LessonThis is a lesson restated in new words. Any figure in it is an example from the source, not this desk's record.
A chart of Apple before earnings, a tight range, a gap up, a gain: the lecture plays that story and then throws it out. Saying there was edge does not mean the trade was positive in expectation. There is no other version of that earnings print. Replay the day as a fan of paths and, in the animation, only one path gaps up. The rest gap down. You only ever walk one. Edge on a single fill is not edge over time.
Why it matters
A posted 10,000 percent on a levered option is one path. Positive edge, as a classroom wish, means the average trade makes money, not that every trade wins. If that average is positive, wealth trends up. If not, the account can reach a broke state you do not climb out of. The process is not ergodic: your one path's average need not match the average across imagined paths.
You still cannot use that classroom average as a machine. The inputs change. A good earnings print can gap down. If the direction were known, the lecture asks what a straddle would cost.
The idea in plain words
Write the average as the chance of a win times the typical win, plus the chance of a loss times the typical loss. The trouble is that none of those four pieces is stable, and you do not know what enters them. A line on a chart is not a proof that the line caused anything. The number of ways to choose an entry and an exit inside one bar is, in the lecture's word, unfathomable. A one-hour candle in which someone says they traded a mean-reversion idea is not a recipe you can copy from the sentence.
The lecture dislikes both the claim that a drawn level must bounce and the claim that no drawn level can matter. Whether one does is a research question. A coach can teach the game and still not play your hand. Someone with a repeatable ability is not, here, going to take your risk unless they manage the capital.
The last two problems are the ones they call largest for retail systematic trading. First, implied probability collapses onto one event. You never see the other paths, so you cannot tell whether the trade was positive in the time average of your curve or only in a cloud of alternate days. You will want to have been right. A backtest that was true for ten years need not be true next year, because the convergence assumptions do not hold out here.
Second, strategy criticism moves the goalposts. "Never sell a covered call, it caps your upside" is the example. On the paths in the lecture's picture, most expiries are out of the money. The gap-up path is the one where the cap binds. Walk a gap-down path and you kept the premium and took the stock's drawdown. You did not cap an upside you did not get. "On average you cap it" may be true as a cross-section and still be the wrong sentence about the path you walked. Comparing that path to buy-and-hold raises another unset question: where you bought, and through which date you held. The lecture calls that a trading question, not an investing one, and says calling tops is the trap.
How to use it
Separate one fill from the average. If the win size, the loss size, and the two probabilities are moving, you do not have the stable object the lecture calls edge. When you criticize a strategy, say whether you mean the path that happened. The covered-call complaint fails that test, because most drawn paths never went through the cap.
What usually goes wrong
One green trade, or a screenshot up 10,000 percent, gets called a method. The other paths in the animation were losses.
A ten-year backtest gets treated as convergence. The lecture says the real series does not owe you that.
A rule gets rejected because a slogan says it caps upside, without asking which paths were actually capped. The reverse error is trusting the slogan that names a support level. Both skip the research question.
Questions people ask
Did the Apple earnings trade have edge?
The lecture refuses the question. One path gapped up. The alternate paths in the picture gapped down. There is no counterfactual in the real day, so a single gain is not an average.
What would positive edge have to look like?
A positive average trade, win chance times typical win plus loss chance times typical loss, stable enough to repeat. The lecture says those pieces change, so the classroom trend in wealth is not something you can assume.
Does selling a covered call always cap you?
Only on the paths that finish above the strike. In the picture, most paths expire out of the money. You cap the rally you actually get, not the rally in a path you did not walk.