Insights
Methodology deep-dives, market commentary, and founder notes. Educational content — not financial advice.
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 — including the uncomfortable fact that, on our live verified record, our home market has so far run below a coin flip.
Australia stock prediction (ASX): the honest record
Our ASX model shows a modest edge over a coin flip on the live, verified record — not a money printer. Here's the resource-and-financials reason it's only modest, and why the always-current public audit beats any number frozen in a blog post.
India stock prediction (NSE): a real but modest edge, honestly framed
India (NSE) is one of our better markets — a real but modest edge over a coin flip, not a money printer. Here's the structural reason a momentum model reads Indian large caps relatively well, and where the live, always-current record behind that claim is published.
Vietnam stock prediction (HOSE): a frontier market and an honest, modest edge
My model shows a modest, honestly-stated edge on Vietnam (HOSE) — not a money printer — with real frontier-market and data-coverage caveats I own. The live, scored record is public, call by call.
Indonesia stock prediction (IDX): the honest record and why my model lands below even
My model's live directional record on Indonesia (IDX / .JK) sits below a coin flip — one of my weaker markets, and every resolved call is public. Here's why thin liquidity, the rupiah, and foreign flows feed a momentum model noise it misreads.
Thailand stock prediction (SET): my single worst market, and why I'm telling you
Thailand (SET) is my single worst market: a large verified sample with a live record clearly below a coin flip — and I'm telling you anyway. Here's the market-structure reason why, and why a record this bad doesn't mean you can just bet the other way.
Malaysia stock prediction (Bursa): the honest record and why it's hard
Malaysia (Bursa) started as my worst, thinnest-data market — then the verified sample grew and it converged to roughly break-even, mid-pack territory. Still no edge, and here's the market-structure reason why.
Singapore stock prediction (SGX): the honest record and why it lands near a coin flip
My model's live, verified record on Singapore (SGX) sits about as close to a coin flip as it gets — and I publish that instead of hiding it. Here's why REITs, defensive banks, and low momentum turnover keep it near even.
Hong Kong stock prediction (HKEX): the honest record and why it lands on a coin flip
My model's live, verified record on Hong Kong (.HK / HKEX) lands almost exactly on a coin flip — and I publish every resolved call to prove it. Here's why China-linked names, policy shocks, and Southbound flows keep it from finding an edge.
China stock prediction: the honest record and why it's a harder market
On our full public log, China (Shanghai) is one of our weaker markets — below break-even so far, and we say so openly. Here's the honest reason: it's a retail-driven, policy-sensitive market that's hard for a momentum model.
TW stock prediction with machine learning: the honest record and why it's hard
Most global AI stock tools ignore Taiwan entirely. We don't — and I'll be honest about the record: our live directional accuracy on TW stocks has run below break-even so far, one of our weaker markets. Here's the verifiable evidence, why Taiwan is genuinely harder to predict than markets like the US and Canada, and what that teaches you about market efficiency.
Korea stock prediction (KOSPI): our honest record on a retail-driven market
On KOSPI, our blended live record sits at roughly a coin flip — honestly disclosed, on a now-large verified sample — with the 3-day horizon the one relative bright spot. Here's why retail 'ant' flow and chaebol concentration wash out the technical signal.
Japan stock prediction (Nikkei / TSE): the honest record and why it's a coin flip
On our full public log of verified predictions, Japan lands almost exactly at a coin flip for our model — essentially no edge, up or down. Here's the honest, structural reason the Nikkei and TSE come out even, and where to check the live record.
UK stock prediction (FTSE): the flashy fluke that collapsed — exactly as I warned
Our UK model once posted an eye-catching directional accuracy on a tiny sample. I told you not to trust it. Now, with a far larger verified record behind it, it has settled where I said it would: a modest edge just above a coin flip, not a money printer. Here's the lesson, playing out in the open.
Canada stock prediction (TSX): our best market, on the live verified record
The TSX is currently our single best market on the live, publicly scored record, strongest at the one-week horizon. A real but modest edge — above-average for us, not a money printer — with the always-current figures on the public self-audit.
US stock prediction with machine learning: where our model actually works
The US is the one market where our model genuinely earns its keep — a real, sample-backed edge, but a modest one, not a money printer. The live figure lives on our public self-audit, not frozen in a blog post.
Walk-forward validation — why most stock-prediction backtests lie
The 'backtested on years of data, huge annual returns' pitch you see in finance Twitter ads is almost always smoke. Here's what's actually happening, why even a rigorous walk-forward backtest still flatters itself, and why we publish our live forward-tested record instead — gap and all.
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 an honest look at our own live record — including the markets where our model is genuinely weak.
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.