The world changes; some people don't get that

I read something by a smart person that basically said: "90% of attempted X have failed, therefore X won't be viable."

The problem is that what that person really should have said was: "90% of the attempted X up to now have failed, therefore unless the underlying conditions have changed, X won't be viable."

Here's an example, adapted from a real situation in the past, with disguised (and simplified) numbers; smart person would have spoken up in year 20 and said: "people have been trying that for 20 years now, and trying in ever increasing numbers of attempts, and over 90% of the attempts have failed."

(click for bigger)

And while that's true, by the year 25, sixty-eight percent of all attempts made that year succeed! The secret is a change in underlying conditions (could be infrastructure, social acceptance, technology, economics, or other change).

The details of the model are the following:

1. Number of attempts in year $N$ is $A[N] = 2^{N-1}$.

2. Given a probability of success $p[N]$ in year $N$, the number of successes is the expected number (so we get rid of any stochasticity, for simplicity only):

$S[N] = p[N] \, A[N]$.

3. The probability of success in year $N$ is given by a deterministic difference equation,

$p[N+1] - p[N] = (a + b\, p[N]) (1 - p[N])$,

where the parameters $a$ (innovation rate) and $b$ (imitation rate) are between zero and one, with the specific values for the model above being $a=0.00001, b=0.6$, and $p[0] = 0$.

In other words, very slow innovation, but a large imitation rate (which leads to slow adoption at first then a cascading effect at some point, before saturation sets in). 

[The general form of this process is called a Bass model of evolution (used to analyze product diffusion in the past); I don't like it for data analysis or decision support (for a variety of technical reasons), but it's a convenient model for simulation when we take the parameters as given.]

Here's a real-world example of things changing, possibly one of the reasons of the current interest (institutional as well as individual) in cryptocurrencies:

But some people have this blindness: that just because something has been tried before and hasn't worked then, it must never work. And that something could be SpaceX landing its boosters, or the feasibility of Bitcoin, electric vehicles, small nuclear reactors, etc.

Fear is the mind killer. And not just for scions of House Atreides.

Fun with numbers for October 4, 2019

It's flu season, let's talk product diffusion

One of the classic marketing models people learn in innovation classes is basically a SIR(1) model without the R part: the Bass model of product diffusion.

The idea is that some fraction $a$ of the consumers are "innovators" who adopt a product without social pressure, while another fraction $b$ are "imitators" who adopt a product when they see others with it. The fraction $x$ of the market that has adopted the product at a given time is given by the following differential equation

$\dot x = (a  + b x)(1-x)$, 

and the behavior looks like a traditional product life-cycle curve (an S-shaped curve):

The process for a viral infection is similar: some people get the virus from the environment (those would be the $a$ fraction), some get it from contact with other people (those would be the $b$); the infection process has a third element, recovery, which we ignored here.

Growth confusion and punditry, part 1

Pundits throwing around growth numbers seem to be unaware that there are significant differences even with very small growth numbers.

Growth confusion and punditry, part 2

A pundit: "it's important to get the economics high-growth first, so that the slower growth starts from a higher number." (Paraphrased.)

Me: Gah! Multiplication is transitive. The order doesn't matter, what matters is that the high-growth period be the longer period.

Consider two periods, with $t_1$ and $t_2$, with associated growth rates $r_1$ and $r_2$. Starting from some value $x_0$, the result of period 1 before period 2 is:

$\left( x_0 \, e^{r_1 t_1} \right) \, e^{r_2 t_2}$,

and the result of period 2 before period 1 is

$\left( x_0 \, e^{r_2 t_2} \right) \, e^{r_1 t_1}$,

in other words, the same result.

These pundits get paid to go on television and say these things and to write them in Op-Eds. And influential people take them seriously. The innumeracy is staggering.

Having some fun with Tesla data

Downloaded some historical data from Yahoo Finance (yes, I have other better sources, but this one is public and can be shared) and played around with smoothing. Here's a nice view of the TSLA closing price for the last year using the same triangular smoothing I did for my bodyweight (in other words, a second-order moving average of (5,5)):

Throughout the first half of 2019 Tesla boosters on Twitter were fully convinced that this would be the year that heralded the end of the internal combustion engine car. In reality, this seems to be the year in which Tesla's financial shenanigans are likely to bring its valuation to a more appropriate level.

CYA statement: I have no personal position on Tesla and will not initiate one in the next 72 hours. This is not intended as financial advice and represents my personal views (of making fun of Tesla boosters) not those of my employer or our clients.

