Conviction Makes a Terrible Position Size

Conviction is the worst number to put into a position size. It isn’t a number at all. It is a feeling about an idea, and the feeling is usually strongest right after you have spent an evening reading about the stock, which is exactly when your judgment is least independent of the story you just absorbed. Size up because you feel sure, and the mood is deciding how much of the account is on the line, with nothing to check it against once the trade goes wrong and the mood changes.

Better inputs sit on the screen already: the account size and the stop distance.

Size from the stop

Take a hypothetical $25,000 account that risks 1% on any single trade. The setup: entry at $62, stop at $58. The stop is where the idea would be proven wrong.

Notice what the position value is doing. At $3,844 it’s about 15% of the account. That figure falls out of the sum. You never chose it. You chose the loss you can accept and the price where you’re wrong, and the share count followed from those, which is how a position size should arrive: as the output of a sum you could show to anyone. The position size calculator runs this for any entry and stop.

The high-conviction version

Now the same trade, with the trader feeling unusually sure, at 2%.

Double the risk, double the shares, double the position. Nothing about the trade changed. The stop is where it was. The chart is the same chart.

What changed is the size of the hole if the stop is hit: $500 instead of $250, on a trade whose odds are whatever they were a minute ago, before the extra confidence arrived. Picking 1% in the first place is a separate decision, and the lesson on risk per trade in dollars works through it. Whatever figure you settle on should hold from one trade to the next.

What a streak does to each

One loss proves nothing. Losses come in runs, though. A run is where the sizing choice shows up. Take three losses in a row, each sized on the remaining balance.

At 1% the account has lost about as much as one bad afternoon. At 2% it has given up nearly 6%, and if those were the trades you felt surest about, you have also learned that the feeling was wrong three times running while it cost twice as much. A 5.9% drawdown needs a gain of about 6.2% just to get back to $25,000. It’s recoverable. It is also how deeper holes start, and the course on risk management is built around the habit of never letting a string of ordinary losses turn into one that takes months of good trading to climb out of.

What conviction cannot see

Conviction attaches to the idea. The product is good, the chart is clean, the quarter will beat. It knows nothing about what decides how much a trade can hurt.

It can’t see how far away the stop is. A wide stop on a trade you love still means fewer shares. Otherwise the budget breaks.

It can’t see a gap. A stock that opens below your stop fills below it. The bigger the position, the bigger the surprise.

It can’t see the calendar. An earnings date inside the holding period, a central bank decision, a lockup expiry: none of them care how you feel about the company, and each can move the price past your stop overnight while the order sits there doing nothing.

A good reward-to-risk ratio is a better reason to take a trade than a strong feeling, and even that belongs in the decision to take the trade, with the share count left to the stop.

The objection from the trader with a journal

The strongest case against this call comes from traders who keep careful records. Their journal shows one type of setup working better than the rest over many trades, and they ask, reasonably, why they shouldn’t put more behind the thing their own records say works, since ignoring evidence you collected yourself looks like its own kind of mistake.

That can be defensible, under conditions. The rule has to be written before the trade. Nobody decides well while looking at a chart they like. The rule also has to rest on your own recorded results for that exact setup. And it needs a hard cap, for instance a rule that no trade ever risks more than 1.5% however good the setup looks, so that a journal-based rule can’t slowly become a feeling in disguise.

What fails is the in-the-moment version. Deciding at the order ticket that this one deserves double is conviction sizing. It just comes with a better story.

Take the trade on conviction, size it on the stop

So the call is narrow. Conviction can decide whether you take a trade. It should never decide how big the trade is. The account size and the stop decide that. Only a written rule, set in advance and capped, may override them.

Where does the call stop holding? A long-term investor with no stop sizes another way. For anyone trading from a stop, the arithmetic above applies every time. The feeling will be there on the next trade too. Let it pick the stock.