Strategy · 6 min read
Strike selection with probability instead of hope: what Delta approximates, how the expected move frames strikes, and balancing win rate against payoff.
Most traders pick strikes by feel: a round number, a support level, whatever makes the credit look attractive. The market, meanwhile, prices every strike off a probability distribution and charges you accordingly. You do not need to out-model anyone to trade better. You need to read the probabilities already embedded in the chain and stop confusing a high win rate with a good trade.
An option's Delta doubles as a serviceable estimate of the probability it finishes in the money. A 0.30-Delta call has roughly a 30 percent chance of expiring in the money; a 0.16-Delta put, roughly 16 percent. The approximation is imperfect, since Delta and the true model probability differ by a drift-and-volatility term and skew distorts the mapping in the tails, but for strike selection it is accurate enough to anchor decisions, and it is sitting right there in the chain.
This immediately turns vague trade ideas into specific ones. "Sell a put below support" becomes "sell the 0.20-Delta put": a strike the market itself says has about an 80 percent chance of expiring worthless. It also standardizes across underlyings and volatility regimes. The 0.20-Delta put might be 4 percent below spot on a quiet index and 12 percent below spot on a volatile single name: same probability, very different distance, and the Delta is the honest common denominator. When a strike that used to sit 3 percent away now sits 6 percent away at the same Delta, the market is telling you expected movement has doubled. Listen to it.
The expected move is the size of underlying move the options market is pricing in over a given period, and you can read it straight off the chain: the price of the at-the-money straddle is a good approximation of the expected move through that expiration. If the underlying trades at $500 and the 30-day at-the-money straddle costs $19, the market expects roughly a ±$19 move, about 3.8 percent, over the month, in the one-standard-deviation sense: about a two-thirds chance of finishing inside that band.
Figure: The expected-move cone: the band widens with time, so a short strike comfortably outside this week's range can sit inside the market's range by expiration.
Use it as a ruler for every strike decision. Selling a call spread with the short strike at $510 when the expected move is ±$19 means selling inside the market's own one-standard-deviation range, so you are being paid well because you will be tested often. Selling at $525 places you outside it: tested rarely, paid little. Neither is right in general; what matters is that you chose the strike knowing which trade you made. Buying options works the same way in reverse: a long call needs the underlying to beat the expected move for the market to have underpriced it, so a thesis of "it will go up a bit" is not enough, because "up a bit" is already in the price.
Here is the trap that catches almost everyone. A trade can win 85 percent of the time and still lose money, because expectancy weighs each outcome by its size. Take a put credit spread: short the 0.15-Delta put, $5 wide, collecting $0.55. Win rate is roughly 85 percent. The arithmetic: 0.85 × $55 = $46.75 expected gain, versus 0.15 × $445 = $66.75 expected loss on the losers if they run to full width. Raw expectancy: −$20.00 per spread before commissions. The high win rate is real. The trade still bleeds.
Management changes the arithmetic, and you should model it explicitly rather than assume it. If you close losers when the spread doubles in value, the loss per loser drops from $445 toward roughly $110, but you also get stopped out of some trades that would have recovered, so the win rate falls, perhaps from 85 percent to 70 percent. New expectancy: 0.70 × $55 − 0.30 × $110 = $38.50 − $33.00 = +$5.50. Slim, and honest, and it shows exactly where the profit came from: not the entry, the management rule.
Hope picks the strike that pays the most. Probability picks the strike where your edge, if you have one, actually lives.
Underlying at $500, 30 days out, expected move ±$19. You want to sell a put spread. Candidate A: short the 490 put at 0.30 Delta, long the 480, collecting $2.40, for about 70 percent probability of maximum profit, risking $760 to make $240. Candidate B: short the 475 put at 0.15 Delta, long the 465, collecting $1.05, for about 85 percent probability, risking $895 to make $105. Candidate A is paid $2.40 to stand inside the expected-move band; B is paid $1.05 to stand outside it. Raw expectancy is negative for both, as it should be, since a fairly priced chain plus friction yields slightly worse than zero. The decision is therefore not "which looks better" but "which one does my edge apply to": a view that implied volatility is 2 points too rich favors A, where more premium is at stake per spread; a purely mechanical income approach with disciplined loss management is usually less bad at B, where the short strike is tested far less often. If you cannot name the edge, the honest answer is neither.
Before any entry, write down four numbers: the short strike's Delta, the expected move to expiration, the probability of profit, and the expectancy under your actual management plan. It takes two minutes with a chain and a modeler. Then let your journal keep score: if your realized win rate runs materially below the Deltas you sold, your fills or your management are leaking; if the win rate holds but the account does not grow, your losers are oversized, and expectancy, not probability, was the number you ignored. Hope picks the strike that pays the most. Probability picks the strike where your edge, if you have one, actually lives.
Educational content, not investment advice. Options involve substantial risk.