About simplifying assumptions: Voltaire warned "beware of the great simplifiers". But the guys who construct mathematical models don't read Voltaire and are as unaware of the limits of their sphere of competence as the quants who busted the financial markets with simplified Gaussian value at risk models in 1998 and 2008 (Read Scott Patterson's The Quants).
In Mediocristan "largely hold" and "right ballpark" are usually just good enough. Human geopolitical systems and financial markets live in emotion and randomness driven Extremistan and are repeatedly Fooled by fat tail Randomness:
1. On June 6 the Fuhrer slept until 12:00 and at 07:00 Jodl told Rommel "I cannot wake the Fuhrer."
2. Lincoln's luck at Fort Stephens in July 1864 when Lt Oliver Wendell Holmes shouts at him "get down you damned fool!" after the army surgeon standing next to him is killed by a confederate sniper.
3. Chicago Mayor Anton Cermak and Lillian Cross' handbag stopping Giuseppe Zangara's bullet meant for FDR. The dying Cermak tells FDR, "I'm glad it was me and not you Mr. President." The US constitution has no succession mechanism for the assassination of a president-elect. Zangara killing FDR meant succession crisis, no March 1933 bank holiday, no New Deal, no Lend Lease for the UK and the USSR. One bullet stood between the outcome we got and living in The Man in the High Castle.
4. At Midway the skies cleared just long enough for 12 American pilots to find four Japanese carriers.
5. On December 31, 1776 Washington's defeated post Battle of LI army crossing NY harbor from LI to NJ was invisible to the British navy due to fog. Clear skies and he and his defeated army get wiped out and the American Revolution is snuffed out in its infancy less than half a year after the Declaration of Independence. No Battle of Saratoga ten months later, no Franklin-negotiated French alliance the next year.
6. The 1755 earthquake that randomly struck and levelled Lisbon, then the richest city on earth, was geological, geopolitical and philosophical.
There is no mathematical model or systemic redundancy that can accurately represent or insure against Extremistan's high impact, low probability randomness. Applying mathematical models to geopolitics and human history is like me, a non-skater basketball player, walking onto a hockey rink. I'd just fall down a lot and get really, really bloody. The idea that your quant modeling skill set can be applied on the historian's field without reading actual history or historiography is pure hubris. Spend the next decade getting an interdisciplinary education learning languages, reading books of history, political science, economics, philosophy, anthropology and doing research in archives discovering how highly contingent human history works, and then come back and tell me how your quant skill set can be applied. I can email you pdfs of the books I've read just the last ten years that exposed how ignorant I was until the age of 57.
Your current approach is the short cut of naive youthful hubris, like Icarus flying too close to the sun.
