The Vertical Spread: What Capping Your Upside Actually Buys

Cheaper is a fact about the invoice

Start with the transaction rather than the label. You buy a call at one strike. You sell a call at a higher strike, same expiry, same quantity. The money you receive for the second reduces what you pay for the first, and the reduction is often large enough to look like the whole point. In the illustrative structure priced below, the single call costs 497.60 points and the pair costs 223.50, so the outlay falls by about 55 percent.

Now describe the same transaction without the word cheaper. You began with a claim on every outcome above one strike, with no upper limit. You then sold, permanently and irrevocably, every outcome above a second strike. What remains is a claim on a band, and you paid for that band with the proceeds of the sale. Nothing was discounted. A piece of the position was sold, and the sale financed the remainder.

Put that way, one question presents itself immediately and is almost never asked: was the piece sold worth more or less than the money received for it? That is a question with an arithmetic answer, and the answer determines whether the structure improved anything at all. It is a different question from whether the trade will work, and it can be settled before any market opens.

The site already treats the surrounding machinery in depth. Our framework for deciding whether an option trade should exist covers the sequence of gates a structure has to pass, and it names the vertical spread as the ordinary directional workhorse without pricing one. The comparison of the four single-option positions derives the kinked payoff and the breakeven arithmetic that everything here builds on, and the strike page explains why cost and odds are always set together. This page does one thing those pages do not: it takes a specific pair of strikes, prices both legs, and follows every number through to the point where it changes a decision.

Two results emerge from doing that, and only one of them is comfortable. The uncomfortable one is that the cap is fairly priced, so the structure creates no advantage on its own. The comfortable one is that a known worst case is worth having anyway, for a reason that has nothing to do with the payoff diagram.

The setup, stated in full so it can be checked

Everything below comes from one deterministic model of an illustrative index sitting at 25,000 with thirty days to expiry, priced at an implied volatility of 15 percent and a financing rate of 6.5 percent, with no dividend adjustment. Those inputs are stated so the arithmetic can be reproduced or disputed rather than taken on trust. One standard deviation of movement over the life of the contract works out to about 1,075 points, which is 4.30 percent, and that figure is the yardstick every strike choice on this page is measured against.

The contract multiplier needs a word of caution. This page uses a lot of 60 units, so one contract is worth Rs 15,00,000, which sits at the lower end of the Rs 15 lakh to Rs 20 lakh minimum contract value the exchanges now work to. That lot is illustrative and nothing more. Exchange lot sizes are revised periodically to keep contract value inside the band: they were raised in November 2024, when the index lot moved from 25 to 75, and revised downwards again from January 2026 as index levels climbed. Anyone reproducing this arithmetic should take the current lot from the exchange rather than from any article, including this one. Our explanation of how the lot is derived from the contract value band covers why the number moves and why it has a floor.

The credit structure here is the one built from puts below the market: short the higher strike, long the lower. It is bullish to neutral and it takes in money at entry. It should not be confused with the bearish put vertical, which is long the higher strike and short the lower and therefore pays a debit. Same instrument, same two strikes, opposite direction, opposite cash flow at entry.

Both payoffs, computed

A vertical spread has exactly one breakeven, not two. A straddle has two because it profits from movement in either direction and therefore crosses zero twice. A vertical is directional: the payoff rises once across the strike range and then flattens permanently, so it crosses zero once and stays on that side. Anyone looking for a second breakeven on a vertical is looking for something the geometry does not contain.

The debit call spread breaks even at 25,223.50, a rise of 0.89 percent, and reaches its maximum of 276.50 points at 25,500, a rise of 2.00 percent. Below 25,000 the whole 223.50 points is gone. Under the model that priced the options, it finishes profitable in about 46 percent of outcomes, reaches its maximum in about 36 percent, and loses everything in about 46 percent.

The credit put spread is the mirror image in shape and the opposite in feel. It takes in 104.80 points, keeps all of them as long as the index holds above 24,500, and breaks even at 24,395.20, which is 2.42 percent below the starting level. It finishes profitable in about 75 percent of modelled outcomes. Against that, it risks 395.20 points to hold 104.80, a reward to risk of about 0.27 to 1, and it hands back everything and more if the index falls through 24,000.