Also:

(Yes, it's sarcastic.Very, very sarcastic.)

Yet another infrastructure photo

Fun with numbers for Sep 1, 2019

Apple declines to burn 17 billion, Tesla boosters disappointed.

Ross Gerber, a Tesla booster who provides endless entertainment on TSLA twitter, had an interesting idea (the same idea he's had for the last 4-5 years), that Apple should burn 17 billion dollars instead of giving them to people who hold AAPL stock.

The first Tesla car that was targeted at the general public (as opposed to tech billionaires and centi-millionnaires who wanted to be thought as forward-thinking) was the Model S, introduced in 2012. So we'll use 2012 as the beginning of Tesla as a real car company.

As I write this it's Sunday, September 1st, 2019 and the last TSLA close was on Friday, August 31st at 225.61. We'll compare this number with the stock closing price for the closest date for all years 2012-2018 and compute the annualized growth. Then we use that growth to forecast the evolution of an hypothetical Apple stake of 17 billion.

Using a 5-year growth rate for that tweet was basically the textbook default, but looking at that table, the last two columns really tell an interesting story… didn't Ross suggest Apple buy Tesla in late 2017 and in late 2018? Because the numbers in that table say something about financial acumen.

Calories, calories, calories... What a bunch of nonsense!

This marvel of mechanical engineering is the Siemens STG5-9000HL gas turbine. Running in single-cycle mode at nominal power it takes in almost 59,000 kg of LNG per hour or around 900 MW (Calories in: 755 million kCal/hour) and delivers around 400 MW of spinning power to a generator, for about 360 MWe (Calories out: 310 million kCal/hour) of electrical power.

Wait, what? Isn’t it Calories-In-Calories-Out? Is the turbine getting fatter or something?

No. Running in single-cycle mode the system loses around 60% of its power to unrecovered heat.

This is the real problem with CICO and 'Just Get a Caloric Deficit' recommendations: because no one measures the energy lost in radiated, conducted, and convected heat, or the energy content of urine, feces, and ‘outgassing,’ and because those show large variation across people (and across different situations for the same person, including changes in diet), the whole thing is nothing more than pretend science: like astrology made with computers, adding the trappings of science to a flawed foundation yields nothing valuable.

Well, of no real value, but monetizable; and there’s also the moral posturing afforded by telling others that being fat is proof of their lack of willpower or moral failings. (And, of course, since it doesn’t work, leads to continuing supply of clients.)

Thermodynamics is not a magical incantation. But some people use as if it were.

(I'd be willing to bet that 90% of the people who invoke 'thermodynamics' as a magical incantation to ward off the evil spirits of low-carb diets couldn't have made the junior-high Physics computations in that first paragraph.)

More infrastructure

"How can you take photos of those ugly things when there are all these flowers and rocks?"

Me: because when you understand what these things are, you marvel at the detail, at the functionality, and at the fact that they work to begin with. Also, I photograph nature too.

Fun with numbers - August 28, 2019

Air conditioning FTW!

Part of what I do is training people, under the fancy name of "executive education," and to match the fancy name we tend to get nice AV equipment, color handouts, markers that work, fancy chairs, and climate-controlled rooms.

The least remarkable of those is also probably the most important. The air conditioning, not the markers.

For a large-ish event with around 60 people in a comfortably large room, with about 50% of heat losses to the exterior, the temperature would increase by almost 15 °C in a 90-minute session:

Okay, there's a lot of approximations in that calculation, but even 10 °C increase means that either it was too cold at the beginning or too warm at the end. So hurrah for air conditioning.

Seriously, how can people fall for this?

Today The Tesla Promotion Network Electrek, posted a news item about batteries. Apparently "[a] startup that spun out of Cambridge University claims a battery breakthrough that can charge an electric car in just six minutes."

That phrasing is unclear: how much charge? Even the slowest chargers in my neighborhood (6 kW) will give any electric car battery some charge in 6 minutes (0.6 kWh). Most people will read that to mean they could give a full charge to, say, a Tesla Model 3 in six minutes.

Say the Tesla has a 80 kWh battery; charging it in 6 minutes requires an average power of 800 kW; even using 480 V, that would mean the current would be almost 1700 A. That would be a really interesting current to see in a lithium-ion battery, or as the fire department calls it, "the initiating event of the fire."

For comparison, a typical gas pump can pump 3 l/s of gasoline, which at 34 MJ/l means that the gas pump can transfer energy at a rate of over 100 MW or around 125 times the rate of that fictional battery.



Not math: some infrastructure