Building a well-tested, properly diversified systematic portfolio is genuinely hard work. Leaving it alone once it’s running is, for many traders, harder still. This isn’t a comment on any individual trader — it’s a well-recognised pattern across systematic trading in general, worth understanding rather than being surprised by. This article is educational, not a diagnosis of anyone’s behaviour; it’s simply a look at common tendencies worth being aware of.
Stopping after losses
A losing streak is uncomfortable to sit through, even when it falls well within a strategy’s known historical range. A common response is to pause or disable a strategy right after a run of losses — often exactly the point at which a strategy with a genuine, if variable, edge is due to revert toward its historical average. Stopping systematically after every drawdown, rather than according to a predetermined plan, can quietly turn a strategy that was designed to be held through its own variance into one that’s only ever run during its favourable stretches — which changes its real, experienced results in ways the backtest never accounted for.
Increasing risk after wins
The mirror image of the same pattern: a strong winning stretch can create confidence that pulls a trader toward increasing position size beyond what was originally planned. This isn’t inherently irrational — but it does mean risk is being adjusted based on recent results rather than a predetermined plan, which changes the portfolio’s actual risk profile in ways that weren’t tested or accounted for in advance.
Switching systems
When one strategy underperforms while another performs well, there’s a natural pull to abandon the underperformer in favour of whatever is currently working. Followed repeatedly, this pattern can leave a trader effectively chasing whichever strategy had the better recent stretch — which is close to the opposite of the deliberate, evidence-based portfolio construction this whole series has been describing.
Chasing recent performance
Related to switching systems: allocating more attention, capital, or confidence to whichever part of a portfolio has performed best recently, purely because it performed best recently. Recent performance is real evidence, but it’s a small, recency-biased sample — treating it as if it predicts what comes next tends to work against the discipline a systematic portfolio was built to provide in the first place.
Overriding entries and exits
Manually skipping a signal that “feels wrong,” or closing a position early because the open drawdown feels uncomfortable, quietly turns a tested system into an untested, discretionary one — even though it still looks systematic from the outside. Whatever edge the backtest demonstrated was demonstrated for the system as specified, not for the system plus ad hoc human intervention.
Abandoning diversification under pressure
During a difficult stretch, there can be a strong pull toward concentrating on “the one strategy I trust most” rather than staying diversified across the portfolio as constructed — which is precisely the concentration risk described in The Hidden Concentration Risk of Trading Only One EA, arrived at through a change of behaviour rather than a change of plan.
Judging strategies on tiny samples
A strategy that loses its first three live trades can feel “broken,” even when three trades are nowhere near enough evidence to draw that conclusion. This pattern connects directly to How to Tell When a Live EA Is Behaving Differently From Its History — the discipline of waiting for sufficient evidence before judging live behaviour is, in part, a defence against exactly this instinct.
The common thread: a documented process
Every pattern above has a similar remedy in principle, even if it’s difficult in practice: deciding the rules — when to review, what evidence justifies a change, how risk will be adjusted and when — before emotion is involved, and then following that documented process rather than reacting in the moment. A system’s tested edge belongs to the system as specified; the moment a trader starts making case-by-case exceptions, the live result stops being a fair test of what was actually built and tested.
Where this fits in the SIPS workflow
SIPS is analytical and monitoring software, not a substitute for a trader’s own discipline or judgement, and it doesn’t provide personalised investment advice — see What SIPS Is / What SIPS Is Not. What SIPS aims to provide is better evidence to make that discipline easier to maintain: clear historical metrics, honest live-versus-history comparisons, and portfolio-level context, so decisions can be based on evidence rather than on however the last few days happened to feel.
The practical takeaway
None of these tendencies are a character flaw — they’re a predictable, well-recognised part of trading a system you can see and feel the results of in real time. The practical defence is the same one systematic trading was built around in the first place: a documented process, decided in advance, followed consistently, and revisited on a schedule rather than in reaction to whatever just happened.