Set the two side by side and the temptation is to prefer the second, because a structure that wins three times in four sounds like a better structure than one that wins slightly less than half the time. The model disagrees, and it disagrees precisely. Integrated against the same distribution that produced the prices, the present value of the expected payoff less the cash paid or received at entry is +0.008 points for the debit call spread, −0.025 for the credit put spread and +0.023 for the single call. All three are zero to within the five paise quoting tick. A seeded simulation of 20,00,000 paths agrees with the integral to within its own sampling error.

This is the first inconvenient result and it is worth stating without softening. Under the assumptions used to price these options, none of these structures has any edge over any other. The frequency of gains is not a measure of quality, it is a description of shape. A high win rate has been purchased with a large loss when losses come, and the exchange rate between the two is exactly fair. Anything that makes one structure better than another has to come from a view that differs from the price, or from a constraint outside the payoff, and the second of those is where this page eventually lands.

The piece you sold was worth what you were paid

Take the sale on its own. The 25,500 call brought in 274.10 points. Under the same model, the expected payoff of that call at expiry, before discounting, is 275.57 points. The two differ by the thirty days of financing on the premium and by nothing else. This is not a coincidence and not a property of these particular strikes. It is what an option price means: the present value of the expected payoff, under the distribution the price itself implies. The tail you sold was worth what you were paid for it, to the paisa, by construction.

That is the answer to the question the first section raised, and it is deflating in a useful way. The sale did not subsidise the purchase. It exchanged one asset for its own fair value. If the structure looks better than the single option, the improvement has to be located somewhere other than in the exchange itself.

The distribution of that fair value is more interesting than its size. The 25,500 call pays nothing in about 64 percent of modelled outcomes and pays something in the rest. When it does pay, the average amount is about 765 points, against the 274.10 you received. So the shape of what you sold is a small, frequent gain against an occasional payment several times larger. That is the shape of the tail, and describing it as the thing you gave up is more honest than describing it as the discount you obtained.

Two consequences follow. The first is that the choice of strike is not a search for a bargain, because at the model price there are no bargains anywhere on the chain. The second is that the whole structure is a statement about which part of the distribution you want, and a bet that the price of that part is wrong in your favour. Our page on what implied volatility is and how it is solved for works through how that price is set and why the market's estimate and the outcome routinely differ. A vertical spread is one way of taking a position on that difference over a narrow range, which is a much more specific claim than the phrase cheaper than a call suggests.

The same view uncapped, and the exact price of the cap

Set the spread against the single call it replaces, on the same view and in the same expiry. The call costs 497.60 points, Rs 29,856 on the illustrative lot, and breaks even at 25,497.60, a rise of 1.99 percent. The spread costs 223.50 points, Rs 13,410, and breaks even at 25,223.50, a rise of 0.89 percent. On outlay and on the move required, the spread is comfortably ahead. It is ahead because it is a smaller position, not because it is a better one.

The exact point where that reverses can be written down. The difference between the two positions at expiry is the short call's own payoff less its own premium, because every other term is common to both. So the single call overtakes the spread at the higher strike plus the premium the higher strike brought in, and at no other level: 25,500 plus 274.10 gives 25,774.10, a rise of 3.10 percent. A grid computed at five point intervals puts the crossing in the same place.

There is a small elegance in that crossing point worth pausing on. At 25,774.10 both positions are worth 276.50 points, or Rs 16,590, and that figure is the spread's maximum profit. The single call does not creep past the spread somewhere in the middle of the range and then extend its lead. It catches up at precisely the level where the spread stops improving, and everything beyond is territory the spread was never going to enter. Under the model, the index finishes above that level in about 27 percent of outcomes.

So the trade is legible: in about 27 percent of outcomes the cap cost something, and in the remaining outcomes the premium received made the spread the better of the two, by up to 274.10 points. Multiply each side by its probability and you are back at zero, as the previous section required. Nothing here is a discovery. It is a description of what was bought and sold, which is what most comparisons of these two positions leave out.

How far apart the strikes sit is the whole decision

Once the direction is chosen, everything that remains is the distance between the two strikes, and that single choice moves every number on the page. Hold the long 25,000 call fixed and vary only the strike sold against it.

