One of the better tools for
assessing risk/reward, and a corresponding probabilistic outcome “yield” is the
Probability Tree.
The way that a Probability Tree works is that the top of the tree you break into branches the different decision forks that you are choosing between, the most likely outcome forks these choices will yield, and finally, the probability that a given outcome will occur for each particular fork.
By multiplying the raw returns that a given path is worth by the probabilities of realizing the particular return, you get a picture of the expected yield for each decision path.
As outcome forks can themselves be further broken down into sub-fork outcomes (each with their own probabilistic outcomes), the probability tree is a relative sophisticated, but intuitive, way of modeling out decisions, outcomes and expected yields.
Should I Stay or Should I Go: Evaluating a Retention Bonus
Case in point, several years
ago, I was working for a startup where a retention bonus had been presented to
me that, on face value, seemed unacceptably weak.
Contemplating my options pragmatically, I saw decisions like “accept offer,” “reject offer,” and “negotiate offer,” the collective of which yielded potential sub-outcomes (with weighted yields), like “receive full bonus,” “receive 50% of bonus,” “receive 150% of bonus,” “lose job – get no bonus” and “leave company – negotiate 125% of current comp at new company.” On top of this, there was a whole set of sub-branches of what the money in hand might yield me under different investment scenarios.
Suffice it to say, this exercise not only helped me avoid making an emotional, potentially polarizing, decision, but it clued me into the fact that there was a high-risk, high reward option that I hadn’t fully considered, which happened to fit my risk profile (at the time).
Yossi Vardi: Properly 'Weighting' the Success Rate of Entrepreneurs
I was thinking about
Probability Trees the other day when listening to a Q&A session with Yossi
Vardi, one of the ‘godfathers’ of Israeli high-tech – as an investor, advisor
and/or board member to more than 40 startups, including Mirablis (makers of IM pioneering software/service, ICQ, acquired by AOL in 1998).
When asked to assess the prospects of a given startup as an investment, Yossi first stated that, “I can never predict if products will succeed so I bet on people,” adding that the entrepreneur behind the specific company being discussed was exactly the kind of person he would like to invest in.
When pressed further about whether that meant that he would be a motivated investor in this particular deal, Yossi offered up some Probability Tree-like logic, noting that the hard data on entrepreneurial success and failure rates presents some interesting facts that mitigate his answer.
One, the data shows that the ‘Success Rate’ of an Inexperienced Entrepreneur is only 23%, while the Success Rate of a Failed Entrepreneur is 24%, only 1% better than the first-time entrepreneur.
However, an Entrepreneur Who Has Succeeded Before (or multiple times) has a 34% probability of succeeding in their next venture; data which, at face value, suggests a staggering 50% better expected outcome than either the first-time entrepreneur or the failed entrepreneur who is trying again.
But, here’s where it gets more complex. As Yossi noted, since an experienced, successful entrepreneur can drive materially higher valuations, this dramatically offsets the expected yield in backing them (all things being equal) since you are typically paying a significant premium for the lower risk of backing a proven entrepreneur. Hence, Yossi's answer was that it depends on the deal price.
Netting it out: even a gold standard entrepreneur’s value is relative to a projected, weighted outcome, something that I have seen many an investor forget in backing a proven winner, largely blind to price.
Related Posts:
