Taught to Trade / Engineering notes for traders
Maintained TradingView indicators and templates built by an engineer with a decade-plus in trading and data analysis. Every tool documents where it works and where it breaks. We never send buy/sell signals—you set your own alerts.
01 / The boundary
Practical tooling and process education for people who would rather inspect the method than inherit a conclusion.
Monthly letter
$19/month
One widely used indicator or strategy taken apart every month: what it claims, what it actually computes, and the conditions under which it will mislead you. Sometimes the verdict is that it holds up. Subscribe and every issue so far arrives immediately as designed PDFs: RSI, MACD, Order Blocks, Fibonacci retracements and Supertrend. A new issue follows each month. No win rates, no backtest results, and no claim that any of it makes money. Arithmetic and definitions only.
Subscribe - $19/monthBilled monthly through Gumroad. Cancel anytime.What do I get?RS = (smoothed average gain over N) / (smoothed average loss over N) RSI = 100 - 100 / (1 + RS)Two details matter more than the formula itself.
02 / TradingView scripts
Each tool is designed to make a familiar chart question easier to inspect—not to decide for you.
An ATR trailing stop that draws its stop line, confirms flips only at the bar close, and counts how many flips were undone within a few bars, so the cost of a tight setting is visible on your own chart. It rebuilds the stop rule behind UT Bot Alerts from scratch, with credit to the original authors. Where it fails: Every flip is late by about the stop distance, and it has no neutral state: it is always on one side, trending or not.
Open scriptEnter how many trades you have and the rate that worked, and the panel shows the range the true rate plausibly sits in (a Wilson interval), whether that range can be told apart from 50 percent at all, and roughly how many trades it would take to narrow it, next to how many this chart's history could even give. Where it fails: It assumes every trade is an independent draw from one unchanging rate; real trades cluster and regimes change, so the true uncertainty is wider than the panel shows.
Open scriptDraws support and resistance zones from recent swing pivots and counts how often price actually returns to them, then compares that rate with a control set of plain round-number levels on the same chart, so you can see whether the zones earn their place. Where it fails: An entry is not a reaction: the count says price came back to a zone, not that it turned there, and one chart is one sample.
Open scriptA retracement grid that shows the two things most Fibonacci rules leave out: which swing was measured and which ratio set was used. Includes an arbitrary 0.4 / 0.5 / 0.6 control grid so you can see how little the exact ratios matter. Where it fails: The swing depends entirely on the lookback, and a new pivot redraws every level.
Open script ↗Marks where the previous session actually traded: its value area, point of control, and the current session's opening range. Where it fails: Prior structure is a fact about the past, not a prediction. Intraday charts only.
Open script ↗Shows how much a strategy must earn per trade before fees and slippage are covered. Where it fails: The hurdle is arithmetic, not an opinion about your edge.
Open script ↗Splits history into three volatility regimes so you can see which one your results came from. Where it fails: Terciles are a crude cut. A regime boundary is a choice, not a discovery.
Open script ↗Flags the structural problems that make a backtest read better than it should. Where it fails: It catches structure, not judgement. A clean scan is not a working strategy.
Open script ↗Ranks a watchlist by daily range so you know where the movement is. Where it fails: volatility isn't direction.
Open script ↗A classic stochastic with selectable SMA, EMA, WMA, or HMA smoothing. Where it fails: whipsaws in strong trends.
Open script ↗Free explainers
Free pages that take one indicator apart: the formula, what the settings change, whether it repaints, and where it fails. The Teardown Letter goes further, one indicator per issue.
The one formula that explains why VWAP chases price in the morning and barely moves after lunch, and what that does to "touches".
Read the explainerWhat the ATR length and multiplier actually change, why there is no best setting, and whether Supertrend repaints.
Read the explainerThe exact formula, why your RSI(14) can differ from someone else's, what the yellow RSI-based MA line is, and whether RSI repaints.
Read the explainerWhat Key Value and ATR Period change, why there is no moving-average filter in it, and when its labels can vanish.
Read the explainer03 / Tools
Short, inspectable downloads that turn process into a repeatable workspace.
Workbook
$39
A trade journal and position-sizing workbook for recording decisions, risk, and review.
Buy Edge Ledger - $39Checklist
$19
The 28 ways a backtest lies—and a practical way to catch yours before relying on it.
Buy the checklist - $19Bundle
$49 - both tools, $9 less than buying separately
Both tools together for building a more inspectable trading process from the beginning.
Buy the bundle - $4904 / Who it’s for
For people willing to state a hypothesis, inspect the failure cases, and keep ownership of the decision.
Discretionary, but not yet sure what actually works for them.
Some Python, wants IBKR automation without weeks of plumbing.
Wants to do it properly from day one and needs scaffolding.
Shipped a strategy that died live and now suspects every backtest.
Free field edition
A compact introduction to the questions worth asking before you trust a backtest.