At the narrow end, a 25,200 strike sold 0.19 standard deviations away funds 80 percent of the long premium. The debit falls to 100.25 points, the breakeven drops to 25,100, only 0.40 percent above the starting level, and the model puts the chance of collecting the full amount at about 47 percent. The catch is that the full amount is 99.75 points. You have built a structure that wins often and wins about as much as it risks.

At the wide end, a 26,200 strike sold 1.12 standard deviations away funds only 20 percent. The debit rises to 400.10 points, the maximum rises to 799.90, and the reward to risk reaches 2.00 to 1. The chance of collecting that maximum falls to about 16 percent. You have built something much closer to the single call you started with, which is exactly what a short leg that barely pays for anything should produce.

The useful way to hold this is that width is measured in standard deviations, not in points and not in strikes. A 500 point width on this index over thirty days is 0.47 standard deviations, so the short strike sits inside the range the market itself considers ordinary. A 1,200 point width is 1.12 standard deviations and sits outside it. Those two structures are not variations on a theme, they are different trades wearing the same name, and describing both as a vertical spread hides the only decision that was actually made.

What the arithmetic will not do is choose. Every one of the five is fairly priced, so the ranking depends entirely on what the distribution is expected to do, which the model cannot supply. That is the honest boundary of a payoff calculation, and it is worth reaching before rather than after money is committed.

The half that is rarely cited: a known loss turns sizing into division

Everything so far has been mildly deflationary. This section is where the structure earns something real, and the reason is not visible anywhere on a payoff diagram.

Take a risk rule of the ordinary kind: an illustrative account of Rs 50 lakh, with no more than one percent of it, Rs 50,000, at risk on any one idea. For the vertical spread the calculation has one step. The worst case per lot is Rs 13,410, known at the moment both legs fill, so the number of lots is Rs 50,000 divided by Rs 13,410, which is 3.73, and therefore 3 whole lots. There is no scenario, no assumption and no judgement anywhere in that line. Rounding down to a whole lot is the only discretion involved and it is not discretion at all, because fractional lots do not exist. It is arithmetic on two known quantities, and it produces the same answer for everyone who applies the same rule.

Now apply the identical rule to a position whose worst case is unbounded, a plain short call at 25,500. The rule requires a maximum loss and the position does not have one, so a level has to be supplied from outside. Assume a four percent adverse move and the loss per lot is Rs 13,554, giving 3 lots. Assume six percent and it is Rs 43,554, giving 1 lot. Assume ten percent and it is Rs 1,03,554, which the budget will not cover at all. Under the model those three levels are exceeded about 21 percent, 11 percent and 2 percent of the time respectively, and all three are defensible choices. The same rule and the same account produce three different positions, and the difference between them is entirely the analyst's taste.

That is the property worth paying for. It is not that the defined-risk structure is safer, and framing it that way is a mistake the site has argued against elsewhere in its treatment of what limited risk does and does not mean. The maximum loss on a vertical spread is a real loss that arrives in full more often than people expect: in this illustration, about 46 percent of the time. What defined risk removes is the estimate. A number that was a matter of opinion becomes a number that is a matter of record, and every downstream calculation inherits the improvement.

The floor deserves an equally honest treatment. A one percent rule and a maximum loss of Rs 13,410 mean the smallest account that can hold one lot of this spread is about Rs 13.41 lakh. The credit put spread, with a larger worst case, needs about Rs 23.71 lakh. The single call needs about Rs 29.86 lakh. So the structure genuinely lowers the entry threshold, by rather more than half against the single call, and it does not abolish it, because one lot remains the smallest quantity that exists and the contract value band keeps that lot large. The lot size page works through why the floor exists and why it cannot be stepped down. The relevant point here is narrower: the spread moves the floor, it does not remove it.

It is worth sitting with what the floor implies. At three lots the position deploys Rs 40,230 of premium against a Rs 50 lakh account, so the capital committed is small while the capital required to be allowed to commit it is not. That asymmetry is the reason defined-risk structures read as accessible and behave as gated. The regulator's own measurement of the population is the relevant backdrop: about 93% of individual traders in equity derivatives made net losses over FY22 to FY24, with aggregate net losses exceeding Rs 1.8 lakh crore (SEBI, September 2024).

