Probability, markets and honest AI
Posts by Kareem Farid, founder of Kunkafa, on forecasting markets as chances rather than certainties: what the numbers mean, what they never promise, and how to think in scenarios.
Probability, Not Prophecy: what an honest forecast looks like
The tagline as a design rule: the other side always a glance away, a chance for every level, a record beside it, and a 50/50 said out loud.
Ten billion data points, seventy experts: how the forecasts are trained and checked
The scale behind Kunkafa's forecasts, how stated chances are checked against outcomes, why the internals stay private, and where to hold us to account.
From candle patterns to measured chances: how Kunkafa reads a market now
Why the old 'next candle up or down' framing was retired for levels, durations and reach rates measured against what happened.
Can AI estimate market probabilities? Look at the record, not the claims
The skeptic's question answered with the rules of the record and the live results, and why 90%-plus accuracy claims are a warning sign.
What a 70% chance actually means, and what a confidence score doesn't
The chance of reaching a level is a rate from the record, checked against outcomes. Why a score of 'confidence' was dropped for a chance you can verify.
The art of building a position
Why experienced investors scale into a position instead of buying at one price, and a free tool that spreads entries around a target.
Risk management with a forecast in hand: a beginner's guide
Position sizing, stop-losses and the Kelly idea in plain English, with every number an example and every forecast a chance, not an instruction.
Why no one can predict exact stock prices
Exact price prediction is mathematically impossible. What honest forecasting shows instead: levels with a chance each, both directions, over a range of durations.
Anti-gambling design: eight decisions inside Kunkafa Predictions
Eight design decisions that keep a forecast from behaving like a slot machine: both directions, no red and green, no countdowns, nothing metered, a record beside every number.
Why ChatGPT can't predict the stock market
Language models learn words, not price patterns, and sound certain without grounding. What forecasting from numbers, checked against outcomes, does differently.
The problem with AI trading apps, and what Kunkafa does instead
Most AI trading apps borrow casino patterns. What a forecast looks like when it is a chance with a record beside it, and why 'signal' was retired.
Understanding uncertainty: why both outcomes remain possible
How to think in chances: a level with a chance in each direction, calibration in plain words, and why a near-even split is the honest normal answer.
Tabular data: the AI problem nobody talks about
Prediction from structured, noisy numbers is harder than it looks and language models do not solve it. Why markets are the proving ground.