How long does it take to learn to trade?

Nobody can tell you honestly, and the one large study that looked found no evidence of learning by day trading at all. What can be said is that unstructured repetition is not practice, and that a record of about a hundred trades is the first point at which your own numbers mean anything.

What should a trading journal track?

The setup each trade was scored against, whether the rules were kept, the planned stop and the actual one, the result in R, and the mistake tags. Dollars last, if at all.

What is a trading playbook?

A short list of setups, each written as conditions you can answer yes or no to from a chart. Written that way, 'was that a good setup' becomes a count instead of a feeling.

Which metrics matter most?

Rules kept as a percentage, the distribution of results in R, expectancy in R once you have around a hundred trades, trades per day, and your worst day. Win rate on its own tells you almost nothing.

Why do so few people learn from experience in markets?

Because markets are a wicked learning environment: feedback is noisy, delayed and sometimes rewards the wrong action. Kahneman and Klein argue intuition can be trusted only where the environment is regular and the feedback is good, and the journal is how you make trading feedback good.

How to learn to trade properly: journal, review, playbook

The short answer

Learning to trade properly means turning a wicked feedback loop into a kind one. Markets reward wrong decisions often enough that experience alone teaches very little, and the largest study of new day traders found no evidence of learning by doing it. What works is the method used in every other difficult skill: write down what you intend to do as setups with countable conditions, record every attempt against that intention, review the record on a fixed schedule for the gap between plan and action, practise the one thing the review found, and measure progress by rules kept and results in R rather than by money. It takes about four minutes a day. Nobody can tell you how long it takes to work, and anyone who does is selling something.

Why doesn't experience teach traders?

Because the feedback lies. Kahneman and Klein, coming from opposite camps on the value of intuition, agreed on the conditions under which it can be trusted: an environment regular enough to provide valid cues, and enough opportunity to learn those cues through feedback. Their paper adds the line every trader should have on the wall: subjective experience is not a reliable indicator of judgment accuracy. Hogarth, Lejarraga and Soyer give the two kinds of environment names. Kind ones match what you learn from to what you will be judged on. Wicked ones do not, and inference in them goes wrong in ways the learner cannot see.

A market is wicked by construction. A trade taken with no setup and a dragged stop can win. A perfect trade can lose. If the only feedback you keep is the P&L, you are learning from noise, and the study of Brazilian day traders found exactly that: following everyone who began between 2013 and 2015, the authors found no evidence of learning by day trading. Barber, Lee, Liu and Odean found the same shape in Taiwan: less than 1% reliably earned positive returns net of fees, but that 1% did it year after year. Skill exists. It is just not transmitted by screen time.

The journal is the device that makes the environment kind. It records the decision separately from the outcome, so the thing you learn from is the thing you controlled.

What is deliberate practice, for a trader?

Ericsson's 1993 paper set out the framework: expert performance is the end result of prolonged, effortful attempts to improve, aimed at specific weaknesses, with feedback, rather than of hours spent performing. The violinists in his study were not better because they had played more concerts. They were better because they had spent more hours on the parts they could not yet do.

Translated: trading is the concert. Practice is the review afterwards and the one adjustment it produces. A trader who takes forty trades a day and reviews none has performed a lot and practised nothing. A trader who takes three, writes each one up, and spends Friday afternoon on the one recurring mistake, is practising.

How do I write a playbook?

A playbook is a short list of setups, and a setup is a name plus a list of conditions you can answer yes or no to by looking at a chart. Not "strong trend"; rather "price above the opening range high, the second pullback, volume on the breakout bar above the ten-bar average." Written that way, each trade can be scored: four of six required conditions held, and here are the two that did not. "Was that a good setup" stops being a feeling and becomes a count that can be argued with.

Steps:

Pick the one setup you actually trade. One written properly beats five written vaguely.
List the conditions. Each one must be checkable from the chart before entry, not after.
Write where the stop goes and why, as part of the setup, not per trade.
Write what disqualifies it: the conditions under which you do not take it even if the rest are present.
Give it a name and use only that name in the journal.

The template you are handed by anyone, including a trading app, is not an edge. It is an example of the shape a written setup takes, which is the part most people get wrong. Edit it into your own.

A setup named ORB pullback in the app's playbook, marked Developing, with its conditions in three groups, each under the app's one-line meaning. Must be true: price above the opening range high; the second pullback; volume on the breakout bar above the ten-bar average. Makes it better: higher timeframe agrees with the direction. Stand down: within 5 minutes of a scheduled release. — A setup in the app's playbook, in its three groups. The must-haves are this guide's example; the other two lines are from the app's starter list. It is not a setup the app hands you.

What goes in the journal, and when?

