Skew is used to measure how symmetric or asymmetric the distribution of ** x** is. And in a normal distribution, the skew is 0, meaning that the distribution is perfectly symmetrical.

Whether retail traders realize it or not, skew is essentially what they are always betting on whenever the jump in the market with a stop loss and take profit, and wait for either to be hit.

This is because, when you enter the market with a fixed stop loss and take profit, you are flattening the distribution of outcomes into a binary one. And by doing this, you are completely removing time from the equation, along with any benefits or disadvantages it may provide in the form of volatility or kurtosis.

In this setup, you are isolating the trade so the only thing that matters is the probability of moving up or down. This appeals to a lot of traders because it makes things simple, but it’s what causes most traders to get chopped to bits.

Unfortunately for retail traders, skew is probably the hardest thing to build your strategy around because on a tick by tick basis, the market is pretty 50/50 for most assets.

Traders often see large swings in price, which create the illusion of a strong skew, but in aggregation, the market’s skew is typically hovering around 0. And any deviations from 0, are never large enough to cover the bid-ask spread.

Take a coin toss for example. In a coin toss, there are only 2 possible outcomes, heads or tails, and they are equally distributed. This means over time, you can expect to win and loss the same amount of time as you continue to flip the coin.

No side has an advantage over the other, so it doesn’t matter which side of the coin you call to win.

This fairness exists in any single toss and in the aggregation of a large amount of tosses.

For example, say we were going to draw a path of 5 steps, flipping a fair coin to determine if we move up 1 step or down 1 step.

These paths will almost always land above 0 or below 0, rarely ever landing on 0.

BUT if we did this for 1000 paths, roughly 500 paths would be above 0, and 500 paths would be below 0. This is how the fairness of each step realizes itself in aggregation.

This path generation example is essentially how financial market prices are determined. Candles are drawn with ticks from the market (orders transacting across the bid/ask spread), and any one tick, theoretically, has a 50% chance of going up or down.

The aggregation of these ticks over a fixed length of time is what determines a candle’s open, high, low, and close values.

Any candle or group of candles can create a picture that looks like price is heavily skewed in one direction over the other, but in aggregation, there are typically an equal number of candles or groups of candles trending upwards as there are trending downwards.

This is why building a strategy around skew can be very challenging. It doesn’t matter how you slice the market into the binary outcomes you want to trade, the corresponding probabilities of ** x** will almost equal the ratio of a linear

**payoff structure.**

*f(x)*