According to most books the ATM option is the option with a delta of 0.50. However, this is only the case when the distribution is normal. The more positively skewed the distribution, the further the 0.50 delta option is out-of-the-money (for calls). According to the following article, the formula to calculate the 0.50 delta option strike is equal to:

S x e^(σ^2/2)

I want to know why this is exactly the case. Looking at the delta defintion I have:

delta = N(d1) = 0.50


d1 = 0



So, how do I get from this well-known formula to the above mentioned formula? Thanks in advance,


2 Answers 2


The delta you mentioned is the Black-Scholes delta. If you let $r=0$, $T=1$ and solve the equation $d_1=0$, you get what is in the article.

  • $\begingroup$ $0=\frac{\ln(S/K)+(0+\sigma^2/2)*1}{\sigma*1}$ gives $0=\ln(S/K)+\sigma^2/2$ gives $-\ln(S/K)=\sigma^2/2$ gives $S/K=e^{-0.5 \sigma^2}$ finally gives $K=S e^{0.5\sigma^2}$ $\endgroup$
    – nbbo2
    Commented Jun 9, 2020 at 8:48

As you point out in your d1 formula:

$$d_1 = \frac{ln \left( \frac{S}{K} \right)+\left(r+0.5\sigma^2 \right)T}{\sigma \sqrt{T}} $$

Therefore, $N(d_1)$ (where $N(.)$ stands for the Standard Normal CDF) is only equal to half when $d_1$ is exactly zero. When an option is ATM, then $S=Ke^{-rT}$. So $N(d_1)$ won't be exactly 0.5, because:

$$d_1 = 0.5\sigma\sqrt(T)$$

For short dated options, $N(d_1)$ of the above will be close to 0.5, whilst for longer-dated options (like 10-year expiry) it will be higher than 0.5.

Indeed, if you set: $S=Ke^\left(-0.5\sigma^2T-rT \right)$, you will set $d_1$ to zero.

People who say that:

(i) $N(d_1)$ for ATM options is exactly half

(ii) ATM option has $N(d_2)$ equal to half because $N(d_2)$ is the probability that the option will end up in the money

Are (in my experience) mostly option traders who lack the technical knowledge to understand how option pricing works. $N(d_2)$ is the risk-neutral probability, so has nothing to do with "likelihood" or "real-world probability" as we humans like to interpret probability. "Risk-neutral" probability is a mathematical construct invented for pricing options.

  • 2
    $\begingroup$ $N(d_1)$ is not the standard risk neutral probability of the event $\{S_T\geq K\}$ though. This is $N(d_2)$. You need a change of numeraire. $N(d_1)$ correspond to the exercise probability under the stock measure which uses $S_t$ or $S_te^{qt}$ as numeraire. $\endgroup$
    – Kevin
    Commented Jun 9, 2020 at 8:37
  • $\begingroup$ @KeChen: you're totally right, careless mistake. $\endgroup$ Commented Jun 9, 2020 at 8:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.