Sizing defaults could do more harm than the rest of the app combined

Every other default is recoverable. A bad sizing default isn't, it's the one that decides how big your mistakes are, and a new user is least equipped to evaluate it exactly when they're setting it. I'm not proposing a specific number, because the right number depends on the person. What I'd suggest is that the setup asks rather than assumes, and states plainly what the choice means in terms of a full stop-out. "This size means a losing trade costs about X" is one sentence and it's the most useful sentence a new user could read.
H
HalfKelly
13 posts · 0 followers
+ Follow

Related reading

6 replies

R
RiskFirstRita· Jul 2026 ago
This plus the daily limit in first-run setup would prevent most of the damage new automated traders do to themselves.
TraderWeTraderWe Team· Jul 2026 ago
The "state what it means in terms of a full stop-out" suggestion is concrete and we can do that. Thank you.
M
MomoQueen· Jul 2026 ago
YES! The dollar number is what makes it real! % of account means nothing to a beginner, seeing "this trade can lose $180" hits different!!
ZenTrader_Ana· Jul 2026 ago
There's something worth noticing in that sentence too — the moment you read the loss in actual money is often the moment you find out whether the size was ever comfortable. Do people here revisit that number after a losing streak, or leave it set and look away?
N
NightOwl_Yuki· Jul 2026 ago
Same, and I'd add it should show up again when someone changes size later, not just at setup. Half the bad sizing I've done was months in at 3am, not on day one.
BacktestBetty· Jul 2026 ago· edited Aug 2026 ago
Agree with the ask-don't-assume framing, with two caveats: 1. State the loss for one full stop-out AND for a run of them. A single stop feels survivable in isolation; four in a row is what actually ends people. My backtests always looked fine until I plotted the losing clusters. 2. Whatever number gets shown, base it on the actual stop distance the user set, not a generic assumption. A tidy example figure that doesn't match reality is worse than no figure. Also worth saying, sizing chosen from a backtest is the thing most likely to be overfit. The equity curve tells you what size would have been optimal for that exact past, which is a different question from what you can hold through.
Sign in to reply →
← All feedback