Everything I touched lost money

I went through a year of my own trades this week. Every buy and every sell, matched against the broker's record.

Twenty-two positions closed. Sixteen of them lost money. After charges I was down 8,458.

Then I looked at what I had not sold. The fund I put money into every month. An index fund I bought once in April and forgot about. Those are the ones that are up.

I was not investing. I was fidgeting.

The money went round more than twice

Across the year I bought about 7.2 lakh of things. What I actually hold is about 3.2 lakh.

The gap is not money I lost. It is the same money going round again. Bought, sold, bought something else with it.

Every lap paid a small toll. Brokerage, tax, stamp duty. On its own each one is nothing, twenty rupees here, forty there. Over a year of laps it came to 2,301.

That is not really what hurt me. What hurt me is that every lap was also a decision, and my decisions were worse than doing nothing.

Every sell had a good reason at the time

This is the part I do not enjoy writing.

I did not churn on purpose. Nobody does. Every single sell felt sensible on the day I made it.

It had run up, so I took the profit. It had fallen, so I cut the loss. Something better came along. The news changed. I needed the cash for a different buy.

I look at the holding times now — eighteen days, nineteen days, twenty-one days, thirty-three days — and I can still remember the thinking behind most of them. The thinking was fine. The pattern was the problem, and you cannot see a pattern one decision at a time.

My own screen was hiding it

Here is what made it worse.

The dashboard I built for myself shows me what I own. Value, profit, return. It said I was up.

But what you own is whatever you did not sell. And what you sell is mostly the stuff that disappointed you. So my screen was showing me the survivors and quietly leaving out every mistake.

A year of losses had walked off the page, one sell at a time. I had to go and add them back by hand before the real number showed up.

I do the same thing when I build

I liked noticing this even less.

There is a product I have rewritten more than once. Not because it was broken. Because I had a better idea about how it should work. New structure, better screens, cleaner code underneath.

Every rewrite felt like progress. Not one of them added a single user.

Meanwhile the parts I shipped and left alone are the parts people actually use. Not because they are well made. Because they have had time to be used.

Sitting still feels like nothing is happening. So you do something. And doing something, in a portfolio or in a product, usually means undoing something you already did.

Doing nothing is the hard version

I always thought discipline meant working harder.

Here it means not touching it. Which is harder, because there is nothing to show at the end. No commit. No trade. No feeling that the day moved.

Activity is the easiest thing in the world to mistake for progress. Both of them make you tired.

What I am actually changing

Three things.

I stopped buying single stocks. My own record is the argument. Sixteen losses out of twenty-two is not bad luck, it is a method that does not work.

The monthly fund keeps going, untouched. It is boring and it is ahead of everything I picked myself.

And I added every closed position to the sheet. Wins and losses. If a scoreboard only counts the survivors it is not a scoreboard, it is a highlight reel.

I would rather look at the real number and feel bad for a day than look at a comfortable one for a year.