Breakouts drove the week while sweet spot setups stayed selective

Weekly Research Report: June 29 to July 03, 2026

THE EDGE THIS WEEK

The dominant pattern this week was selectivity over breadth. Backtest results stayed strong across the full sample, but live scanner conversion was weak, which suggests that many setups met structural criteria without receiving the participation needed to follow through.

The clearest research takeaway is that quality concentrated in a narrow group of symbols, while the broader opportunity set produced too many failures. In practical terms, edge existed, but it was unevenly distributed and punished traders who treated all qualifying setups as equal.

BY THE NUMBERS

The headline divergence was between historical system strength and current scanner efficiency. A 64.7% average backtest win rate across 2,042 trades is healthy, but a 19.9% live breakout rate shows that this week's opportunity flow was noisy and less reliable in real time.

That gap matters. It usually points to an environment where entries still look statistically valid in aggregate, but timing and confirmation become more important than raw setup count. Traders looking for fresh opportunities should see live setups in the scanner with extra attention on confirmation quality rather than frequency alone.

SWEET SPOT REPORT

The sweet spot condition underperformed sharply. Setups with pressure below 60 and 3 to 5 bars produced only a 34.2% win rate this week versus a historical 58.8% across 187 trades, which is a meaningful breakdown rather than random drift.

This kind of underperformance usually suggests that low-pressure consolidations were not resolving with enough sponsorship. In stronger tape, these quieter structures often act as efficient continuation bases. This week, they behaved more like indecision zones, with insufficient urgency after trigger.

The implication is not that the setup class stopped working permanently. It is that the usual advantage of low-pressure compression did not translate because the market was not rewarding patience in the expected way. Traders should treat this as a regime warning: when a historically stable subset drops this far below baseline, confirmation variables deserve a heavier weight than pattern purity alone.

SYMBOL SPOTLIGHT

Symbol Spotlight

LLY stands out not just for joining the 100% Club with 9 wins from 9 trades, but for doing so with an average return near 1R per trade. That profile suggests consistency rather than one oversized outlier. When a symbol can repeatedly deliver full-risk-unit outcomes, it usually reflects orderly trend structure and dependable post-entry behavior.

GDX is interesting for a different reason. Its 94.44% win rate across 18 trades came from a larger sample than most of the top list, which gives the result more analytical weight. This implies that the edge was not confined to isolated single-name momentum. There were pockets where repeated participation in the same theme kept paying, and $GDX may have been one of the cleaner expressions of that persistence.

XLB offers the opposite lesson. A 25.0% win rate across 12 trades and a negative 0.5R average points to a pattern class that repeatedly looked acceptable at entry but failed after commitment. When an ETF posts that kind of result, it often reflects broad internal weakness rather than one company-specific distortion. In research terms, it is a reminder that sector-level participation can invalidate otherwise decent chart structures.

Other high-efficiency names such as V, ZM, PANW, and VRT reinforce the same idea. The strongest symbols did not merely win often. They converted setups into high-quality outcomes with strong average R, indicating smooth follow-through rather than choppy survival.

WHAT THE BOTS MISSED

The bots missed 11 trades worth +23.8R, and the distribution of those misses is revealing. With only 1 TP3 runner missed, most of the lost value appears to have come from standard winners rather than rare explosive outliers. That means the issue was not failure to catch unusual expansion. It was over-filtering ordinary opportunity.

The main blocker was rvol_threshold, responsible for 10 of the 11 misses. From a research perspective, that suggests relative volume requirements may have been too strict for this week's tape. In quieter or less decisive environments, waiting for elevated volume can improve quality. But when strong names advance in a more orderly fashion, the same filter can reject valid breakouts before participation becomes obvious.

The single spy_alignment block is less concerning by itself, but it does raise an important question. If leadership is becoming more selective, strict index confirmation may occasionally suppress good single-name trades that are decoupling from the broader market. Traders who want to compare discretionary versus systematic behavior can watch the bots in the Edge Lab and evaluate how filter strictness changes opportunity capture.

SECTOR HEAT MAP

Sector Heat Map

Breakouts concentrated most heavily in Technology with 15, followed by ETFs and Financials with 10 each. Healthcare posted 8, while Consumer-related categories were split across two labels with 8 and 7 respectively. Industrials and Chinese ADRs were materially quieter at 3 each.

The pattern here is mixed leadership rather than clean sector-wide dominance. Technology leading the list is notable, but not decisive enough on its own to imply broad momentum expansion across the full growth complex. The presence of ETFs near the top supports the idea that some of the tradeable movement was thematic or basket-driven rather than purely single-name.

Financials matching ETFs at 10 breakouts adds another layer. It suggests rotation did exist, but the weak overall scanner breakout rate indicates that sector activity did not consistently translate into durable follow-through. In other words, concentration was visible, but persistence was inconsistent.

For research purposes, the best interpretation is that traders were better served focusing on local strength clusters than assuming broad market confirmation. Sector concentration pointed to where attention should go, but symbol selection inside those sectors remained the deciding variable.

RESEARCH NOTE

Research Note

This week offers a useful reminder that pattern statistics operate on two layers: setup validity and environment fit. The backtest universe remained strong enough to confirm that the underlying framework still has edge. But live execution conditions weakened enough to drag down breakout conversion and crush the sweet spot subset.

That divergence is the real lesson. Traders often respond to a weak week by questioning the setup itself, when the better question is whether the market is rewarding that setup's required path. Low-pressure bases, for example, depend on orderly accumulation becoming visible after entry. If the environment is producing hesitation instead of expansion, the same setup can move from favorable to fragile without changing its visual structure.

The deeper takeaway is that regime detection should not rely only on index volatility. Even with an unknown vol regime, the internals of this dataset already describe the market's behavior. Strong backtest expectancy, weak live conversion, poor sweet spot performance, and high missed R from volume filtering together point to an environment where edge existed in pockets but required more adaptive confirmation logic.

In practical terms, next week's research focus should be simple: watch whether selective leadership broadens, or whether the market continues to reward only a handful of clean names while punishing average-quality setups. If the latter persists, the advantage stays with traders who rank signals aggressively instead of trading scanner output at face value.