The Overtrading Tax: Why Doing Less Often Earns More

Ask a new trader how to make more money and the instinct is almost always the same: trade more. Watch the market more closely, react to more signals, take more positions. Activity feels like effort, and effort feels like it should be rewarded.

The data says the opposite. Across decades of academic research and the live records of professional managers, one of the most reliable findings in all of trading is that higher turnover correlates with lower net returns. The traders who win are not the busiest ones. They are usually the most patient ones.

There is a tax on overtrading. It is paid in spread, slippage, commissions, and taxes — and then paid again in the form of more decisions, more chances to be wrong, and more noise mistaken for signal. This post breaks down both halves of that tax, the math of how it drags on compounding, and how Lukra is deliberately built to trade less.

The Explicit Costs: What Every Trade Actually Pays

Every time you enter or exit a position, you pay a toll. Most of these tolls are small enough to ignore on a single trade and large enough to matter enormously across hundreds of them.

• The spread. You buy at the ask and sell at the bid. The gap between them is a cost you pay the instant you transact, before the price has moved at all. On liquid large-caps it is tiny; on thin names it is not.
• Slippage. The price you actually fill at is rarely the price you saw when you decided. The larger your order relative to available liquidity, and the faster you demand execution, the worse this gets.
• Commissions and fees. Even in a zero-commission retail world, there are exchange fees, regulatory fees, and the cost of payment-for-order-flow embedded in your fill quality. Nothing is truly free.
• Short-term tax drag. This is the one most retail traders underweight. In many tax regimes, positions held under a year are taxed as ordinary income rather than at the lower long-term rate. A high-turnover strategy can convert what would have been efficient long-term gains into a stream of fully taxed short-term gains, year after year.

Individually these are rounding errors. Stacked together and multiplied by turnover, they become a structural headwind. A strategy that turns its entire book over twenty times a year is paying that full stack of costs twenty times a year, regardless of whether the underlying view was correct.

The Implicit Cost: More Decisions, More Ways to Be Wrong

The explicit costs are the visible part of the tax. The larger and more insidious part is behavioral and statistical.

Every trade is a decision, and every decision has an error rate. If your edge gives you, say, a 55% chance of being right on any given high-conviction trade, that edge only compounds in your favor when you are selective about when to use it. Dilute your trading by acting on dozens of marginal, low-conviction signals and you are mixing 55% decisions with a flood of coin-flip decisions. The average quality of your decisions falls toward random.

This is the core problem with high turnover: most short-term price movement is noise, not signal. The more often you trade, the more you are reacting to noise — random fluctuations that look meaningful in the moment and mean nothing in aggregate. A strategy that waits for a small number of statistically strong setups is sampling from the signal. A strategy that trades constantly is mostly sampling from the noise.

There is also a documented behavioral cost. Frequent traders tend to sell winners too early and hold losers too long, chase recent performance, and overreact to short-term volatility. Each additional trade is another opportunity for those biases to express themselves. Doing less is, in part, a way of getting out of your own way. We explored a related version of this idea in Why Consistency Beats Prediction in Trading.

The Math of Frictional Drag on Compounding

Frictional cost is corrosive precisely because it compounds against you in the same way returns compound for you.

Suppose two strategies both generate the same 12% gross annualized return before costs. Strategy A turns over its portfolio twice a year. Strategy B turns it over twenty times. Assume a conservative all-in round-trip cost — spread plus slippage plus fees — of 0.15% per round trip.

• Strategy A: roughly 2 round trips × 0.15% ≈ 0.3% annual cost. Net return ≈ 11.7%.
• Strategy B: roughly 20 round trips × 0.15% ≈ 3.0% annual cost. Net return ≈ 9.0%.

That is a 2.7-percentage-point gap, created entirely by friction, with identical underlying skill. Now compound it. Over twenty years, $100,000 growing at 11.7% becomes about $920,000. The same $100,000 growing at 9.0% becomes about $560,000. The high-turnover version surrenders nearly 40% of the final wealth — not to bad decisions, but to the cost of making more of them.

And this ignores taxes entirely. Layer in short-term tax drag on the high-turnover strategy and the gap widens further. The lesson is not that costs are large per trade. It is that turnover multiplies a small per-trade cost into a large drag on the only number that matters: terminal compounded wealth.

Why Low-Turnover, Rules-Based Systems Compound Better

Put the explicit and implicit costs together and a clear principle emerges: a trading system should only act when the expected edge exceeds the full cost of acting. Most of the time, for most signals, it does not.

Low-turnover, rules-based systems have three structural advantages here:

• They raise the bar for action. A discretionary trader can always find a reason to trade. A rules-based system only trades when its defined conditions are met, which naturally filters out marginal setups.
• They pay the tax fewer times. Fewer trades means fewer spreads crossed, less slippage, fewer taxable events, and more gains that qualify for long-term treatment.
• They are more honest about their edge. A system that needs constant activity to justify itself is often disguising the absence of a durable edge with the appearance of work. This is one of the recurring reasons most trading bots fail — they confuse activity with alpha.

Discipline, in this framing, is not a personality trait. It is a cost-management strategy. The discipline to do nothing when the conditions do not warrant action is one of the highest-return behaviors available to any trader, because it directly removes friction from the compounding equation.

How Lukra Is Built to Trade Less

Lukra is designed around the conviction that fewer, higher-quality trades beat constant churn. This is not a marketing posture; it is encoded in how the models actually behave.

Conviction-weighted entries. Lukra's models do not act on every signal that crosses zero. Position size scales with statistical conviction, and leverage only expands into the 2x–3x range when the edge is strong and the volatility regime is favorable. Weak signals produce small positions or none at all, which keeps the model out of low-edge, high-cost trades.

Regime overlays that suppress activity. Our models use regime filters — including 50/200-day SMA guards and VIX-aware sizing — that pull the system toward caution or cash when conditions are hostile. The practical effect is fewer trades during exactly the periods when noise is highest and the cost of being wrong is greatest.

Trailing stops instead of constant repositioning. Rather than churning in and out trying to time every wiggle, Lukra's models tend to let winners run under trailing stops. This captures trend while minimizing the number of round trips — and the friction that comes with each one.

Scheduled rebalancing, not reactive trading. Where rebalancing is appropriate, Lukra favors deliberate, periodic rebalances over reactive, intraday churn. Rebalancing on a defined cadence keeps the portfolio aligned with the model's intent without manufacturing a new taxable, cost-bearing event every time the market twitches.

Transparent reporting that counts the cost. Because we report performance on a net basis and publish risk-adjusted figures like Calmar, Sharpe, and Sortino alongside raw returns, the cost of turnover has nowhere to hide. A strategy that traded too much would show it in the net numbers, and we would rather see that and fix it than paper over it.

The result is a system that, by design, would look boring to a day trader. It often does very little. That is the point. The cheapest, most reliable improvement available to almost any strategy is to trade less — and to make each remaining trade count.

---

For a closer look at how Lukra chooses which conditions justify acting at all, see Mean Reversion vs. Momentum Strategies.

You can review Lukra's live, net-of-cost performance across all active strategies. View strategy performance →

Past performance is not indicative of future results. Algorithmic trading involves risk of loss. Cost and tax figures used in this article are illustrative and depend on your broker, jurisdiction, and individual circumstances.