Moving from Static Sensitivity to Stochastic Modeling for SMB Diligence

I’m heading to two conferences this spring to dig into the same problem: how we can move from "gut-feel" metrics to actual system rigor in the SMB space. Most deal models I see are dangerously static. We hard-code a 10% churn rate or a 50 bps interest rate shift, run a sensitivity table, and call it a "Downside Case." But real systems aren't static; they are probabilistic. If churn, growth, and interest rates all move simultaneously, what is the actual probability of ruin? I’m currently moving my own deal analysis into Python to stop looking at single-point IRR estimates and start looking at p-values. I’ll be at these two events this spring to dig deeper into this: - MIT SSAC (Boston, March 6-7): I’m returning for the 20th anniversary to see if the influx of private capital has finally turned sports into a consistently profitable asset class through better technology and yield management. - SMBash (Dallas, April 22-24): This is my first time attending. I’m looking to connect with operators who are moving beyond the "viral thread" version of ETA and actually dealing with the math of Main Street. I am particularly interested in how these models perform in the tax, accounting, and B2B services sectors - industries with high seasonal variance where a single-year EBITDA multiple tells almost none of the story. If you are an operator or a searcher in Boston or Dallas and want to trade notes on building more rigorous diligence models or data partnerships, let’s find a time to connect.