I Broke My Own Rules 19 Times in One Month

Ask a trader how many times they broke their own rules last month. You'll get a shrug, or a range. "A few." "Not that often."

Nobody has the number. I didn't either for my first two years. I had written rules, I believed I was following them, and if you'd pushed me I'd have guessed three or four slips a month.

Then I started logging every deviation — the date, which rule, and what I told myself at the time. Month one came back at 19.

The rules weren't the problem. The gap between what I thought I did and what I actually did was the problem, and it stayed invisible until it had a number attached to it.

Measurement does most of the work

There's real research behind why counting changes anything. Harkin and colleagues published a meta-analysis in Psychological Bulletin in 2016 pulling together 138 studies and just under 20,000 people, all testing one question: does monitoring your progress toward a goal actually help you reach it?

Two details in that paper matter more than the headline number. Monitoring worked better when the information was physically recorded rather than just held in someone's head. And it worked better when the results were reported to another person.

That's the entire design of an accountability system, handed to you by researchers who were mostly studying diets and exercise programs. Write it down. Show it to someone.

Here's why it bites so hard in trading. A log doesn't stop you at the moment of temptation — nothing stops you at the moment of temptation, that's what makes it temptation. What a log does is attach a small cost to the break, arriving later.

You move your stop, the trade recovers, you feel clever, and then you still have to type "moved stop, told myself the level was arbitrary" into a file you're going to read on Sunday. That one sentence is where behavior actually shifts.

Richard Dennis's turtles are the cleanest version of this. Same rules, same capital, same two weeks of training, wildly different outcomes. The ones who failed didn't fail because the system stopped working — they failed because they didn't run it. Dennis put it bluntly: you could publish his rules in the newspaper and nobody would follow them.

Where most people get this wrong

Most traders who try this build a journal instead, and journals drift toward the market. Screenshots, entry logic, what BTC did at 3am, a paragraph about the Fed. Useful sometimes. But a journal mostly describes the market's behavior, and the only behavior you control is yours.

The second mistake is logging only the breaks that cost money. Those feel like lessons, so they get recorded. The break that worked — the one where you sized up on instinct and caught the move — quietly vanishes from the record.

That's backwards. A profitable rule break is the most expensive event in your month, because the market just paid you to do it again.

Third: no witness. Private tracking dies in about eleven days, in my experience. Every version of this that survived longer than a month had one other person receiving the count, even when they never replied.

Try the smallest possible version

For the next two weeks, keep one line per deviation. Date, rule broken, what you told yourself at the time. Don't fix anything yet, don't set a target, don't promise to do better. Just count, and send the weekly total to one person who trades.

Then tag your trades. Every entry gets marked "followed" or "not followed," and after thirty trades you compare the two buckets. I've never once seen that comparison flatter the second bucket. And after you've seen your own version of it, the argument about discipline is finished — it isn't a character question anymore, it's a number with your name on it.