Jamie Mai: Seeking Asymmetry

Mai’s investment strategy has five main pillars:

1. Finding mispricings in a conceptually priced world. Prices, specifically for derivatives, are based on standard pricing assumptions which may be inappropriate according to the prevailing market conditions. When these assumptions are unjustified, it creates mispricings and trading opportunities. Mai strives to identify these trade opportunities.
2. Selecting trades in which the probability is considerably inclined to a positive result. As a rule, if the trade succeeds, Cornwall expects that the estimated gain multiplied by the probability of a positive result must be at least twice as large as the estimated loss. These profits and losses and their probabilities must be based on subjective measures. The rigid criterion for qualifying trades will lead to a concentrated portfolio. Generally, Mai has only 15 to 20 independent risks at any one time. Although his portfolio is very concentrated, he constructs his trades asymmetrically to constrain the downside when he is wrong.
3. Implementing trades asymmetrically. Mai structures a trade where the downside is limited and the upside is open-ended. One broad way of achieving such a return/risk profile is by being an option buyer (of course, when there is a mispricing).
4. Waiting for high-conviction trades. Mai is patient enough to wait for a trading opportunity that meets his guidelines. This trait improves the return/risk of a trade.
5. Using cash to target portfolio risk because the portfolio mostly consists of derivatives. Mai holds a large cash component in the portfolio (commonly, 50% to 80%). He targets his desired risk level by altering this cash component. 

Option prices are determined by pricing models using certain assumptions. These models provide relatively good approximations. However, sometimes those assumptions are inappropriate for a prevailing market situation. Mai recognized five such assumptions that are not valid at times.

1. Normal distribution of prices- Options pricing is based on the assumption of a normal distribution. This indicates that the probability of extreme moves on either side is abysmally low. However, in some cases, large price moves are more likely than suggested by the normal distribution. In such situations, options are likely to be mispriced, especially OTM calls/puts. This mispricing creates profit opportunities. Another implication of a normal distribution is that it assumes equal probabilities for an up move/down move. Although this assumption may be acceptable, there are times when a market is more likely to go up by a given percentage than down by the same or vice versa. In such situations, the odds of a big price upswing are quite greater than the odds of an equivalent decline. The conclusion is that many times the market price is wrong and such situations provide trade opportunities.

2. The forward price is a true forecaster of the future mean— This assumption implies that options are going to be priced with probabilities centered at the corresponding forward price. However, sometimes when the forward price is far from the present price, it should not be reasonable to assume that a price change adequate to the difference between the forward and damage is the presumably market outcome. Frequently, there is also good reason to assume that some price between the forward and price is more likely than the forward price. If this can be true, out-of-the-money options (puts if the forward price is higher and calls if it's lower) could also be underpriced.

3. Volatility is a measure of the square root of time—This assumption in option pricing is logical only for short periods. For longer time intervals this assumption may understate probable volatility, especially if current volatility is low, for two reasons. First, the longer the period, volatility will be more likely to return to the mean from current low levels. Second, longer periods allow more trend opportunities that result in larger price moves than indicated by the volatility assumption.

4. In calculating volatility, the trend can be avoided—Option pricing models calculate the probability of price moves based on time and volatility. Trend is not part of the calculation. The implied belief is that the daily price direction change is random. Hence, a trending market can result in price moves that would be considered unlikely by the pricing model. If there is a reason to anticipate a trend, then OTM options will be underpriced. 

5. Existing correlations are good predictors of future correlations—Some market correlations like gold and platinum, tend to be fairly consistent, whereas other market pairs like the Australian dollar and Swiss franc may show inconsistent correlation patterns. Trades based on correlation will tend to presume that future correlation will be equal to the past correlation. Such a belief may be invalid for markets that have a varying correlation.

One of the great myths is relating risk with volatility, which is wrong for several reasons. First, the most important risks don’t reflect in the track record and hence are not shown by volatility. For example, a portfolio of illiquid positions may have low volatility during a risk-on period. However, it may have a large risk if market opinion shifts to risk-off. The other side of the coin is that large, abrupt gains may sometimes increase volatility but the theoretical risk is limited. 

Flexibility is one of the distinctive features of Market Wizards. Mai habitually changes his view as demanded by the research. A good trader will always get out of a position when he realizes a mistake. Great traders are skilled at taking an opposite position when they discover their original idea was wrong.