How to test an idea before you risk money on it.

ADX trend breakout and Keltner channel breakout post nearly identical win rates and risk to reward ratios on the same 220-stock universe. Their median tickers land 0.66 percent and 2.33 percent, more than three times apart, and the gap traces back to how often each rule actually trades.

Dual momentum 12 minus 1 trades about once a year per ticker, keeps one of the smallest drawdowns and tightest tails on this site, and its win rate and risk to reward ratio both look healthy. The median ticker still finished the decade slightly underwater.

Two studies with almost the same win rate and almost the same risk to reward ratio produced very different returns and very different drawdowns. Trade frequency is the number that actually moved, and it did not move by the same multiple as the results it left behind.

Sorted by average maximum drawdown, the safest-looking strategy on this site is also the only one of seven compared here that lost money, and the strategy with the deepest drawdown has the best return.

Every other strategy study on this site has an average return sitting above its median, pulled up by a handful of large winners. Bollinger squeeze breakout runs the other way: its 0.27 percent median beats its 0.17 percent average, because its worst tickers, not its best, are doing the pulling.

Among the fifteen largest single positions tracked super investors currently hold, one name is shared by 30 other managers. Six others are held by no other tracked manager at all. The crowded ones are mega caps. The uncrowded ones are the real idiosyncratic bets.

The single number at the top of a backtest is the least useful thing on the page. The shape underneath it tells you whether the edge is real or whether two lucky stocks carried the whole result.

Thirty four of the eighty one super investors this site tracks hold Microsoft. That is the most crowded name in the group, and crowded is not the same word as convincing.

52 week high momentum's best single result is one NVDA trade returning 3300 percent. It helps explain the study's 94.33 percent average far more than its 48.05 percent median suggests, and it still captured only a fraction of what buying and holding NVDA would have.

Two managers, the same 76-investor pool, two rankings that crown almost opposite winners. Burry's typical trade compounds faster than all but one tracked investor. Buffett's does not, and his headline weighted return of 98.8 percent is the reason both facts can be true at once.

Golden cross missed almost all of NVDA's, TSLA's, and AMD's decade. It also avoided most of WBD's collapse and turned a losing OXY position into a winning one. The 9.02 percent average return is what both halves of that trade look like added together.

The worst weighted return Tenachine tracks belongs to a manager with 153 disclosed trades, one of the largest samples on the whole leaderboard. Most of the other nine names on the same bottom-ten list are still profitable. Worst here does not mean losing.

The most active strategy on this site fires roughly 102 trades per ticker over ten years and finishes almost exactly where it started. Trade count turns out to be another number that predicts nothing about the result, in either direction.

Two oversold bounce strategies land within four hundredths of a percent of each other on average return. Everything that usually explains a return, win rate, payoff ratio, trade count and drawdown, is different between them. The same number arrived by two different routes.

Win rate is the easiest number on a backtest report to misread. Four studies on the same 220 stocks show it climbing from 10 percent to 69 percent while return does the opposite, the same and then the opposite again.

Gap down fade wins more than twice as often as golden cross on the same 220 stocks. Its return is a fraction of golden cross's. The payoff ratio behind each win explains why, and it is not close.

The Kelly criterion, run on Li Lu's own 14 disclosed trades, comes out at 62 percent of the portfolio on a single position. The formula is not wrong. It is just answering a narrower question than the one a reader wants answered.

MACD zero cross and MACD signal cross trade the same indicator with two different trigger rules. One fails to complete a backtest on 35 percent of its universe. The other fails on 3 percent. Win rate and drawdown barely differ between them, but return and coverage both do, by a wide margin.

A backtest built from today's list of large caps cannot see the companies that were removed from that list along the way. That gap has a name, a rough size, and a completely different cousin that shows up in the coverage numbers on every study page.

Every one of Tenachine's 22 published strategy studies has an aggregate risk to reward ratio above the 1.0 breakeven line, except one. NR7 volatility breakout sits at 0.6964, and the gap to the next lowest study is not small.

Triple ma pullback's headline average return is positive. Its median ticker lost money. Neither number is wrong, and the gap between them is wider than the usual mean-pulled-up-by-winners story, because here the sign itself flips.

Same channel breakout rule, two lookback windows. The 20-day version returns more than double the 55-day version's average, and its drawdown and failure rate both grow along with it.

Ma crossover 20/50 draws down further on average than any other strategy Tenachine has studied, deeper even than Williams percent R reversal. It also returns less and wins less often. A bigger drawdown did not buy a better trade here.

A one percent risk rule caps what a single trade can cost you. It says nothing about what a losing streak costs, and the gap between those two things is worth walking through with a calculator rather than taking on faith.

One manager's entire disclosed 13F book is three names. Another's ten largest positions do not even add up to his whole book. Both facts come from the same kind of filing, and both are only half the picture.

The single best headline return across every strategy study on this site belongs to the study with the worst completion rate. Read one number without the other and the 94 percent looks like a discovery instead of a warning.

Two managers on the same conviction buy screener, one whose overall record is strong and one whose is not. The screener alone cannot tell you which case you are looking at, and neither can either single position on its own.

Some of the biggest current bets tracked here were opened years or decades ago. Holding period turns out to be another number that cannot be read on its own, the same lesson the site keeps landing on from a different angle each time.

Two top 10 rankings, drawn from the same 81 tracked investors, share exactly one name. The other nine slots on each list are different investors, and the gap between win rate and weighted return explains why.