The diagram above was produced by my father in an effort to predict lotto numbers. Usually in a given year, I read at least a few blogs on using data science to help improve a person’s odds gambling. Gambling might include casino games, lottery tickets, and perhaps even stocks. On the other hand – and I admit that I might be an oddball in this area – I feel that data science should be able tell a person if a rabbit’s foot “brings luck.” I have a British copper penny from 1945 that I consider lucky although I hardly gamble enough to test its capabilities. I suppose such notions are humorous. However, I distinctly remember a priest once blessing children’s lunch boxes; the blessing is meant to protect the children rather than the food in the lunch boxes. Priests can be asked to bless all sorts of things. One would think that “sacred objects” should by now be tested for their potency. I don’t consider this a test of God but rather a test of priests. If my lucky penny isn’t lucky, I suppose I can throw it into a wishing well.
What is more scientific – dismissing beliefs and practices outright – or testing them? Then, apart from whatever those tests have to say, there is a question of whether the outcomes should be considered relevant; whether or not they serve an important purpose. I don’t know how often people are asked to “shut everything down and reboot the computer.” For me, this technique intended to resolve some sort of computer operating problem almost never works. I do it, however, to appease the masses – to show that I support their ritualistic shamanism. A freshly-minted graduate might believe in the power of the “perfect cheer,” the “mission statement,” the “formula to success.” But a company that has been around for some time has probably tried making adjustments to the outward suit; it quickly becomes apparent that such superficialities don’t necessarily lead to increases in revenues or profitability. Such sweet treacle from the mouths of babes doesn’t always stand the test of time.
I was stopped in my car waiting to turn left at an intersection when a chain reaction occurred behind me. A brand-new BMW had struck another vehicle at a fairly high speed. One car after another released a loud bang or thump; then the reaction reached my car which likewise lurched forward after being bumped. Thankfully, the chain stopped with me. The next day I was able to properly inspect the damage. There was barely any evidence of the collision, really – unlike the awful damage to the sedan that had struck my car. I noticed an old palm sticking out from where my car had been hit. Although I have no idea how it got there, all of my palms are blessed. I thought to myself, wow this is pretty powerful stuff, these blessed things. I then decided to put quite a lot of palms in my car. Moreover, I fastened a blessed band of St. Christopher on the steering wheel. I believe he is the patron saint of travelers. This is the sort of experience that might be dismissed as unscientific while companies busily follow their formula to success.
For me, testing whether or not particular conditions lead up to specific measurable outcomes is a simple matter regardless of the exact context. Fortunately, I am not in enough car accidents to test whether or not the palms are actually useful. Moreover, I never remove the palms. I therefore cannot assert that the palms add to my protection even if I were in many accidents. I do, however, attempt to guess the cards in a shuffled deck; and I try to determine whether for example the foods that I eat can improve my ESP. I don’t really focus on food specifically but anything in the data. With a deck of cards, it is easy to determine success or failure. Much less likely is measuring my ability to release telekinetic energy. For example, do eating hamburgers or avocados help a person shoot out telekinetic force? I would need a device capable to detecting a broad spectrum of ultra-low-level emissions. I realize all of this sounds rather foolish except that I am in a perpetual state of “testing.” What “sounds” foolish might be less so after one bothers to gather data and conduct tests.
It has been claimed that faith healers can heal people with their hands using the power of the Holy Spirit. To me, this is a kind of emission; and it should be measurable using proper equipment. If somebody had told me as a child that I could gain the ability to heal people with my hands if, say, I eat a box of raisins every day for several decades, for sure I would eat the raisins. I recognize that I am practically entering the realm of superstition. A baseball player might wear his undershirt backward in order to win a game; or he might eat apple pie beforehand; or put on a disgusting pair of old socks. My point is that people should not be opposed to testing superstitions and beliefs that are testable. Nor should we dismiss the results if the tests show that the supernatural is able to play a role in our lives. Humans occupy a realm of knowledge inadequacy; and to me it seems perfectly fine to be open to the possibilities even if they seem a bit extraordinary. When surreal things happen in life, it is because we failed to take into account the weird and peculiar in our analysis. When passenger planes crash into buildings, this seems kind of surreal. That Salvador Dali moment confirms that we were completely disconnected from reality.
It is called “data science.” This has caused us to dismiss the weird and magical. When I try to predict from a deck of cards, it is to determine whether, in the face of incomplete data, my instincts should be trusted that day. Now, data science in the traditional sense might interpret the results in a purely statistical manner. I get it. If the circumstances and conditions seem to have no impact on outcomes, it isn’t a problem for me to set aside the card guessing. However, I don’t delegitimize the adventure or journey. I believe in the decades ahead, data scientists will encounter all sorts of beliefs and belief systems for instance in corporate environments. There are often good reasons why beliefs exist – giving rise to certain routines, procedures, processes, and policies. It is important to be open to the idea of putting things to the test and premising one’s assertions on data. As sure as the sun rises, a data scientist will make recommendations that will prove unenlightened; and he or she will be blamed for miserable performance. The data scientist has to work towards change; but it is quite difficult to change people without making some effort to understand them and engaging their perceptions. We should do weird. I certainly do.