Correction: the 20 August 2026 report published a stale forecast

Validated over 1051 out-of-sample sessions, walk-forward: each day is predicted training only on earlier ones. It is right 60.04% of the time against 51.09% for a coin flip. Brier 0.23466 against 0.24988 with no model.

The 120 variables it looks at today

Of the 139 computed, the model keeps the 120 most correlated with the target each month, chosen using past data only. These are today's, grouped by source.

This is the only thing that genuinely changes. Price moves every day and adds nothing; a quarterly or annual report rewrites value, quality and growth at once. Here is the before and after, on the 0-10 scale used on company pages, where 5 is the index average.

Filing date on EDGAR, not the quarter end. Scores come from the rankings calendar: the same ones each company page publishes. No opinion on the results, only the measured effect.

The high of the past 23 months was +12.7 points, in the window ending June 16, 2026: nine weeks later the gap is +6.0. At the other end of the same stretch, -8.6 on August 14, 2025.

Over the past month the value portfolio gained +4.14% while momentum gained +5.95%. Across that six-month window the benchmark rose 9.7%, so this is not the whole market moving: it is a rotation inside one.

Computed from carteras.json · 187 monthly rebalances between 2-Feb-2011 and 17-Aug-2026 (fifteen years) · 181 overlapping windows of 6 rebalances · recent stretch: the last 24 rebalances, the shortest run that still holds 4 non-overlapping windows.

The biggest sale of the day tells you nothing. Hershey Trust Co Trustee In Trust For Milton Hershey School, of Hershey, sold 1.9 million dollars of stock and still holds 1,015,749 shares: that is 0.97% of the position, and the sale was scheduled months in advance under a 10b5-1 plan. That is tax plumbing, not a view on the company.

Of the 7 sales filed that day, 3 were discretionary and 4 scheduled. And here is the finding: the two largest sales as a share of holding were both discretionary.

Source: Form 4 filings with the SEC · alertas_insiders.json, session of 19-Aug-2026.

The 90-day context: 647 sales worth 2,988 million dollars (2 with no declared price, outside that total), of which 300 were discretionary, and of those 106 exceeded 25% of the seller's position. That is where to look.

All three beat the benchmark (+494.9%) on cumulative return over fifteen years, but cumulative return is not the proof: the benchmark carries the bias explained below. The weakest is Value 100, with an alpha of -0.24%: negative, meaning what it earns does not pay for the market risk it takes to earn it. Publishing the figure that comes out worst is the only reason the others deserve any credit.

Source: carteras.json · index based at 100 on 3-Jan-2011. Walk-forward with point-in-time data: on each date scoring uses only prices up to that day and fundamentals already filed with the SEC. 187 rebalances through 17-Aug-2026.

The site's other five portfolios are not here. They are not picked by factor ranking, so their history cannot be rebuilt without looking ahead: their figures are today's holdings applied to the past, which inflates them. Until they have a walk-forward test, they stay out of this report.

Source: carteras.json · index based at 100 on 3-Jan-2011. Walk-forward with point-in-time data: on each date scoring uses only prices up to that day and fundamentals already filed with the SEC. 187 rebalances through 17-Aug-2026.

The three windows where momentum pulled furthest ahead of value are +27.8 (9-Jul-2020), +23.7 (8-May-2020), +18.7 (7-Aug-2020). The widest is July 9, 2020, six years ago: the extreme of the series is not a recent event, and the window closing August 17, 2026 is worth +6.0 points.

At the other end, the three where value led: -43.2 (10-May-2021), -35.9 (9-Apr-2021), -32.8 (10-Mar-2021); the widest, May 10, 2021, five years ago.

And after an extreme? Counted, not assumed: of the 4 extremes in momentum's favour with enough data after them, in 4 the lead had flipped sign six months later; of the 5 extremes in value's favour, in 3. That is a count over 9 episodes, not a rule.

Closing levels of ^GSPC, ^VIX and ^TNX (Yahoo Finance) · session of 19-Aug-2026.

The S&P 500 rose 0.21% in the session of August 19, 2026 and the VIX closed at 14.89. None of that explains the factor rotation in section 02, which is why it sits in section six rather than section one.

• Why each executive sold. The SEC requires the filing, not the motive. Anyone who tells you the motive is making it up.
• What insiders did yesterday. Form 4 can be filed up to two business days after the trade, so the freshest data is always the day before last.
• Whether momentum keeps winning. Fifteen years and 187 rebalances are not enough to claim a factor mean-reverts. We can describe what happened, not what comes.
• What any company is worth. This report computes no price targets. No Monte Carlo, no DCF, no fair value. When you see one elsewhere, ask which cost of capital discounted it and who chose that number.
• Whether the S&P rises tomorrow. We measured it with 139 variables over 1051 out-of-sample sessions: the model calls direction right 54.28% of the time and saying "up" every single day is right 54.12%. There is no signal, so we don't sell one.