Joseph Noko (pronounced “nookoo”) is an equity research analyst. He publishes independent research for institutional investors worldwide, most often on companies that carry no analyst coverage at all: Philippine gold, Hungarian security printing, Estonian construction, Danish regional banking, Nigerian palm oil.
Each thesis is built from the primary filings. Reported earnings are rebuilt into NOPAT, invested capital and ROIC, line by line. A reverse DCF then solves for the growth and returns the current price already assumes. Institutional clients receive the thesis and the model workbook, long and short. Every closed position is published with its realised return, including the losses.
He is ranked in the top 20 (LTM) globally and top 40 all-time on SumZero, among a professional membership in excess of 16,000. SumZero tracks the live performance of every thesis he publishes there: an average total return of 21.96%, 11.28 points ahead of benchmark, and a median annualised return of 23.78%. His model portfolio has run since 2024. Rankings and returns move with the market. He was a top-10 forecaster in the Good Judgment Project’s COVID-19 tournament, a contest scored on probabilistic accuracy against realised outcomes rather than on argument.
His research appears in his newsletter, The Mirandolan, and on SumZero, where it is read by more than 570 analysts and portfolio managers worldwide. New Constructs commissions Long Idea research from him for their institutional client base. Their forensic accounting data powers the Bloomberg New Constructs Core Earnings Leaders Index and reaches institutional investors through FactSet, S&P Capital IQ and LSEG’s Refinitiv platform.
He finds the losses that reported numbers do not show: in the accounts, in the price, and in the risk models that treat a loss like a gain. The method integrates expectations investing, forensic accounting and financial economics, and it rests on a worked theory of risk set out in his master’s thesis, The Nature of Risk. Drawing on Knight, Keynes, Bernoulli, Shannon and Kelly, the thesis argues that risk is subjective, that not all risks are quantifiable, and that because wealth is multiplicative, loss aversion is rationally obligatory rather than a behavioural bias. Lose 50% and you need 100% to recover. That argument is why he builds cost of equity on downside and expected-shortfall beta rather than variance, and why he values companies by solving for what the price already assumes instead of forecasting a future and defending it.
He also builds the systems that do this work. His research engine automates the pipeline from primary filing to valuation, with deterministic logic held in versioned files rather than prompts, two independent derivations of every figure that must reconcile exactly, and language models confined to governed advisory roles. Running an autonomous agent on production research taught him that the failures that matter are not hallucinations but plausible values in the right shape, internally consistent and externally wrong. He serves as a finance expert and AI trainer at micro1, an AI laboratory.
He holds a Master’s in Law and Finance from the Université d’Angers, France, graduating magna cum laude.
Selected engagements are taken alongside the newsletter: earnings-quality reviews, and advisory on the governance of AI in investment processes. He may be contacted through LinkedIn, SumZero, or by email below.