Coincidentally, I came across another example of someone doing bad stats in a piece lamenting innumeracy. In Now for some jaw droppingly bad analysis the author accuses another author of innumeracy regarding the recent Snapchat bid rejection. He then takes that other author’s out-of-the-air number, “imagines” it’s an order of magnitude lower, and does some simple, bad math with it.
Snapchat is a company that turned down a $3B buyout offer. The author supposes Snapchat could be thinking that someday it will be worth $30B and so that means there must be at least a 10% chance of Snapchat reaching a $30B valuation to justify turning down $3B now.
Putting aside discount factors, it’s the probability distribution of future valuations that matters. This does not mean simply guessing the probability of achieving an order of magnitude valuation increase.
I understand that simplifying illustrations can be helpful and even necessary, but simplifications should always enhance understanding. In this case, the simplifying example is built on faulty framing using an analysis that is too far removed from any valid approach. (I appreciated CL53‘s comment in the original post.)
Aside: I guess part of blogging is doing that kind of foot-in-mouth thing and I’m sure I’ll do it. Not sure if anyone will know between the typos, grammar and the “does a falling tree make a sound if nobody’s there to hear it” thing.