bfd725621d
#24 v0: three-book reconciliation — backtest story is 82-112% unseen+missed; +312% is decoration, rankings only
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-07-23 19:15:00 -04:00
ecd9f49593
copybot: live paper feed [skip ci]
2026-07-23 23:08:10 +00:00
cf734a1aa9
companion missed rows carry entry price
2026-07-23 19:07:21 -04:00
d2f58c5fef
copybot: live paper feed [skip ci]
2026-07-23 23:05:35 +00:00
8edcabac7d
copybot: live paper feed [skip ci]
2026-07-23 23:03:33 +00:00
2c87bb8c5d
full missed-row companion feeds (portfolio[_follow]_missed.json): wallet-card would-be totals now reconstructable row-by-row in the dash modal
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-07-23 19:01:17 -04:00
d638ff3abb
follow-only backtest book (--follow-only -> portfolio_follow.json, daily-wired): followed-7 replay +312% vs +41% bench — bench crowding was starving the follow set
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-07-23 18:49:37 -04:00
b962ba9bcc
copybot: live paper feed [skip ci]
2026-07-23 22:41:31 +00:00
0c18b5b2e5
backtest book regenerated under band-guarded mirror (sells >=90c)
2026-07-23 18:37:56 -04:00
3fc7cafc46
copybot: live paper feed [skip ci]
2026-07-23 22:35:26 +00:00
cd0bc9ff00
copybot: live paper feed [skip ci]
2026-07-23 22:34:54 +00:00
54f4c511ab
band-guarded mirror exits ( #21 successor): mirror sells >=90c only, sub-band sells ignored+ledgered; sell-band study slice; backtest parity
...
Sell-band evidence: every band <90c LOSES by mirroring (<30c +$79/sell
given up, 50-70c +$7.29); >=90c SAVES (-$0.85/sell, n=68).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-07-23 18:34:20 -04:00
a534766ede
copybot: live paper feed [skip ci]
2026-07-23 22:30:30 +00:00
94af7e9d15
backtest book regenerated under maker/hold parity
2026-07-23 18:21:40 -04:00
7fe52849a6
copybot: live paper feed [skip ci]
2026-07-23 22:19:18 +00:00
7baabe2286
copybot: live paper feed [skip ci]
2026-07-23 22:16:32 +00:00
004911bba3
backtest execution follows the paper bot's follow flags (#20/#21 parity): maker entries at the sharp's price, hold-through exits; sold-unresolved carry as open at exit print
...
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-07-23 18:09:11 -04:00
7abde0ab7f
copybot: live paper feed [skip ci]
2026-07-23 21:37:10 +00:00
97737fee2d
copybot: live paper feed [skip ci]
2026-07-23 20:17:27 +00:00
60fab12e8b
copybot: live paper feed [skip ci]
2026-07-23 20:17:11 +00:00
32f2946b5d
copybot: live paper feed [skip ci]
2026-07-23 19:55:30 +00:00
e36eb7d2c2
copybot: live paper feed [skip ci]
2026-07-23 19:55:16 +00:00
829e7634dd
copybot: live paper feed [skip ci]
2026-07-23 19:46:43 +00:00
6dec0eec63
copybot: live paper feed [skip ci]
2026-07-23 19:46:19 +00:00
05a42ea605
copybot: live paper feed [skip ci]
2026-07-23 19:44:33 +00:00
c743711927
copybot: live paper feed [skip ci]
2026-07-23 19:44:06 +00:00
9799368af0
copybot: live paper feed [skip ci]
2026-07-23 19:33:55 +00:00
2fd88aa4b3
copybot: live paper feed [skip ci]
2026-07-23 19:33:32 +00:00
ce59ebaa88
copybot: live paper feed [skip ci]
2026-07-23 19:01:08 +00:00
e1700e88c8
copybot: live paper feed [skip ci]
2026-07-23 19:00:55 +00:00
96aa0426ad
copybot: live paper feed [skip ci]
2026-07-23 18:36:39 +00:00
0108ccadb3
copybot: FLIP #20 + #21 — entry_mode=maker, exit_mode=hold on both books (forward windows open at this commit)
...
User-directed flip 2026-07-23 (ahead of the Friday read; the user go is
the gate, per both pre-registrations). Maker entries: GTC at the sharp's
price, 60s registry TTL, no chase. Hold-through: mirrored exits ignored
and ledgered to copybot_ignored_exits[.live].jsonl as the live
counterfactual. Verdict bars unchanged: #20 keep/revert at n>=30 resolved
(fill>=70% or ROI>=parity edge / revert fill<55%@40, ROI<=0@30); #21
keep/revert at n>=30 resolved ignored-exits (hold>=mirror / revert
-$50@30 or 25% attributable DD).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-07-23 14:36:12 -04:00
2d3c5dbfa8
copybot: live paper feed [skip ci]
2026-07-23 17:31:01 +00:00
2b37d29b11
copybot: live paper feed [skip ci]
2026-07-23 16:50:32 +00:00
cc8dfa7623
copybot: live paper feed [skip ci]
2026-07-23 16:50:11 +00:00
9afe5e88ba
copybot: live paper feed [skip ci]
2026-07-23 16:43:20 +00:00
b146a6595c
copybot: live paper feed [skip ci]
2026-07-23 16:41:41 +00:00
826bb6891c
copybot: live paper feed [skip ci]
2026-07-23 16:31:00 +00:00
6ef8898568
copybot: live paper feed [skip ci]
2026-07-23 16:30:09 +00:00
9688daea13
copybot: live paper feed [skip ci]
2026-07-23 16:19:19 +00:00
2bcd9940d4
copybot: live paper feed [skip ci]
2026-07-23 16:18:14 +00:00
354a0c8a13
copybot: live paper feed [skip ci]
2026-07-23 16:17:07 +00:00
48e941dd94
copybot: add JuiceFarm (7th seat, auto p80 floor) to paper + live follow sets
...
Bench promotion per the #13 method run early: +17.4%/+33.5% both windows
(n=5), insider z 2.5 (p_luck 0.7%), 64% day-of entries, 1% refunds.
gkmgkldfmg stays per user call. Live mirrors paper exactly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com >
2026-07-23 12:16:25 -04:00
2228b9fae7
copybot: live paper feed [skip ci]
2026-07-23 15:45:05 +00:00
702fec7a65
copybot: live paper feed [skip ci]
2026-07-23 15:40:42 +00:00
3ffd71092c
copybot: live paper feed [skip ci]
2026-07-23 15:38:25 +00:00
dd2561051e
copybot: live paper feed [skip ci]
2026-07-23 15:27:30 +00:00
4a83efee79
copybot: live paper feed [skip ci]
2026-07-23 15:25:40 +00:00
e0704d353e
copybot: live paper feed [skip ci]
2026-07-23 15:23:34 +00:00
86550ef489
copybot: live paper feed [skip ci]
2026-07-23 15:23:10 +00:00
jaxperro