## Because I am all about the base rate

Tis the season for predictions. If one has an audience one seem compelled to make predictions.  You are better off reading the book Superforecasting than this article. The book explains in depth the simplest elements you need in making predictions and forecast.

It starts with – Base Rate – which is how frequent does the said event happen in general relative to all other events. For example

1. What percentage of tweets are retweeted?
2. What percentage of people are named Bill?
3. What percentage of startups achieve \$1B valuation?
4. What are the chances of you winning Survivor when you start the season with 19 others?

The next step is an iterative process that refined this prior knowledge by seeking new information and refining your estimate. That is the posterior probability.

Most likely you won’t read the book, so I present here these two concepts set to the tune of Megan Trainor’s song.

Because you know I’m all about that base-rate

‘Bout that base-rate, no tails

‘Bout that base-rate, no tails

‘Bout that base-rate, no tails

‘Bout that base-rate… base-rate… base-rate… base-rate

Yeah, it’s pretty clear, I ain’t no sigma two

But I can predict it, predict it, like I’m supposed to do

‘Cause I got that Bayesian that all the gurus chase

And all the right tunables in all the right places

I see the magazine workin’ that Crystalball

We know that shit ain’t real, come on now, make it stop

If you got logic, logic, just raise ’em up

‘Cause every inch of you is curious from the bottom to the top

Yeah, my mama she told me “don’t worry about your data size”

(Shoo wop wop, sha-ooh wop wop)

She says, “Bayesians like a little more posterior to hold at night”

(That booty, uh, that booty booty)

You know I am wont to be stick figure xkcd comic doll

So if that what you’re into, then go ‘head and move along

## Prepare for the Onslaught of Predictions

This is that time of the year when we will see magazines, blogs, thought leaders and their ilk polish their crystal balls, look deep into it and make predictions. Predictions that range from big to extremely big, outrageous to ridiculous, all trying to outdo others in clairvoyance. The more ridiculous the prediction higher the number of page views, sharing and Retweets. And more such a prediction is shared the more believable it comes, even becoming a foregone conclusion.

Popularity and pervasiveness alone come to define the quality of predictions, with no one stopping to check the likelihood of the event or the reasoning of the guru making the prediction. Never mind none of the predictions made in the past year or the years before borne out to be true. There is is always weasel language to get out of it. Or who remembers, who checks or who cares?

This year you may do well to simply ignore all such predictions. But I recommend a better options. Before you read any of the predictions read this recent book on science of prediction by  Berkeley and Wharton professors titled, Superforecasting. The concepts and methods explained in this excellent read are meant to help you make better predictions but they also help you evaluate predictions from others.

Armed with concepts like understanding of base rates, conditionals, likelihoods and outside-in process you can see how most predictions can be ignored regardless of pedigree of who made it, their popularity or even if one or two turn out to be true due to sheer luck.

Take for example the following prediction from Fortine magazine,

Apple has announced plans to build an electronic car, targeting 2019. Apple could dramatically accelerate this timetable by buying Tesla  TSLA . With over \$200 billion cash on hand, the iPhone-maker has more-than ample resources to absorb the purchase, especially now that some of the bloom has come off Tesla’s once-rosy stock. In addition to its automobile know-how, Apple gets access to Tesla’s battery technology, which CEO Elon Musk claims can help change “the entire energy infrastructure of the world.” Of course, Apple would also get Musk—a worthy heir to Steve Jobs’ “think different” legacy and ideally suited to be Apple’s futurist, chief technologist and CEO-in-waiting.

You notice here there are no timelines so we have to assume this is meant for 2016. There is also no likelihood based metrics, only certainties – “Apple will acquire Tesla”. If Fortune magazine bothered to look at base rate for acquisitions by Apple in the past 5 years, there is none even close to the size of Tesla’s \$30 market cap. Beats acquisition for \$3 is one tenth of Tesla’s current market cap and it is not really about Apple’s launch of Apple music.

The biggest problem with this prediction is it is inside-out, it started with the idea and looked for specific aspects that will make the idea true without looking for what would make it false.  That is,

• Apple wanted to enter electric car business (we know this only from rumors)
• Apple has enough resources (base rate of M&A will show you really don’t need all the resources to make an acquisition)
• Tesla stock is looking attractive (neglecting the fact that despite current valuation Tesla owners will demand hefty premium if at all they want to sell)
• Apple gets access to Tesla’s battery tech (neglecting there are no other ways or options)
• Capping it all, getting Elon Musk as CEO in waiting for Apple (you can see the flaws here)

However the prediction is outrageous, sounds plausible and seems like something pundits and avid fans want it to be true. So this will get traction. But that does not  make it a solid prediction.

If you give even a cursory review of all the 2016 predictions you will see how woefully unrefined they are.

## I don’t know about your conclusion but how you arrived at it is flawed

It has become common place in the tech and other marketing blogs – popular bloggers  making bold predictions about the future of a firm or about a technology. They state it with utmost confidence without any uncertainty. All black and white, no room for gray areas.

They back it up with scant evidence which is mostly selectively chosen to support their preconceived notion. Or these evidences are anecdotal or manufactured based on their own biased reading of the market. They fail to seek any evidence to the contrary or refuse to ask questions on what evidence will disprove their theory.

On top of this poor evidence seeking behavior we see even poorer analysis. All set up to support their foregone conclusion.

The latest in that mold comes a blog post that boldly proclaims how an enterprise software business that

• hauls in \$35 billion in revenue
• has \$29 billion in cash
• has almost limitless borrowing power
• has a market capitalization of \$142 billion

is really doomed.

So writes Sarah Lacy in her blog.

I am not going to argue with the conclusion. I do not know what the future holds. But let us raise some questions on her  analysis (if we can call that):

1. What does doomed mean? \$35 billion in revenue wiped out? Cut in half?
2. What are the assumptions and preconditions for this outcome? Remember, a model is only as good as its assumptions.
3. What time frame are we talking? 5 years? 10 years? 20 years? Did she consider how long will it take to lose any portion of the \$35 billion revenue or any of the \$29 billion cash? What other technology and market shifts could happen during that period?
4. One way to make a prediction is to run a scenario analysis and state in what percentage of the scenarios (and under what conditions) a particular outcome is possible. An example of such an analysis is the one by Jeremy Siegel when he predicted Dow will hit 15000. Has she run through such an analysis? What are the alternative scenarios and how likely are they?
5. Let us consider a case where the revenue drops by 10%. Will the decision makers stand still and let the slide continue all the down to 50% and finally to 0?
6. How many millenials will go on to install what she calls as skunkworks implementations? And if they do, will the business in question standstill and not use their cash pile and borrowing power to buyout these smaller solution providers?
7. The story  is titled,   “\$142 billion market cap business is really doomed”, backed by anecdotes quoted from selective memory and adorned with flashy narratives sounds plausible, but is it probable?

If the analysis is flawed or if the analysis is just a figment of one’s imagination, if you and I cannot reproduce the same results with the data set (in this case there is no data set) we have to reject the conclusion regardless of how plausible it sounds or how popular the blogger is.