From 807eb1b506d0482a4d2bb76001b553919f4b781d Mon Sep 17 00:00:00 2001 From: Theodore Song Date: Tue, 18 Aug 2026 10:01:43 -0400 Subject: [PATCH] Fix adaptive confidence and trading costs --- README.md | 14 ++++++++++---- index.html | 56 +++++++++++++++++++++++++++++++++++------------------- sw.js | 2 +- 3 files changed, 47 insertions(+), 25 deletions(-) diff --git a/README.md b/README.md index 946b50c..b7e5acd 100644 --- a/README.md +++ b/README.md @@ -15,7 +15,7 @@ https://polymarket-site-eta.vercel.app/personal.html The site fetches live Polymarket markets, generates agent suggestions, lets you run frequent paper cycles, and syncs the shared arena state through Neon or -Vercel Blob. Engine v38 also installs an offline app shell and caches timestamped +Vercel Blob. Engine v39 also installs an offline app shell and caches timestamped market snapshots. During an outage, cycles continue locally; cached entries are allowed for 90 minutes, older snapshots become mark-only, and all cached data expires after 24 hours. @@ -26,7 +26,7 @@ small samples toward neutral, caps sizing changes to 0.68x-1.30x, and reserves 15% of candidates for deterministic exploration so a stale regime cannot become permanent. -Engine v38 treats each binary stake as capable of falling to zero even when the +Engine v39 treats each binary stake as capable of falling to zero even when the 18% stop cannot fill. New core positions are capped at 2.5%-4% of equity and aggressive positions at 3%-5%, with lower limits for near-term, extreme-price, reversal, and fast-moving setups. Oversized positions inherited from older @@ -55,7 +55,7 @@ evidence. Set `EVAL_MARKETS`, `EVAL_CONCURRENCY`, `EVAL_HORIZONS`, or `EVAL_COST_CENTS` to change the audit. The first 80-market audit found that reversal signals lost 4.34% on average in both chronological partitions, while crypto and longshot samples were also -negative overall. Engine v38 therefore blocks reversal and sports-trend entries outside the fixed +negative overall. Engine v39 therefore blocks reversal and sports-trend entries outside the fixed 15% exploration lane and applies modest sizing penalties to crypto and longshots. It does not boost any rule from this audit because no positive rule was robust across the chronological split. @@ -66,10 +66,16 @@ segment and averaged -4.13%. Sports trends were negative in train and test and averaged -3.53% at 72 hours. Politics trends were the sole cohort with positive row-level returns in all three 72-hour segments, but its market-cluster interval still crossed zero; that supports a longer hold test, not a larger entry bet. -Engine v38 gives Politics trend positions that 72-hour observation window before +Engine v39 gives Politics trend positions that 72-hour observation window before ordinary signal exits. Stops, profit locks, settlement handling, and risk-budget reductions remain immediate. +Engine v39 also subtracts a half-cent round-trip cost when grading each live +walk-forward signal. Confidence uses the largest independent matching bucket, +not the sum of five overlapping feature buckets, and evidence from older engine +versions is down-weighted. This prevents a handful of duplicated observations +from authorizing larger positions or hiding a modest negative regime. + Run `npm run evaluate:settlements` to evaluate fixed decisions made 1, 3, 7, 14, 30, and 90 days before known binary settlements. The audit uses one observation per resolved market and horizon, includes losing contracts at zero, diff --git a/index.html b/index.html index d52e617..d4bdf42 100644 --- a/index.html +++ b/index.html @@ -341,7 +341,7 @@ footer{margin-top:34px;padding-top:22px;border-top:1px solid var(--border);color