If you have been meaning to get into systematic trading, opening range may be the simplest way. Opening range is simple to quantify and lends itself to automation quite easily. Take the high and low of the first N minutes after the open. Common windows are 5, 15, 30 and 60 minutes. Now you have your ORH and ORL. Volume and information concentrate at the open, so everything that happened overnight gets priced in the first few minutes. And the range hands you defined levels, so there is never any ambiguity about where you are wrong. Four common ways to trade Opening Range: 1. The breakout Price breaks above ORH and keeps going. Enter on the break, stop below ORH or at ORL, exit end of day. Classic. 2. The failed breakout Price breaks above ORH, then falls back inside the range. Short the failure. The longs who chased the break are trapped, and they are your fuel. 3. The retest Price breaks out, pulls back to ORH, and ORH holds. 4. Support and resistance Price never leaves the opening range. Trade the boundaries instead of the break. Higher context can lend insights into when this is the right call. Setups 1 and 2 are exact opposites. Same level, same trigger, but opposite trade. They cannot both be right in the same conditions. This means the level is not the edge (surprise, surprise). The context around it is: volume, volatility, how wide the range is, whether the market gapped, etc. and this is where the fun begins... and what BA makes simple(r). Now you do not need to write code for any of it. Links and video in the comment below.

Four posts this week. One argument. 1. IR ≈ IC × √N Edge comes from breadth, not strength. Fifty signals at IC 0.05 produce roughly three times the risk-adjusted edge of one signal at 0.10. 2. N_eff = N / (1 + (N−1)ρ) Correlation caps breadth. As you add strategies, effective bets converge to 1/ρ and stop. At ρ = 0.30 you can never exceed 3.3 real bets no matter how many systems you own. 3. Sₛ > Sₚ × (√(2 + 2c) − 1) What a candidate is worth is set by its correlation to your book, not by its own backtest. A 0.70 Sharpe strategy at 0.80 correlation takes a 1.00 Sharpe book down to 0.90 despite being a winner on its own. 4. Contracts = (Equity × Risk%) / (ATR × PointValue) Fixed size is a leveraged bet on calm markets. ES ATR went from 20 to 200 in Q1 2020. Same two contracts, ten times the risk. Size based on volatility. Here is what ties them together. In all four, the number that decides the outcome is not on the strategy report. Your Sharpe does not tell you your portfolio value. Your strategy count does not tell you your breadth. Your contract count does not tell you your risk. Standalone metrics describe a strategy in isolation. You do not trade anything in isolation. Portfolio fit and proper sizing is more of the game than you realize. Thanks for all the likes, retweets and follows this week.

Yesterday I posted IR ≈ IC × √N. That version assumes your signals are independent. They almost never are. Here is the formula that accounts for correlation: IR ≈ IC × √(N / (1 + (N−1)ρ)) ρ = average pairwise correlation across your strategies The term under the root is effective breadth. It is how many real bets (unique) you have, not how many strategies are in your portfolio. N_eff = N / (1 + (N−1)ρ) Run it: 5 strategies at ρ = 0.00 → 5.0 effective bets 30 strategies at ρ = 0.30 → 3.1 effective bets Five genuinely uncorrelated strategies beat thirty slightly correlated ones. The crossover is at ρ = 0.17, so almost any real portfolio is on the wrong side of it. Worse, effective breadth has a hard ceiling. As N grows, N_eff converges to 1/ρ. ρ = 0.30 → ceiling of 3.3 effective bets ρ = 0.20 → ceiling of 5 ρ = 0.10 → ceiling of 10 At ρ = 0.30 your 200th strategy takes you from 3.26 to 3.30. You are paying commissions and carrying operational risk for a rounding error. This is why portfolio construction beats strategy optimization. Your 31st variation of the same trading system does nothing. One strategy with a genuinely different return driver lowers ρ, and lowering ρ raises the ceiling for every strategy you already own. Long story short, the portfolio approach is correct. But you can’t just mash things together. They need to be genuinely different to capture the desired effect!

Almost no single signal is strong. The best individual signals at top desks have an Information Coefficient (IC) of only 0.05–0.15 (weak edge). Edge typically comes from combining dozens or hundreds of weak, independent signals. It is actually the Fundamental Law of Active Management: IR ≈ IC × √N IR = Information Ratio (risk-adjusted edge or consistency of alpha) IC = average strength of your signals N = number of genuinely independent signals (after removing correlations) 50 signals with IC=0.05 → IR ≈ 0.35. One strong signal with IC=0.10 alone → IR=0.10. The diversified system is ~3x more powerful. Correlation is the silent killer. What looks like 20 independent ideas is often only 6–8 after you remove shared variance. Sizing as if you have 20 independent bets = massive over-leverage → blowups when correlations spike (exactly what happens in leverage cycle crashes). Overoptimizing to improve a single strategy is often wasted time. The math shows it is far simpler to find and combine uncorrelated strategies to improve overall IR.

Some example strats and alt data uses → buildalpha.com/commitment-of… → buildalpha.com/market-breadt… → buildalpha.com/trading-weath…

AI monitors Build Alpha's automated strategy discovery and when strategies are not significantly improving, AI has free reign to → change simulation settings → modify order types → modify parameter ranges → create its own signals and restart the strategy generation process. No more perfect settings. No more ideas needed. No more prompting or coding one strategy at a time. A continuous quant research desk working for you 24/7. DM or email to get in the beta. Expanding soon..