This is the equity risk premiumThe extra return investors earn for holding stocks instead of risk-free bonds. Over the long run, equities return ~7-10% annually because investors demand compensation for bearing risk.. Stocks go up because investors demand compensation for risk. ES captures this structural tailwind.
The strategy doesn't predict anything. It stays long when the market is trending up and steps aside when it stops. No forecasting. No indicators for entry. Just alignment with the dominant force in equity markets.
We tested both directions. Shorts work sometimes, but the structural tailwind is long. Over 6 years of data, longs-only dominated every metric: higher profit factor, lower drawdown, better walk-forward consistency. Fighting the equity risk premium is a losing proposition for a trend-following system.
Two or more consecutive days where today's high is above yesterday's high.
That's it. No RSI. No MACD. No indicator confluence. Just price making higher highs.
Simpler entry signals work better. We tested SMASimple Moving Average. The mean closing price over N periods. SMA(20) = average of the last 20 daily closes. crossovers, auction cycles, 4-indicator confluences, and combined signals. Higher highs beat all of them.
Adding filters removes valid trades without improving quality. Every additional condition is a chance to overfit. The entry barely matters. The exit does all the work (that's Module 3).
| Entry Signal | Trades | Profit FactorGross profit divided by gross loss. PF of 2.0 means the strategy made $2 for every $1 it lost. Higher is better. | Total PnL |
|---|---|---|---|
| 2+ Higher Highs + Auction confirm | 85 | 1.99 | $12,172 |
| SMA(20) Crossover (raw) | 94 | 1.67 | $10,049 |
| D1 Auction Cycle (raw) | 113 | 1.35 | $4,896 |
All variants use the same SMA(20)-12 trailing stop exit (longs only). Only the entry signal differs. PnL on 1 MES ($5/pt), 2018-2026. Numbers from the patched simulator (2026-04-08). The "Higher Highs + Auction confirm" row is the canonical H12 best config from the buffer sweep.
The Auction Cycle signal alone produces the most trades (113) but the lowest profit factor (1.35) — many false-positive entries. SMA crossover is cleaner (1.67 PF) but still triggers in ranges. Requiring both to agree cuts trade count to 85 but lifts PF to 1.99 — the same edge, much less noise. The lesson holds post-audit: confirmation filters add value when they're orthogonal, not redundant.
Investors have two choices: hold risk-free assets (Treasury bonds) or hold risky assets (stocks). Stocks can lose 30-50% in a crash. To accept that risk, investors demand higher expected returns.
Over the last century, U.S. equities have returned roughly 7-10% per year above inflation. This isn't a prediction. It's structural. As long as the economy grows and investors are risk-averse, this premium persists.
ES (S&P 500 futures) captures this premium directly. A strategy that stays long during uptrends is harvesting this structural return. It's not alpha. It's beta, captured efficiently.
We tested them extensively. RSI, MACD, CumRSI, Williams %R, Stochastic, ADX. None improved results. Most made them worse.
The reason: indicators are lagging transformations of price. They can only confirm what price has already shown. Higher highs IS the information. Running it through an indicator just adds a delay and a parameter to overfit.
The one place indicators help is the exit: the SMA(20) trailing stop. That's because the exit needs a smooth, slow-reacting reference point. For entry, raw price is better.
Which entry signal had the highest profit factor in our backtest?
C) Correct. Higher highs + auction confirm: PF 1.99, 85 trades, $12,172 total. The auction signal alone overgenerates entries (113 trades, PF 1.35); the SMA crossover alone misses some real trends (94 trades, PF 1.67). Requiring both to agree filters out the noise without losing the edge.
You know the edge. You know the signal. Next: the exact entry and exit rules, with real trade walkthroughs.
Module 3: The Rules →