NEW YORK ( TheStreet) -- The market has been pulling back recently and biotech stocks have not been immune to the selling. Rather than focus on a single stock in this week's column, I want to write more broadly about valuation and how it relates to expected utility theory. In general, expected utility theory argues that individuals act to maximize the difference between expected benefits minus expected costs. While this seems intuitive, it highlights some interesting aspects of biotech investing.

Let me explain using an example of a simple coin-flip game in which a coin landing heads earns the player \$100 while tails returns nothing. The expected benefit of the game is \$50, calculated as the odds of heads times the payout of heads (0.5x\$100=\$50) plus the odds of tails and the payout of tails (0.5x\$0=\$0.) Obviously, this looks like a great game but it ignores the expected costs. If I am charging to play this game, how much would you pay?

If the game cost \$20, for instance, then you should certainly play as the expected utility is \$30 (expected benefits of \$50 minus the expected costs of \$20.) As long as the cost of the game is less than \$50 the player has a positive expected utility and should play. Obviously, one could generate a number of different game structures but the basic analysis is the same.

So how does this relate to biotech investing?

When investing in a clinical stage biotech company, investors are implicitly playing an analogous game. There is a probability of success (regulatory approval), expected payouts from success (discounted cash flow), probability of failure (FDA rejections), payouts from failure (additional costs), and costs for playing the game (share price.) I'm simplifying things, of course, but it's still a good heuristic approach.

Take, for instance, a biotech company that has a 50 percent chance of drug approval and the discounted cash flow on approval is expected to be \$100 per share. To keep it simple I will assume that the costs of failure (drug rejection) are \$0. Given these assumptions, the expected benefit from investing in this company is \$50 (0.5x\$100 + 0.5x\$0).

Is this biotech company a good investment? Like with the coin flip game, the answer depends on the cost -- in this case the price of the biotech company's stock. A share price below \$50 would mean the company is undervalued and therefore a good investment. If the current stock price exceeds \$50, the company is overvalued.

Of course, biotech investment decisions are not nearly as simplistic as my example, mainly because the nailing down the correct probability of a drug approval and the value of that drug approval (the discounted cash flow) is a significant challenge. In reality, these numbers have to be estimated with the associated errors.

Even setting that aside, this sort of analysis highlights two critical points that biotech investors should always keep in mind. First, a company could have a very low probability of success and be undervalued. For instance, one could assume a 2 percent chance of success with a \$100 DCF, which creates an expected return of \$2. If the share price is under \$2, then an investment has a positive expected utility (profit) even though you expect the drug to ultimately fail. In many ways, I think this is what many Biodel ( BIOD) bulls argue: The odds of success for its rapid-acting, mealtime insulin are low but the stock is cheap enough to make it a good investment.

Second, a company with a high probability of success can be overvalued. For instance, one could assume a 98% chance of success with a \$100 DCF, which generates a \$98 expected benefit. If the price is over \$98 then the company is overvalued. A good real-life example is the bear thesis on Seattle Genetics ( SGEN): The Adcetris label will be expanded but the company's current valuation more than takes that into account.

I've simplified discussion of how expected utility theory plays a role in biotech investing. Clearly, the method can be made much more complicated with the introduction of more variables. The key take-home point is that not all biotech companies with a high probability of success are good investments, nor are all biotech companies with a low probability of success bad investments.

Perhaps the only truism is that holding all else constant, as the share price decreases the more undervalued a company becomes. The key there is "hold all else constant" because you need to be sure that the falling share price is not caused by changing probability of success or lower commercial opportunity.

Sobek is long Seattle Genetics.
David Sobek has been writing on biotech for a number of years through various outlets with a general focus on small cap oncology and antibiotics companies. He received his PhD in political science from Pennsylvnia State Univeristy in 2003 and a BA in international relations from The College of William and Mary in 1997.