This year I ran a strategy across 4.4 million one-minute candles, on three different market regimes: a rising market (gold +128%), a falling one (gold -42%) and a flat range. Almost nothing I learned was about win rate. 1) WIN RATE WITHOUT EXPECTANCY IS NOISE A 70% win rate with a 1:0.3 payoff loses money. A 35% win rate with a 1:3 payoff pays. The number that decides is expectancy: (win rate x average win) - (loss rate x average loss). If your journal does not show it, it is a diary, not a journal. 2) YOUR COST PER TRADE IS BIGGER THAN YOU THINK, AND IT CHANGES THE ANSWER I measured my own friction from a broker export: commission on more than 3,000 positions, plus the median spread. Total, 3.11 pips round trip. At that cost a 20-pip stop turned a winning system into a losing one, -0.055R, and a 40-pip stop turned it back, +0.129R. Same entries, same exits. Only the cost assumption changed. Measure yours before you optimise anything else. 3) ONE GOOD MARKET MAKES EVERY STRATEGY LOOK BRILLIANT Raw profit factor, 1.87. Half of it came from gold trending up for four years. Tested on a falling market and on a range, the number I could defend was 1.33. If your sample covers one regime only, you do not have an edge yet, you have a trend. 4) R MULTIPLES FROM OTHER PEOPLE ARE NOT WHAT YOU THINK Educators often publish R as maximum excursion, meaning how far the trade went your way before you closed it. Realised R is a fraction of that. On the sample I checked, the ratio between the two was around 4.3. Never quote a number without its denominator. WHAT TO DO THIS WEEK Export your last 100 trades and compute three things: expectancy in R, average cost per trade in R, and both of those split by market condition. If that takes you more than twenty minutes, your journaling is the problem, not your strategy. Post your numbers here and I will go through them with you. That is what this group is for. Full disclosure so nobody has to guess: I build Tradoshi, a journal that connects to your broker and computes all of the above on its own. https://tradoshi.com/en/ - a spreadsheet gets you to the same place, it just takes longer.