Every trade, on the day, including the ones you would rather forget. A journal you have curated is a journal about somebody else. For each trade:

The setup it was scored against, by name, and which conditions held.
The planned stop and size, and the actual ones. The difference is the most important field on the form.
The result in R, and only then in dollars.
Whether each of your standing rules was kept: stop not moved, size not increased after a loss, daily limit respected.
Tags for the mistakes, from a fixed short list, and one or two for what you did right.

"Stop not moved" can be checked rather than remembered. If you trade futures at Tradovate, a free page here reads its notifications log and counts every move that gave a trade more room.

Then a day card: the goal written before the open, the review written after the close. About four minutes, most days. The morning line writes tomorrow's first one.

Fills should come in on their own where the broker allows it, and by export where it does not. A journal that is a typing exercise gets abandoned in week three. The fields as a table you can copy are in what to write in a trading journal.

One trade's card in the app: SPY 668 C · long, ORB pullback, in 09:52:14, with a green ring and "kept the rules, lost". Risk $250.00, the stop that was live; Result −1.00R, −$250 realized. Execution and Risk are graded A on an S-to-F switch and Rules is set to Kept every rule. By the book, Setup grade B: three must-haves held, shown as solid dots; the extra and the stand-down did not, shown as rings. — A trade in the app's journal: which conditions held, a grade for setup, execution and risk, and a mark for whether the rules were kept and whether it paid.

How should I review?

Weekly, at a fixed time, with the platform closed. The first question is the only one that matters: did I follow the plan? If yes, the outcome is information about the setup. If no, the outcome is noise, and the review is about why the plan lost the argument.

Count rules kept as a percentage of trades. This is the number to move first.
Look at the results in R by setup. A setup with a negative average over fifty trades is being told something.
Look at the mistake tags. The one that appears most is the practice assignment for next week.
Look at the worst day. What was the third trade that day for?
Write down one change. Not five.

After roughly a hundred trades, expectancy in R per setup starts to mean something. Twenty settles nothing. Anything an app or a person tells you about your edge before that is a guess with a chart on it.

Which metrics matter?

In order:

Rules kept. Percentage of trades on which every standing rule held. Under your control on every trade.
R distribution. How many trades at -1R, how many worse than -1R (those are moved stops), how many at +1R, +2R and beyond.
Expectancy in R, per setup, once the sample is a hundred or so.
Trades per day, and how the results change as the count rises.
Worst day in R. The number that decides whether you are still here next year.
Win rate, last, because on its own it says nothing. A 40% win rate with winners at 2R is a good record; a 70% win rate with losers at 3R is not.

Dollars are never hidden from you. They are just not what the letter grade is about, because a grade you can raise with one lucky trade stops measuring anything you control, and wanting the result at 3 p.m. on a red day is the whole problem. The experiment behind that, and the six lines a weekly card carries instead of a dollar figure, are in process over profit.

How long does it take?

Nobody knows, and the honest studies are discouraging: 97% of the Brazilians who persisted past 300 days lost money, and the most active fifth of 66,465 US households trailed the market by 6.5 points a year. What those studies do not show is anyone who kept the kind of record described here, so they cannot tell you the method fails either. What can be said is that the method has a cost of four minutes a day, and that in return you get the first accurate description of your own trading you have ever had. Whether that is worth it is your call. Anyone promising more is selling something else. The rest of those numbers, and a checklist that scores whether what you are doing is a process at all, are in is your day trading gambling.

Where Degen Pro fits

Degen Pro is a Mac app for Interactive Brokers built around this method. Options fills arrive from Interactive Brokers as they happen and futures come in from a Tradovate or TradingView export, and each trade is graded on setup, execution and risk, marked for rules kept and whether it was planned, and tagged for what went wrong and what you did right. The playbook holds setups as conditions, must be true, makes it better, stand down, and over time the app says which conditions earn their place and which cost you trades for nothing. The day gets a report card graded on process, not money, and a review the app writes from the record; the week gets a card with rules kept and a mark for every trade, and no dollar figure and no R on it. Position sizing works from the stop and the budget and says when the record supports sizing up and when it does not; a coach reads the same record and answers from it, with your own API key kept in your Keychain. It gives no signals — it can place an entry you choose, calls and puts on SPX and SPY, but it never tells you what to take; without Interactive Brokers the journal, the limits and the grades still work from an export, on a Mac.

Start tomorrow morning

Write one setup as conditions.
Decide R, the daily limit and the maximum trades, and write them down before the open.
Log every trade on the day.
Review on Friday, change one thing.
Repeat until a hundred trades are in the record, then look at expectancy for the first time.

None of this is investment advice. Trading involves risk of loss, and most people who try it lose money. The method above does not change that base rate. It changes whether you will be able to tell which side of it you are on.