There is a reason cheap options feel like they should work more often than they do. Each individual condition sounds reasonable on its own. It is the conjunction that kills them.
Three conditions, not one
A far out-of-the-money call needs all of the following to be true before expiry:
- Direction — the stock has to move up rather than down.
- Magnitude — it has to move far enough to cross the strike, plus the premium you paid.
- Timing — all of that has to happen before the contract expires.
Miss any one and the contract is worth zero. Not reduced — zero.
Compare that with a slightly in-the-money contract, which mostly needs condition one, benefits from condition two, and is fairly forgiving about condition three.
| Far OTM — Δ 0.15 | Slightly ITM — Δ 0.65 | |
|---|---|---|
| Direction | required | required |
| Magnitude | required | scales profit, not required |
| Timing | required | tolerant |
| Partial credit | none | yes |
| Outcome if two of three hit | total loss | — |
| Outcome if move is half-size | — | roughly half the profit |
That bottom row is the crux. The in-the-money contract pays partial credit. The out-of-the-money contract does not — it is closer to pass or fail.
Why conjunctions are brutal
Suppose, generously, that you are right about direction 60 percent of the time, that when you are right the move is big enough 50 percent of the time, and that it arrives in your window 60 percent of the time.
Those are decent individual numbers. Multiplied, they give roughly 18 percent.
The exact figures do not matter — your own will differ. What matters is the shape: requirements multiply, they do not average. Three conditions at 60, 50 and 60 percent do not produce something in the fifties. They produce something in the teens.
This is the arithmetic behind the market's pricing. The GOOGL 350 call at eight days carries a delta of 0.150, which is roughly the market's estimate that it finishes in the money at all. The chain is not being pessimistic. It is doing the multiplication.
So why buy them at all
Because when all three conditions do hold, the payoff is not slightly better — it is categorically different. Here is the GOOGL ladder at eight days, by move size.
| Strike | Δ | +$1 | +$3 | +$5 | +$10 |
|---|---|---|---|---|---|
| 315C | 0.894 | +5% | +15% | +25% | +50% |
| 330C | 0.590 | +8% | +26% | +45% | +97% |
| 340C | 0.325 | +11% | +36% | +64% | +150% |
| 350C | 0.150 | +14% | +44% | +81% | +203% |
Across that row, the 350 call turns a ten dollar move into 203 percent while the 315 call manages 50 percent. That gap is not linear in delta, and it is not an accident of pricing.
Convexity, mechanically
The mechanism is gamma, and it is worth stating step by step rather than as a slogan.
Start with the 350 call at delta 0.150. The stock rises three dollars. That contract's delta does not stay at 0.150 — it rises, because the strike is now closer and more plausible. Say it becomes 0.24.
The next dollar of movement is therefore captured at 24 cents rather than 15. If the stock keeps rising, delta keeps climbing — 0.35, 0.50, higher. Each dollar is captured more efficiently than the one before.
Price rises → delta rises → the option responds harder → delta rises againThat feedback loop is convexity. It is the entire reason out-of-the-money options can produce returns that look disproportionate to the move that caused them.
And it runs in reverse. If the stock stalls, delta drifts down, the contract responds less to any eventual move, and theta — already nearly 20 percent of premium per day at that strike — keeps collecting. Convexity is not a free option on good outcomes. It is a mechanism that amplifies whatever actually happens.
Why does gamma peak near the money rather than far out?
Gamma measures how fast delta changes, and delta changes fastest where the outcome is genuinely uncertain — right around the strike, where a small move flips the contract between worthless and valuable.
On the GOOGL eight-day ladder, gamma runs 0.0122 at the 315 strike, peaks around 0.0276 near the money, and falls back to 0.0149 at 350. The far out-of-the-money contract does not have the most gamma in absolute terms.
What it has is the most gamma relative to what you paid. The 350 call's gamma of 0.0149 sits on a $1.18 premium; the 315 call's 0.0122 sits on $18.73. That ratio is why convexity feels explosive out of the money even though gamma itself is lower there.
The honest use case
There is a legitimate way to use the bottom of the ladder, and it follows directly from the three conditions: buy it when you have specific evidence for all three, not just the first.
Evidence for direction is what most analysis produces. Evidence for magnitude means something in your process forecasts a move of a particular size — a measured move, a range expansion, a level with room above it. Evidence for timing means something says now rather than eventually — momentum expansion, a completed break, volume confirmation.
If you have all three, your probability estimate genuinely differs from the chain's generic 0.15, and the contract may be correctly priced for you but not for the market. That is the definition of an edge.
If you have only direction, you are accepting the market's odds on the other two while paying for the privilege of needing them.
The reframe
Contract selection is really a question about how many things you are willing to need.
- Need one thing: high delta, comfortable DTE. Lower ceiling, far more reliable.
- Need three things: low delta, short DTE. Much higher ceiling, and each additional requirement multiplies against you.
Neither is wrong. What is wrong is needing three things while having evidence for one, which is the most common way a correct directional call still ends at zero.
Delta as a second opinion
There is a useful habit buried in all of this: read delta as the market's estimate, then compare it with your own.
Delta is approximately the probability the contract finishes in the money. So every strike on the chain is the market quoting you odds.
| Contract | Δ | Market's implied odds |
|---|---|---|
| GOOGL 330C 8D | 0.590 | About 3 in 5 |
| GOOGL 340C 8D | 0.325 | About 1 in 3 |
| GOOGL 350C 8D | 0.150 | About 1 in 7 |
Now hold your own analysis against it. If you believe GOOGL reaches 350 within eight days, you are claiming the true probability is meaningfully above one in seven. That is a strong claim, and it should be supported by something specific — a measured move, a catalyst, a range expansion with room.
If you cannot articulate why your estimate beats the market's, you are not finding value. You are accepting the market's odds while paying a spread for the privilege.
This also explains why low-delta contracts feel like they "almost work" so often. A one-in-seven shot that comes close is still a loss, and the near-misses accumulate in memory much faster than the arithmetic does.