercised for a dividend). The 30-day rule never bites here: each re-entry is a new put, and shares arrive only on assignment, weeks after the previous call-away. A writer who sold the shares and was assigned back within 30 days would meet it (30-day matching with options).

Three kinds of year: the Wheel, covered calls only, and holding

One path proves little, so the same rules were run through three assumed years, each starting at 450p on 17 August 2026: a steady rise to 540p; a flat year that zig-zags about 3.5% either side of 450p; and a fall to 360p after an early drop. The Wheel writes each put at the listed strike at or below 96% of the price and each call at or above both 102% of the price and the share cost. "Covered calls only" buys 1,000 shares on 17 August and writes a call at or above 104% of the price each month, buying the shares back the next Monday if they are called away; for tax, that re-purchase inside 30 days is matched with the call-away (s106A), so the realised gain is measured against the new shares and the holding keeps its old cost, the covered call page's 30-day trap. Holding buys and keeps. Every option is priced on the model at IV 22%, filled on the tick, and costed as in the ledger.

The ranking changes with the year. In the rising year the Wheel never gets its shares: all 12 puts lapse, it collects put premiums and misses a £900 rise, finishing £313.20 against £1,022.30 for holding. In the flat, zig-zagging year it is the best of the three, because each dip below a put strike is followed by a recovery to a call strike. In the falling year it is assigned in October at 420p and then, holding shares worth less every month, can only write calls at or above its share cost; the premiums shrink from 5.50p to 0.25p, and the last calls pay less than their own costs: £1.40 of commission and £5.00 of spread each. It loses £327.20, less than holding's £777.70 only because it bought later and lower. None of these paths is a forecast; together they show that the Wheel is a bet on a range, paid in premiums.

Greeks: the jump when the shares arrive

The Wheel's exposure does not change smoothly. In the put phase the position behaves like 272 shares, and a one-standard-deviation fall takes that to 658. Assignment then makes it 1,000 overnight, and writing the call brings it back to 654. The call phase repeats the covered call's asymmetry: after a fall the position is almost all shares (919), after a rise barely a quarter (273).

What the underlying changes: Tesco and BP on the model sheet

Two structural points matter more than either name. The Wheel needs physical delivery, so it runs on single-stock options, not on FTSE 100 index options, which settle in cash and never deliver shares. And because the whole cycle is a way of buying and holding a particular share, the share's own risks (results, takeover, dividend cuts) are the Wheel's risks; the premium does not change them. BP and Tesco are model underlyings here, not a view on either.

The same year on the Tesco mini

Run on the 100-share Tesco mini (8TC), the same twelve options need £450 of cash instead of £4,500. Every premium, share amount, dividend and SDRT charge is a tenth of the standard contract's; the commission is not.

The mini turns the Wheel from a £4,500 commitment into a £450 one, which is the difference between tying up a large part of a small account and a fraction of it. The price is the commission, charged per contract and on each assignment: with the spread it takes 49% of the premiums, against 12% on the standard contract; the mini's result is £41.10, not a tenth of £676.20. The mini rate is unconfirmed, and so is broker access to the minis; the cash-secured put page sets out the per-contract arithmetic.

A US wheel turn in pounds: four dates, four exchange rates

On a US share the contract is 100 shares, there is no SDRT, and IBKR charges no commission on assignment, but each amount is converted to sterling on its own date (HMRC CG78310). Take a hypothetical $100 share on Monday 17 August 2026: an 18 September $96 put written at $1.75 (model $1.74, IV 30%, US rate 3.625%), assigned with the share at $93; then a 16 October $96 call written on Monday 21 September 2026 at $2.15 (model $2.17) with the share at $94, called away. Commission $1.00 a trade.

The first three rates are ECB reference-rate crosses; the 16 October rate is an assumption, because that date is still ahead. In dollars the turn made $388.00, the two premiums less commission, since the shares were bought and sold at $96. Converted at one rate that would be £288.48; the correct figure is £231.39, because the same $9,600 of shares cost more pounds in September than they fetched in October. Currency moves on the shares are part of the gain.

A Wheel on a US exchange-traded fund adds a UK trap of its own. If the fund is not on HMRC's list of reporting funds, the gain when the units are sold is an offshore income gain, taxed as income rather than as a capital gain, and after an assignment and a call-away that sale carries both premiums. HMRC's list (edition of 4 September 2026) includes the Invesco QQQ Trust and the iShares Russell 2000 ETF, but not the SPDR S&P 500 ETF Trust (SPY) (US funds and the offshore-fund rules).

Where the worked plan changes course

Rolling mechanics, the credit-only habit and when closing beats rolling are on the rolling page. Two re-entry points deserve care in the UK. Buying the shares back within 30 days of a call-away changes the tax on that call-away, which the covered call page works through; and selling shares at a loss shortly before a put is assigned has the mirror effect, worked on the cash-secured put page.

UK tax of a Wheel turn

Costs, and the account the Wheel needs

Across the year, the costs that the premiums had to clear were £23.80 of commission (£1.40 on each trade and each assignment, IBKR UK tiered rate), £65.00 of spread (half an illustrative 1.00p quote, each way) and £44.50 of SDRT. The spread is the largest: on an ICE Tesco option paying 5p, a 1p-wide quote costs a tenth of the premium each way. SDRT is paid only when a put is assigned, because the writer is then the buyer of the shares; when a call is assigned the exercising holder buys and pays it (who pays SDRT). Tesco's standard contract is 1,000 shares with a 0.25p tick worth £2.50, monthly expiries only (contract sizes).

The Wheel or its parts: what changes in pounds

How these numbers are calculated