Hacker News Reader

Comment by hso9791 | Hacker News Reader

Thinking is hard. That's why we must strive to avoid it at all costs. :)

Probabilistic thinking is useful to wrap one's head around.

Having worked with decision support for a decade, I've come to realize that probabilistic thinking may or may not be useful to work into organizational decision making regarding change.

It's still a thought experiment, but I'll put in a shameless plug for the first of a series of blog entries: https://blog.henrikscorner.net/2018/02/25/a-modest-proposal/

In case of concerns: The blog itself is unmonetized.

View on HN

The thing is people need certainty. Imagine a guidance from Apple saying they would make $2 per share with 47% probability. Investors will freak out. Which is why most companies sandbag their results. I want to know if any orgs actually use probabilistic decision making in their day to day operations.

That is an excellent point!

That is the very reason I believe the current system should be augmented with a new first class financial statement, rather than replaced. This will give investors better information, and managers will gain better decision support.

Those who prefer the old way of doing things remain free to carry on as before. Those who embrace it may compete better.

Of course, the same mechanism may prove useful for governmental organizations.

I forgot to answer your question: "I want to know if any orgs actually use probabilistic decision making in their day to day operations."

The short answer is "yes!" The long answer is, well, rather lengthy.

Check out the fields of advanced analytics and BI.

Imagine a guidance from Apple saying they would make $2 per share with 47% probability. Investors will freak out.

So? People freak out all the time about things that are true. It doesn't change that it's true.

Strictly speaking, one may end up with a separate set projections for e.g. dividends. These may in a standardized way present ranges for each KPI with confidence intervals associated.

I'd rather invest if that type of data were included. :)

As this is a somewhat large topic, I'll elaborate further throughout the series of blog posts.

> Imagine a guidance from Apple saying they would make $2 per share with 47% probability. Investors will freak out.

if that's all it said, sure, because that sounds like apple just made some insanely high-variance bet. if instead apple provided some nice smooth probability distribution over possible earnings, folks would react much more reasonably.

One big problem with that is that the uncertainty in the variance is probably larger than the uncertainty in the mean.

Reply on news.ycombinator.com