The Lab
Every trading day at 9:25 ET the Lab forecasts where SPY will be at 10:00, noon, 2:00 and the close, then learns from what actually happened. The bar it has to clear is the simplest honest forecast: the latest pre-market price holds.
Skill at the close +1.5% ± 1.1% · promising, not proven
Edge hit rate 63% direction beyond the latest price, 60 leaned days
Range coverage 71% target 68%
Days learned 129 first 15 are warmup and not counted
Call
Mon, Jun 8, 2026 Graded All inputs inClose target $740.34, range $734.62 to $746.10, 68% chance of closing above the prior close ($737.55). Closed at $739.22.
| Checkpoint | P(up) | Target | Range | Actual | In range | vs naive |
|---|---|---|---|---|---|---|
| 10:00 | 70% | $739.08 | $736.28 – $741.89 | $742.59 | No | Beat (0.615 vs 0.648) |
| Noon | 70% | $739.67 | $736.00 – $743.36 | $743.51 | No | Beat (0.489 vs 0.603) |
| 2:00 | 68% | $739.76 | $734.75 – $744.81 | $741.08 | Yes | Beat (0.227 vs 0.298) |
| Close | 68% | $740.34 | $734.62 – $746.10 | $739.22 | Yes | Lost (0.236 vs 0.235) |
pushes downpushes up
- VIX futures (VIXY) yesterday
- IWM minus QQQ yesterday
- SPY yesterday
- Distance to gamma flip
- Put skew
- Dollar (UUP) yesterday
- High yield (HYG) yesterday
- Put/call ratio (log)
Track record
-10.0%-5.0%+0.0%+5.0%+10.0%
naive+1.5%
May 22Jul 16Sep 8Oct 28
40%60%80%100%
68% target75%
May 22Jul 16Sep 8Oct 28
- Kalman: everything, slow memory 13%
- Gaps keep going 11%
- Kalman: options positioning 11%
- Analog days (10 nearest) 11%
- Yesterday's move reverses 9%
- Everyone else 44%
0%50%100%
Apr 27Jun 29Aug 27Oct 28
0%50%100%
22
16
15
19
42
0%25%50%75%100%
Leaderboard
Average CRPS at the close in sigma units (lower is better), and each expert's edge over "latest price holds". Weight is how much the ensemble listens to it today.
| Expert | CRPS | vs naive | Weight | Verdict |
|---|---|---|---|---|
| Kalman: everything, slow memory | 0.4713 | +0.0063 ± 0.0097 | 13.1% | ~970 more days to confirm |
| Gaps keep going | 0.4724 | +0.0052 ± 0.0075 | 11.5% | ~847 more days to confirm |
| Kalman: options positioning | 0.4734 | +0.0041 ± 0.0077 | 11.4% | ~1469 more days to confirm |
| Yesterday's move reverses | 0.4766 | +0.0009 ± 0.0044 | 9.2% | ~9974 more days to confirm |
| Latest price holds (random walk) | 0.4775 | — | 9.1% | the benchmark |
| Analog days (10 nearest) | 0.4776 | -0.0001 ± 0.0080 | 10.7% | no edge so far |
| Kalman: price action | 0.4790 | -0.0015 ± 0.0094 | 7.2% | no edge so far |
| Kalman: Jev text reads | 0.4804 | -0.0028 ± 0.0036 | 7.8% | no edge so far |
| Dealer hedging drift | 0.4830 | -0.0055 ± 0.0046 | 6.6% | no edge so far |
| Yesterday's direction continues | 0.4854 | -0.0079 ± 0.0045 | 6.3% | no edge so far |
| Kalman: everything, fast memory | 0.4897 | -0.0121 ± 0.0148 | 3.4% | no edge so far |
| Gaps fade | 0.4985 | -0.0210 ± 0.0079 | 3.7% | no edge so far |
By checkpoint
| Checkpoint | Skill | From range | From edge | Coverage | Edge hit rate |
|---|---|---|---|---|---|
| 10:00 | +10.4% ± 1.8% | +2.2% | +8.1% | 73% | 77% (70) |
| 12:00 | +1.6% ± 1.0% | +0.7% | +0.9% | 69% | 65% (37) |
| 14:00 | +2.0% ± 0.9% | +1.8% | +0.2% | 71% | 52% (31) |
| Close | +1.5% ± 1.1% | +0.8% | +0.7% | 71% | 63% (60) |
How it learns
- Senses. Options positioning from the 9:25 snapshot, yesterday's moves in stocks, bonds, credit, the dollar and VIX, and Jev's reads of the morning headlines and WSB.
- Experts. Simple fixed hypotheses ("gaps fade"), Kalman filters whose weights drift slowly with the market, and an analog expert that asks what happened on the most similar past days.
- Ensemble. Experts earn weight by forecasting better (Hedge), and every expert keeps a small share so one that fell out of favor can come back when the market changes.
- Range. The range widens after a close lands outside it and narrows after one lands inside, so over time about 68% of closes fall inside it.