Tuesday Toolkit 9/15/2020

Photo by Monstera on Pexels.com
Tuesday’s Tool
Today we’re talking interaction effects. Definition: A situation in which the effect of one causal variable on an outcome depends on the state of a second causal variable (that is, when effects of the two causes are not additive).

In English, that is: When an outcome depends on multiple variables, the way those variables interact with each other may have a large impact on the outcome.

Let’s look at an example using cookies. The two variables that contribute to the number of edible cookies made are time in the oven and temperature of the oven. The output, which you presumably want to maximize, is the number of cookies made that are edible (yield). If we consider two possible states for time and temperature, we end up with four possible scenarios. 

– Low temperature + short time in oven = few edible cookies (raw)
– Low temperature + long time in oven = a lot of edible cookies
– High temperature + short time in oven = a lot of edible cookies
– High temperature + long time in oven = few edible cookies (burnt)

Using programs like MiniTab, R, or MatLab, one can run an Analysis of Variance (ANOVA) or interaction plot to find out if there is a strong interaction effect, a weak one, or none at all. 

Using this data; 
Temperature   Time   Yield
    1                     1     30
    1                     1     35
    1                     2     60
    1                     2     58
    2                     1     60
    2                     1     64
    2                     2     30
    2                     2     35

Produces the below graph ⤵

Image description

The lines are not parallel which indicates an interaction effect between Time and Temperature on the Yield of cookies. If the two variables had no impact on the outcome, the lines would be parallel. Where is this used? Six Sigma projects, Design of Experiments, Quality Control, and any other domain that collects data can use an interaction plot to understand if any independent variables have an interaction effect with response variables. Some examples are; 

– Pharmacology. The interaction of multiple elements to make drugs might create the desired effect, cancel it out, or multiply the effectiveness. 
– Risk for diabetes. The interaction of genetic risk markers and diet might mean a higher chance of diabetes. 
– Food taste. The interaction of condiments and food on overall taste. 
– Manufacturing. The interaction between temperature and pressure on over all strength. 
– Sales. The interaction between using a meeting agenda, the time of day, and a positive or negative response of a buyer. 
– Drug effectiveness. The interaction between medications and a patient’s medical response. 

Statistics is terrible, don’t @ me. It is useful though. Having a framework to measure, analyze and think about ways to improve an outcome is important. Next time you find yourself trying to achieve a better outcome, take a step back and look at what the input variables are. Collect some data, and run an interaction plot in MiniTab (there is a free version). 

So, What does happen when too many people try to decide? 
There is an interaction effect that changes the outcome. When someone is angry, the way other people respond to that anger is an interaction effect. 
When someone’s infectious laughter takes over a meeting, the resulting decisions will likely be different than if there had been no laughter. The state of one person effects the state of others, which in turn effects the outcome. 

PayPal Mafia Valuation
The original founders and builders of PayPal have long been known as the “PayPal Mafia.” Peter Thiel, Elon Musk, Reid Hoffman, Keith Rabois, Max Levchin and about twenty others have moved on from PayPal to create dozens of companies, write books, create Venture Capital firms, and much more. YoutTube, Yelp, Yammer, LinkedIn, Tesla, SpaceX, Founders Fund, Affirm and OpenAI are a few examples of companies the mafia went on to create. 

The best estimate I’ve been able to come up with is that the total valuation of the companies they have touched since leaving PayPal is $682,536,900,000. With limited access to private valuations, ambiguity in company valuation, and not enough time in the world to calculate the valuations of all the companies funded* by the mafia, this is the best I’ve been able to come up with. Source. The real valuation is more than likely higher.
For comparison, here is the GDP of a few countries you have on your “will travel there someday” list. 

– Iceland – $26,000,000,000 
– Norway – $ $432,000,000,000
– Ireland – $382,000,000,000
– Chile – $299,000,000,000
– New Zealand – $ 205,000,000,000 

Twenty five people have contributed to building nearly $700 Billion dollars in companies in the last two decades. It’s hard to fathom. 

Best Tweet of the Week
Sticking with the mafia theme, this felt appropriate. Check out the article here

Image description

Best Looking Museum?
The State Hermitage Museum in St. Petersburg. The colors, gold accents, and detailed ornaments are something spectacular. 

Image description
Photo credit.

Athlete of the Week 
Naomi Osaka. A 22 year old Tennis player from Japan. She is one of the top players in the world, having won three Grand Slams and two Premier Mandatory contests. She has twice defeated legendary player Serena Williams in what Osaka called “bittersweet” victories over her childhood hero. Osaka is one of the highest paid athletes in the world with expected 2020 earnings of $35 Million. 

Image description
Photo credit.
*Notable companies funded by the PayPal Mafia; Facebook, AirBnb, Stripe, Spotify and Flexport. Plus many, many more. 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: