Leaner, Faster, Humbler: Value Betting Engine Drops a Few More Bytes

Bertrand Russell said, “Mathematics, rightly viewed, possesses not only truth, but supreme beauty — a beauty cold and austere, like that of a sculpture, without appeal to any part of our weaker nature…”

In my pursuit of the perfect Value Betting Engine (VBE), my weaker nature won on numerous occasions.

It won when incorporating a finite impulse response (FIR) filter, which caused the algorithm to perform too many real-time calculations — sometimes retroactively — on the hard right edge, just where one least needs the waters muddied.

It won when attempting to stretch the Principle of Commonality beyond its natural tolerances by converting real variables into derivations.

It won when bullheadedly determining, absolutely, to solve for problems of variance, although it was becoming clear that variance wasn’t a mere nuisance-artifact. What is more, obstinacy in the face of variance opened up a new gate to a hitherto uncharted and bedeviling region of weakness where quandaries go to be solved with standard error.

These manifestations of my weaker nature stemmed from a deeply rooted desire to make the math say what I thought it should (pride) rather than to simply permit the math to tell me what it wanted me to know (humility). I was also suffering from an unacknowledged attachment to FIRs.

To be fair, even while saddled under a hefty load of hubris, poignant and actionable truths about cyclicality were being revealed in the latter iterations of the VBE, which is a testament to its robustness, but I knew a personal character flaw was going unaddressed: a predilection for denial.

I was in denial that the truths that I was discovering throughout the development of the VBE were simpler than I thought they should be. I was in denial that those same truths didn’t need my assistance with sophisticated statistical analysis. And finally, I didn’t want to believe that I was neither telling a new story nor contributing much to an old one.

But as soon as it became clear that I wasn’t breaking new ground in the field of cyclical analysis, I was able to sublimate pride, cut source code down by a third and settle for humbly facilitating a pedestrian equation that tells an old story in a novel way…

…in 243 bytes.

It may not be art, it may not be original, and it may not evince, as Russell would say, ‘a true spirit of delight,’ but it is lean, fast, accurate, and accomplishes automatically everything that J.M. Hurst used to do laboriously with a pencil and graph paper.

VBE Opportunities, 12/22/21


Leave a Reply