Overtrading — taking marginal setups because you are watching rather than because they are good — is covered generally in overtrading. The behaviour is not crypto-specific. The ceiling on it used to be the session.
Why the cost compounds
Each marginal trade pays the full round trip: spread twice, fees twice, impact twice, plus funding if it is carried. Those costs are fixed and certain, while the marginal trade’s edge is small and uncertain.
So a doubling of trade count more than doubles the cost drag, and does it against the weakest half of the opportunity set. It is arithmetic rather than psychology once the trades are taken.
The specific crypto trap
The best conditions and the most available conditions are not the same. Liquidity is deepest during main sessions and thinnest overnight and at weekends — which is exactly when a bored trader is most likely to be looking.
So the marginal trade is not merely marginal on setup quality. It is taken in a worse market, at wider spreads, with less depth to exit into. Both effects push the same way.
What actually helps
A defined session, so “the market is open” stops being a reason. A trade count limit, which is cruder than a quality filter and works because it does not require judgement in the moment. And writing the setup criteria down beforehand, since the failure is nearly always a real criterion being loosened rather than abandoned.
The honest test
Would I take this if I had only three trades left this week? Most marginal trades fail it instantly, and the ones that pass are the ones worth taking. Scarcity is what a session used to impose; without it, it has to be chosen.