From the March 01, 2011 issue of Futures Magazine • Subscribe!

To scalp or swing trade? That is the question

Expected value

An analysis of the two strategies begins with a model that quantifies the alternatives. To keep it as simple as possible, we begin with a single-position trade; no scale-in/out techniques are used. The goal is to make a relative comparison of the expected value of basic scalp and swing trades.

imageTo do this, a directed trade graph is introduced (see "Trade directed," right). In the graph, x is the trader’s winning percentage; 1-x the losing percentage; PT is the trader’s profit-target in ticks; SL is the trader’s stop-loss in ticks. The graph’s equations and outcomes easily can be coded in a spreadsheet to evaluate trade strategy expected value.

A scalp trade with a 70% winning percentage on one contract with a one-point (four ticks) ES profit-target and stop-loss is used because we expect the scalper to have a high winning percentage with fast profit-taking and tight stop-loss. This gives a per-trade expected value of 1.60 ticks. We quickly notice that, under these assumptions, a high winning percentage is required (subtract approximately 0.25 ticks to cover commissions). Traders can increase the expected value by increasing the profit-target and stop-loss values equally, or by multiplying the value by the total number of contracts traded.

imageThe baseline scalp is compared against a variety of swing trades where the swing trade winning percentage is fixed to match the profitability of the scalp (see "Technique comparison," right). The scalper matches the profitability of the swing trader at various swing trade winning percentages and profit-target and stop-loss levels. For example, the baseline scalp is as profitable as a swing trade that garners 10 points of profit (40 ticks), using a two-point initial stop-loss (eight ticks) and a 20% winning percentage.

As with any model, the "garbage-in/garbage-out" rule applies. The goal is a simple model that is useful. We find that with just a modest improvement in trade-winning percentage, the five- and 10-point swing trades become significantly more profitable than the scalp. This is highlighted in the bottom two rows of "Technique comparison."

Regarding profit-target and stop-loss values, the scalping baseline (first row) uses tight and balanced profit-target and stop-loss values, as we might expect when scalping. The swing trade cases use a modestly larger stop-loss based on the goal of staying in a trade longer. Winning percentages were chosen across the swing trade examples to create a normalized comparison with the scalping baseline. A winning percentage roughly will follow the ratio of stop-loss to profit-target; a small stop-loss to profit-target ratio will give low winning percentages. Large stop-loss to profit-target ratios can give high winning percentages. When working with the expected value models, reasonable stop-loss and profit-target values are chosen and then winning percentages are derived.

The first conclusion drawn is that scalping is only preferred when swing trade winning percentages are low. If the winning percentages can be increased to a moderate level, then swing trading becomes more profitable. We might expect that successful swing traders claim their percentages are at, or above, the 33% and 45% levels, thus making their trading more profitable than scalping.

comments powered by Disqus