I think one could quite confidently call this a stock pickers market since there is more idio dollar spread available than in any non-crisis year since 2002 (see first chart comparing demeaned, windorized ESTU total returns decile ranked in 2019, 2025, 2026 (all through the 1/1-10/12 YTD period). We note the total return spread between top and bottom decile is larger in 2026 than either of the other two sampled years which further supports this thesis. However, a larger proportion of the x-sectional volatility this year has been factor-driven which the next two charts further expand on. The second chart tracks how far apart stocks have drifted from each other in 2026, split into the component explained by factor exposures and the component that is stock-specific. The dashed lines show where each would be if no moves had reversed. From this we note that idio dispersion has compounded steadily all year while factor dispersion ran ahead of its line until June 22, then collapsed through late July. The third chart shows stocks are ranked by how much factor exposure helped or hurt them through June 22. From this we note that the biggest factor winners had gained 47pp, then gave back 25pp by July 29. The biggest losers recovered 11pp. Every decile reversed in order meaning this was a systematic unwind. The takeaway here is that through June, factor tilts created a 79pp spread between stocks and almost half of that reversed in five weeks, whereas the same window in 2025 gave back 1%. So pnl generated by stock-specific returns were relatively more stable, while gains carried by factor exposure largely have not been. This all generally supports the notion that there IS spread to capture in 2026 but that a sloppy approach to this has resulted in underperformance relative to the market given the broad factor return reversal the market has been experiencing in 2H 26'. Methodology: Top 3,000 stocks by market cap in the risk model's US estimation universe, Jan to Oct 12 each year: daily returns are split into factor (total minus specific) and idio (risk model specific), winsorized at the 1st/99th percentile and demeaned across stocks each day to remove the market. They are then summed additively to compare cross-sectional dispersion, decile spreads and the build-up and reversal of factor vs idio returns against 2025 and 2019.