Four legs, four bills

A two-leg structure pays the charge stack twice going in and twice coming out. Whether that matters is an arithmetic question, and the answer is more interesting than either the dismissal or the alarm it usually attracts.

The components are these. Securities transaction tax is charged on the sale of an option at 0.15 percent of the premium, so it falls on the short leg at entry and on the long leg at exit, and not at all on a purchase. Stamp duty is 0.003 percent of the premium on the buy side only. The exchange transaction charge is levied on premium turnover, and goods and services tax at 18 percent applies to the brokerage and the exchange charge but not to the transaction tax or the stamp duty. Brokerage is modelled as a flat Rs 20 per order, which is illustrative rather than sourced, because naming a rate card would mean naming a firm.

On one lot of the illustrative spread the four events come to Rs 35.95, Rs 54.58, Rs 80.11 and Rs 23.60, a round trip of Rs 194.23. Converted into index points that is 3.24, which moves the breakeven from 25,223.50 to 25,226.74. As a share of the capital at risk it is 1.45 percent, and as a share of the maximum profit it is 1.17 percent.

Here is the inconvenient result running in the unexpected direction: at this size, on these premiums, the charge stack is not what makes or breaks the structure. Two legs cost about Rs 194.23 against about Rs 116.05 for the single call, which is more, and neither figure changes any conclusion reached above. Publishing the opposite would have been a tidier story and it would have been wrong.

Where the stack does bite is where the premiums are small, because roughly 49 percent of the bill is a flat fee that does not shrink with the trade. Move the same 500 point width four percent further out, to 26,000 and 26,500, and the debit falls to 76.15 points, Rs 4,569, while the round trip only falls to Rs 160.80. The drag rises from 1.45 percent of capital at risk to 3.52 percent, and the flat component rises to 59 percent of the bill. Narrow, cheap, far-from-the-money spreads are where a multi-leg structure quietly stops being viable at retail size, and it is a fixed cost rather than a tax rate that does it.

One mechanical detail changes the exit bill materially. Index options settle in cash, so a structure held to expiry does not need to be closed. Letting it settle costs the exercise transaction tax on the long leg's intrinsic value, about Rs 45.00 here, against Rs 103.71 to square both legs off in the market, a difference of about Rs 58.71. Against that stands the risk of holding through the final session, which the expiry day page covers in detail. The point is only that the two routes are priced differently and the difference is computable in advance.

The diagram is a photograph of the last day

Every payoff diagram on this page, and every payoff diagram anywhere, describes the position at expiry and says nothing whatsoever about the days before it. For a vertical spread that gap is unusually wide, because the short leg keeps its time value for as long as the long leg keeps its own, and the two do not decay in step.

Suppose the index does exactly what the structure needed and arrives at 25,500 with fifteen days still on the clock. The spread is then worth 325.05 points of a possible 500, and against the 223.50 paid that is a gain of 101.55 points, which is 37 percent of the maximum profit. The view was correct, the target was reached, and roughly a third of the available profit exists. With seven days left it is 44 percent. With two days left it is 62 percent. The flat top of the diagram is not a place the position visits early; it is a place it arrives at on the last day or not at all.

The same effect operates in reverse and is gentler than the diagram suggests. If the index instead drifts to 24,700 with fifteen days left, the spread is worth 130.23 points rather than the 223.50 paid, a loss of 93.27 points, or Rs 5,596, against a maximum loss of Rs 13,410. Time still to run cushions the downside in the same way it withholds the upside. Both are the same fact seen from opposite ends.

Volatility does something more surprising. With the index sitting on the short strike and fifteen days remaining, the spread is worth 343.01 at an implied volatility of 12 percent, 325.05 at 15 percent and 312.20 at 18 percent. It gains when volatility falls. The single call over the identical three settings moves from 625.65 to 669.17 to 717.90, in the opposite direction. Once the index has arrived, the short leg is closer to the money than the long leg and dominates the position's volatility exposure, so the structure that began life as a purchase of optionality has quietly become a sale of it. Nothing on the payoff diagram hints at this, and it is the single most common way a correct directional view produces an unrecognisable position.

