Breakouts dominated the week while sweet spot setups stayed selective
Backtest patterns, sweet spot performance, and missed-trade analysis for September 28 to October 02.
THE EDGE THIS WEEK
Participation Was Broad, Conversion Was Selective
This week showed a familiar split between activity and efficiency. The scanner stayed active with 448 setups, but only 22.3% converted into breakouts, which suggests breadth of opportunity was not matched by follow-through quality.
The strongest pattern was concentration of successful behavior in a narrow group of symbols and themes, especially selective technology names and a few standout defensive or commodity-linked pockets. In contrast, several widely watched large caps and retail-linked names produced repeated failure, reinforcing that this was a market where selectivity mattered more than raw setup volume.
BY THE NUMBERS
Core Weekly Metrics
The top-line read is mixed but informative. Backtest win rates remained healthy across the broader universe, yet the live scanner saw a relatively low breakout conversion rate, indicating that many setups triggered in conditions that did not support sustained expansion.
For traders reviewing see live setups in the scanner, the implication is straightforward: this was not a week to equate setup count with edge. The environment rewarded waiting for cleaner leadership and avoiding lower-conviction breakouts that looked acceptable structurally but lacked sponsorship after trigger.
SWEET SPOT REPORT
The Usual Compression Edge Did Not Show Up
The sweet spot setup underperformed materially versus its historical baseline. A 39.4% win rate against a longer-term 55.9% win rate is not noise on its own. It suggests that low-pressure, short-duration compression was less predictive than usual.
That pattern matters because sweet spot logic depends on stored energy converting into directional expansion. This week, compression often resolved into hesitation or failure instead of clean continuation. In practical terms, traders should not assume that all tidy 3 to 5 bar pauses carry the same expectancy. Context still dominates structure.
Research takeaway: when sweet spot performance drops this far below baseline, it often means the market is accepting setups mechanically but rejecting follow-through behaviorally. The pattern is still there, but the participation behind it is weaker.
For anyone monitoring developing patterns, see live setups in the scanner and compare whether current compressions are occurring in true leadership names or in symbols merely pausing inside unstable trends.
SYMBOL SPOTLIGHT
Where the Signal Was Cleanest
NOW stands out not just for a 100.0% win rate across 11 trades, but for the quality of its average outcome at 0.9993R. That combination implies both consistency and clean path efficiency. It is one thing to avoid losses over a small sample. It is another to do so while nearly capturing a full R on average across repeated opportunities.
GDX is interesting for a different reason. Its 95.45% win rate across 22 trades gives it one of the larger meaningful samples among top performers. That makes it less of an outlier and more of a candidate for pattern confirmation. When a symbol produces high conversion across a larger number of tests, it often reflects sustained thematic sponsorship rather than a one-off streak.
COST deserves attention on the other side of the distribution. A 16.67% win rate across 12 trades with an average return of -0.6669R signals repeated rejection, not random underperformance. Weak names often reveal more about market character than strong ones. In this case, the data suggests that some mature, heavily watched names were poor vehicles for breakout participation even when the setup shape appeared valid.
Other notable names included LLY, CVX, ZM, and ZS, while weak behavior persisted in XRT, MA, and RTX. The pattern across both lists points to a market where relative strength and thematic alignment mattered more than brand recognition or index prominence.
WHAT THE BOTS MISSED
No Operational Leakage This Week
The cleanest operational result in this report is also one of the most useful. With zero missed trades and no R left on the table, execution did not distort the research read. That means weak scanner conversion cannot be blamed on process gaps or excessive filtering.
This is important because it sharpens the conclusion: the issue was market quality, not system participation. When the bots are capturing everything available and the output still shows low breakout conversion, traders should focus their review on setup context, not execution mechanics. You can watch the bots in the Edge Lab to study how disciplined participation behaves during lower-efficiency weeks.
SECTOR HEAT MAP
Breakouts Clustered in Technology and ETFs
Technology led the breakout count with 31, well ahead of every other category. ETFs followed with 15, then Financials with 9, while Consumer and Industrials each contributed 7. Chinese ADRs produced 6, Healthcare 5, and an additional consumer bucket added 6 more, suggesting some classification overlap but still confirming broader participation outside a single theme.
The interpretation is less about headline leadership and more about concentration. Technology generated the largest number of breakouts, but the broader scanner data shows that not all of that activity translated into uniformly strong conversion. In other words, tech had the most opportunities, but traders still needed to separate clean leaders from noisy participants.
When breakout counts cluster in one sector while overall scanner conversion stays low, the usual failure mode is overtrading the theme. The better approach is to narrow further into names with repeatable follow-through rather than assume the whole group shares the same edge.
The ETF count is also meaningful. Elevated ETF breakouts often point to macro or basket-driven participation rather than purely stock-specific conviction. That can support directional movement, but it can also flatten dispersion and make individual setups more sensitive to broader market reversals.
RESEARCH NOTE
Good Backtests and Weak Live Conversion Can Coexist
The most important research question this week is why a universe with a 64.2% average backtest win rate produced only a 22.3% breakout rate in the scanner. The answer is that these numbers measure different stages of edge. Backtests summarize pattern behavior across a larger sample, while weekly scanner conversion measures the market's willingness to reward those patterns right now.
That gap is where traders often lose discipline. They see a strong historical edge and assume current conditions will express it immediately. This week argues for a different approach. Historical edge should define what you are willing to trade, but live conversion should determine how aggressively you trade it.
In practical terms, when current breakout rates are this low and sweet spot setups are running well below baseline, the research-backed adjustment is not to abandon the system. It is to tighten selection, reduce assumptions about follow-through, and emphasize symbols already proving they can convert efficiently.