Each January, put everything into the S&P 500 stock that rose the most the year before and hold it for twelve months. Sell early only if it falls 35% below the price paid.
Hold. The stop needs a 88% fall from here to fire. It has not traded below its January price this year (lowest: +2.8%).
Chance each stock finishes 2026 first. Moderna leads Sandisk by 192 points with 52 sessions left. At today's price its stop would sit near $141.71.
Every S&P 500 member ranked by 2026 price gain; first place at the Dec 31 close is the January buy. Chance of #1 replays the last six months of real daily moves 4,000 times over the sessions left, with each stock's trend removed so nobody gets credit for momentum continuing.
| # | Stock | 2026 gain | Path | Low vs Jan | Off high | Chance of #1 |
|---|---|---|---|---|---|---|
| 1 | MRNA Moderna |
+639.3% | +1.1% | −10.0% | 55% | |
| 2 | SNDK held Sandisk |
+447.4% | +2.8% | −44.8% | 32% | |
| 3 | DELL Dell Technologies |
+258.8% | −12.4% | −24.1% | 6% | |
| 4 | MU Micron Technology |
+226.3% | +3.3% | −25.8% | 6% | |
| 5 | MPC Marathon Petroleum |
+194.0% | −0.2% | −1.5% | <1% | |
| 6 | VLO Valero Energy |
+173.2% | −0.4% | −3.6% | <1% | |
| 7 | MRVL Marvell Technology |
+168.1% | −16.8% | −30.9% | <1% | |
| 8 | BE Bloom Energy |
+154.7% | +2.3% | −37.0% | <1% | |
| 9 | AMD Advanced Micro Devices |
+146.9% | −12.1% | −19.7% | <1% | |
| 10 | INTC Intel |
+143.5% | +2.3% | −36.9% | <1% | |
| 11 | STX Seagate Technology |
+132.0% | +0.8% | −44.2% | <1% | |
| 12 | HPE Hewlett Packard Enterprise |
+129.5% | −17.4% | −25.1% | <1% | |
| 13 | PSX Phillips 66 |
+127.2% | −0.2% | −1.4% | <1% | |
| 14 | LITE Lumentum |
+123.6% | −13.9% | −27.9% | <1% | |
| 15 | FTNT Fortinet |
+113.0% | −7.4% | −14.2% | <1% |
Test period: January 1997 to Oct 16, 2026. Benchmark: SPY with dividends reinvested (2026 so far is price only). Returns are price-only through 2011 and total return from 2012, matching the source exhibits. No taxes or trading costs. The stop sold in 2000, 2002, 2008 and 2011, and has not fired since 2011.
The stock held each year and its return. Faded bars show the loss the stop avoided. Returns above +260% are cut off at the top.
Ending value of $100 at each stop level. Tighter stops sell winners on ordinary dips; looser ones ride each crash further down. 35% sits just past the deepest dip any non-collapse year produced (Reebok, −32.3% in 2001) and short of the shallowest collapse (Netflix, −64.5% in 2011). Faded segments: years without intraday data could pull a tight stop down to the low end.