Where the structure behaves well, and where it behaves badly

Payoff arithmetic describes behaviour, not suitability. The table below sets out the conditions under which the structure does what its diagram suggests and the conditions under which it does something else, with the reason in each case.

The seventh row deserves emphasis because it is a structural risk rather than a market one. Every claim about defined risk on this page assumes both legs exist. A spread entered as two separate orders is not a spread until the second one fills, and in the interval it is a single option with the risk profile of a single option. Whatever the fill mechanism, the arithmetic only becomes true when the pair is complete.

What the cap actually bought

Two conclusions sit at the end of this arithmetic, and they point in different directions. The first is that the vertical spread confers no advantage in expectation. The tail sold was fairly priced, the money received matched its value, and the resulting structure has the same expected value as the single option it replaced and as the credit structure that looks nothing like it. Anyone reaching for a spread because it is cheaper has bought a smaller position and called it a better one.

The second is that a maximum loss known before entry is worth having for a reason that never appears on the payoff diagram. It converts the sizing step from an estimate into a division. That sounds like a small administrative improvement and it is not, because sizing is the one decision that compounds across every trade a person will ever place. A rule applied to an estimate is a rule applied to an opinion, and it will be quietly bent in whichever direction the opinion favours. A rule applied to a known number cannot be bent without the bending being visible.

Which leaves a fairly plain summary. The cap did not buy a discount, it bought certainty about the worst case, and certainty about the worst case is what makes a position sizing rule enforceable rather than decorative. Everything else on this page, the breakevens, the crossover, the width table, the charge stack, is arithmetic that anyone can reproduce in an afternoon and that changes very little about whether a trade should exist. The sizing consequence is the part that changes something, and it is the part most descriptions of the structure leave until last or omit entirely.

There is a broader habit underneath all of this, which is refusing to accept a structural description without pricing it. It takes an afternoon, it produces numbers that can be argued with, and it reliably dissolves at least one comfortable belief per exercise. If working through the arithmetic here was the interesting part rather than the tedious part, that is the method we teach.

Every figure comes from one deterministic model, seeded at 20260815 so that it reproduces identically on each run. Option prices use the standard closed-form model on an illustrative index at 25,000, thirty days to expiry, implied volatility of 15 percent held flat across strikes, a financing rate of 6.5 percent and no dividend adjustment. Quoted premiums are rounded to a five paise tick, which is the only reason the expected values come out near zero rather than exactly zero. Probabilities are taken from the same lognormal distribution the prices imply, so they are the market's own estimate rather than a forecast, and a seeded simulation of 20,00,000 paths is run as a cross-check on the closed-form integrals.

Charge rates are taken from the site's verified statutory research. Securities transaction tax on the sale of an option is 0.15 percent of premium and on an exercised option 0.15 percent of intrinsic value, both set by section 159 of the Finance Act 2026 with effect from 1 April 2026. Stamp duty on an option is 0.003 percent on the buy side, under Schedule I Article 56A(d) of the Indian Stamp Act 1899. The exchange transaction charge is modelled at Rs 3,250 per crore of premium turnover, the rate notified for index options by exchange notice 20240927-37 with effect from 1 October 2024. Goods and services tax at 18 percent is applied to brokerage and to the exchange charge, and not to the transaction tax or the stamp duty. Brokerage is an illustrative flat Rs 20 per order. One item could not be verified for the derivatives segment: the regulator's turnover fee is verified at Rs 10 per crore for the cash segment under regulation 41(1) of the 2026 broker regulations, and no derivatives figure is asserted here. At the cash rate it would add about Rs 0.05 across both entry legs, which does not affect any conclusion.

All results are illustrative and simulated. They are not a track record, they are not a forecast, and they are not an indication of what any structure would produce in a live account. Lot sizes, strike intervals and charge rates change; the lot used here is illustrative and was chosen only so that contract value sits inside the current minimum band. Nothing on this page is advice to enter, avoid or adjust any position.

• The framework that decides whether an option trade should exist →
• The four single-option positions, and their payoffs →
• What the strike does to the payoff →
• Lot size, and the floor it puts under every position →
• Build a payoff diagram from your own legs →
• What a hedge costs, and why there is no free one →