Secrets to Short-Term Trading Ch. 4: Trading Frequency and the Cost of Impatience

Investment Background

Larry Williams's decades in short-term markets produced a specific, recurring observation about his own students and readers: given a genuinely profitable trading approach, the most common way people damage its returns is not by breaking its rules on entries or exits — it is by trading it more often than the edge actually supports.

way-of-the-turtle covers position sizing, exits, and diversification across markets in detail. What it does not address — and what belongs specifically to a short-term trading discipline — is the separate question of trade frequency itself: how often to look for a new trade, and the specific psychological pull toward looking more often than the strategy calls for.

The Wall Street Translation

The Arithmetic of Overtrading

Here is the mechanism, worked through with numbers. Suppose a strategy has a genuine positive expectancy of $150 per trade after all costs, generating roughly 100 qualifying signals per year when followed strictly. Followed exactly as designed, that is $15,000 in expected annual profit.

A trader who grows impatient between signals and begins taking marginal setups — trades that fall just outside the strategy's actual criteria — might double their trade count to 200 per year. If those additional 100 trades have a expectancy of even slightly negative $20 per trade (a realistic outcome for lower-quality setups plus added transaction costs), total expected profit falls to $13,000 — fewer dollars from twice the trades, twice the transaction costs, and twice the screen time.

The frequency was never free. Each additional trade carries its own cost (commissions, slippage, bid-ask spread) and its own risk, and past the point where the strategy's genuine signals are exhausted, every additional trade is diluting the average quality of the whole set.

Why the Pull Toward More Trades Is Specifically Strong in Short-Term Trading

This is where short-term trading differs meaningfully from longer-horizon systematic approaches: the feedback loop is fast enough that boredom, restlessness, and the sense of "missing out" between signals accumulate on a timescale of hours or days, not months. A position trader waiting weeks for a signal experiences this pull rarely. A short-term trader watching a screen for hours experiences it constantly, and each idle period is an opportunity to convince oneself that a marginal setup is actually a qualifying one.

The rule that follows is uncomfortable precisely because it fights this pull directly: the correct response to "nothing is happening" is usually to do nothing, and the strategies that survive are disproportionately the ones whose rules make "do nothing" the default state rather than something that has to be actively chosen against restlessness.

Division of Labor With the Rest of the Library

Book	Owns
way-of-the-turtle	Position sizing, exit mechanics, and diversification — how much to risk once a trade is taken
misbehaving ch5	Why knowing a bias is not an edge — the general limits of self-awareness against behavioral error
This book	Trade frequency specifically — the separate question of how often to act, distinct from how much to risk once acting

Executable Trading Rules

1. Set a maximum trade count per period in advance, based on the strategy's historical qualifying-signal rate — not on how the trader feels. If the strategy historically generates roughly two signals per week, a week with five signals should trigger suspicion of the signals, not celebration of the opportunity.

2. Track the expectancy of "marginal" trades separately from core-criteria trades. Tag every trade at entry as either meeting the strategy's full criteria or a near-miss taken anyway. Reviewing the two buckets separately almost always reveals the near-miss bucket is dragging down overall results, well before it feels obvious in the combined number.

3. Treat idle time as the strategy working, not the strategy failing. A strategy's edge exists in specific, identifiable conditions; the absence of those conditions is information, and trading through it to generate activity destroys exactly the selectivity that produced the edge in backtest.

4. Add a cooling-off rule after any impulsive trade. A fixed pause — one full session, non-negotiable — after any trade taken outside the strategy's stated criteria interrupts the pattern before a single lapse becomes a habit of lapses.

Relevance to a Retirement Portfolio

Overtrading in a retirement account rarely looks like short-term speculation — it looks like excessive rebalancing, frequent fund-switching in response to short-term news, or "improving" on a target allocation more often than the plan calls for. The mechanism is identical to this chapter's arithmetic: each additional discretionary change carries transaction costs, tax consequences, and the risk of moving at exactly the wrong moment, and past a very low frequency of intervention, added activity subtracts value rather than adding it.

The transferable rule mirrors Rule 1 directly: set the rebalancing and review schedule in advance — annually, or on a fixed drift-band trigger — and treat any urge to act outside that schedule as the same restlessness this chapter describes in a short-term trader, not as new information the plan failed to anticipate. A retirement portfolio's edge, such as it is, comes substantially from the discipline of not touching it — the same discipline this chapter asks of a short-term trader between signals.

Chapter 5 addresses a related but distinct discipline: holding a position through news-driven noise without either panicking or growing careless.