Insights
Methodology deep-dives, market commentary, and founder notes. Educational content — not financial advice.
Why most AI stock-picking tools are lying — and the one question that exposes them
Almost every 'AI stock predictor' advertises an accuracy number you can't verify. There's a single question that separates the honest tools from the marketing fronts — and most of the market fails it. Here's the question, why it works, and our own real numbers (including the markets where our model is bad).
TW stock prediction with machine learning: the honest numbers and why it's hard
Most global AI stock tools ignore Taiwan entirely. We don't — and I'll be honest about the numbers: our model is currently around 41% directional accuracy on TW stocks, below our US benchmark on an early, still-growing sample the model keeps learning from. Here's the real data, why Taiwan is genuinely harder to predict than the US, and what that teaches you about market efficiency.
NZX AI forecast: can machine learning predict New Zealand stocks?
Every global AI stock tool ignores the NZX — too small, too illiquid, not worth the engineering. As a New Zealand company, we don't. Here's an honest look at whether machine learning can forecast NZX-listed stocks, the unique challenges of a small market, and our early (small-sample) numbers.
US stock prediction with machine learning: where our model actually works
Our model hits ~54% directional accuracy on US large-caps across ~830 verified calls — a real, sample-backed edge, but a modest one, not a money printer.
Japan stock prediction (Nikkei / TSE): the honest numbers and why it's hard
We score ~37% directional accuracy on Japan — below our US benchmark, on an early and still-growing sample. Here's the honest reason the Nikkei and TSE are hard for our AI model.
Korea stock prediction (KOSPI): our honest numbers on a retail-driven market
Our model lands around 46% directional accuracy on KOSPI — close to even, a slight bias below 50% on a moderate sample. Here's why retail 'ant' flow and chaebol concentration wash out the technical signal.
China A-share prediction: the honest numbers and why it's hard
We score ~39% directional accuracy on China A-shares on an early, growing sample — below our US benchmark. Here's the honest reason: retail flow, ±10% price limits, and policy.
Singapore stock prediction (SGX): the honest numbers and why it's hard
My model scores ~42% directional accuracy on Singapore (SGX) — below our US benchmark on an early, growing sample. Here's why REITs, defensive blue-chips, and offshore flow make it hard.
Australia stock prediction (ASX): the honest, early numbers
Our ASX model sits at ~43% directional accuracy on just ~42 verified calls — below our US benchmark, on an early and still-growing sample. Here's the mining/commodity-cycle reason.
Canada stock prediction (TSX): promising, but too small a sample to trust
Our model hits ~56% on the TSX — above a coin flip — but on only ~36 verified calls. That's far too small to trust, and here's why sample size matters.
Why I built Trading Agent — and why one person, not a team
There are dozens of stock-prediction tools out there. Most are either gambler bait or expensive enough that nobody you know can afford them. Trading Agent is my attempt at something in the middle — built by one person, in New Zealand, on a laptop.
UK stock prediction (FTSE): why I don't trust our own 72% number yet
Our UK model shows ~72% directional accuracy — but on only ~36 verified calls, that's almost certainly luck. A lesson in sample size, not skill.
Walk-forward validation — why most stock-prediction backtests lie
The 'tested on 5 years of data, returned 70% a year' pitch you see in finance Twitter ads is almost always smoke. Here's what's actually happening, and how a different way of testing — walk-forward — kills most of those numbers down to honest.
Vietnam stock prediction (HOSE): a frontier market and honest early numbers
My model is ~40% on Vietnam (HOSE) so far — an early number on a tiny ~30-prediction sample, with frontier-market and data caveats I own.
Malaysia stock prediction (Bursa): our honest, early numbers
~32% directional accuracy on Bursa Malaysia, on a small ~34-row sample — our thinnest-data market, and the one with the most room to improve.