# A1.27 Buy early on the map + A1.28 non-kit check (Analyst 1, 7 Oct 2026 10:22 Bangkok)

## One line
**A1.27 is a KILL by the pre-registered bar. The fresh one-hand kit rule at $300k makes money in both halves, but it trails the live book (A $2,043 vs $3,117 a day, B $10,546 vs $12,038) and in half B it beats only 80% of random kit draws. A1.28: keep the non-kit coins. They earn +$1,424 a day in half A.**
**Unregistered lead: buying EVERY kit coin at its fair $300k milestone beat the live book in both halves (A $3,416, B $14,471 a day). Pre-registered below for half C. It is not a pass.**

## A1.27 rule (pre-registered 10:16 in my status, picked on half A only)
- Coins: kit (Kit A, Kit B, 3,587 family) AND pre-$300k buyer median nonce at most 1 AND all traced fresh buyers funded by ONE sender (A1.26).
- Entry: first swap 30 s after the fair $300k milestone ($15k bought at $300k+), $1k, 1% fee, price impact from the in-range ETH (virtual reserves).
- Exits = live book: 0.3x stop, 5x, then kit hold with a 50% trail from the peak (48 h cap), pull exit (in-range ETH at half of entry), 12 h time stop before 5x. Every exit sold at the end of the next block after the trigger.
- Universe: every kit coin that reached a fair $300k, 13 Sep to 6 Oct: 85 in half A, 92 in half B. Features for the 109 coins that never reached $1M were traced the same way as A1.26 (7,710 more buyers), so there is no selection on the outcome.
- Days: half A 14 to 30 Sep (17 days), half B 1 to 6 Oct (6 days). Live book on the same halves and days: A2.21 wide trades plus kit hold.

| half | set | trades | mean x | median x | $ a day | worst day | beat random (200 draws) | winners caught | also in live gate | live $ a day | live + rule $ a day |
|---|---|---|---|---|---|---|---|---|---|---|---|
| A | **rule** | 22 | 2.58 | 0.29 | **$2,043** | -$1,486 | 95.0% | 6 of 9 | 4 | $3,117 | $5,160 |
| A | base (no one-sender) | 53 | 1.79 | 0.29 | $2,461 | +$1,843 | 71.5% | 8 of 9 | 15 | $3,117 | $5,578 |
| B | **rule** | 43 | 2.47 | 0.28 | **$10,546** | -$4,256 | 80.5% | 5 of 10 | 10 | $12,038 | $22,584 |
| B | base (no one-sender) | 56 | 2.18 | 0.28 | $11,024 | -$1,697 | 74.0% | 8 of 10 | 12 | $12,038 | $23,062 |

- Verdict by the pre-registered bar: below the live book stand-alone in both halves, and half B beats random only 80.5% (bar 95%). **KILL.**
- Entry caps: the fair $300k mark comes late on kit coins (their dev bundle jumps the price), so the median entry cap is $717k (A) and $556k (B), with a range of $290k to $2.5M.
- The money is in a few runners. Without the top 2 trades the rule makes $610 (A) and $3,240 (B) a day. Most entries hit the 0.3x stop: the median trade is 0.29x.
- Winners caught (entry cap, result): `0x7608c38ea8cc2082b043e886185979942cdeb58e` $326k 28.97x, `0xb4b0dfeaa2e2a838f0f7a304f005ebe32390d4ef` $619k 16.86x, `0xeb1646dbfbeacf76483df1092c208388c029c534` $486k 14.24x, `0x3939ce55ec4e0ff8df9aa6469a795c793da06d55` $1.30M 13.96x (open at data end, valued at the last price), `0x6e8a6a1af8ec53f3f09c7303e5403558dee42d0a` $728k 13.53x, `0xe1f227ad6f50b3c8905e75ed621610becf4510d9` $522k 12.12x, `0x94bbe853481f40d049d98eb830c57276d9727e5b` $1.19M 9.77x.
- The one-sender filter LOWERS money in both halves vs the base group. It picked winners in half A (6 of 9) but cut 3 of 8 winners in half B.

### What is different from the killed A2.26 early entry and A1.15 kit early
- A1.15 bought Kit B at a price mark ($250k cap or $500k sellable) with an exit at the $1M gate price plus 5x, and died on the dev bundle dumping before $1M. A2.26 bought before the gate with no coin filter.
- Here the entry needs real money at the price ($15k bought at $300k+, so the dev jump alone does not count), the fresh-buyer gate, and the live exits including the kit hold. That is why the base group makes money now. It still does not beat random kit draws, so the gate itself adds nothing.

## Unregistered lead (seen after the pricing, so it cannot pass)
- **Every kit coin at its fair $300k milestone, live exits:** A 85 trades, $3,416 a day; B 92 trades, $14,471 a day. Live book on the same days: $3,117 and $12,038. It caught 9 of 9 and 10 of 10 winners.
- **Pre-registered now for half C (from 7 Oct 00:00 Bangkok, fixed):** paper book "kit at fair $300k": every Kit A, Kit B or 3,587-family coin, $1k, buy 30 s after $15k has been bought at $300k+ cap, live exits as above. Test once at 40+ trades. PASS = $ a day above the live book on the same days AND worst day no worse AND beats 95% of 200 random same-count draws of all v3 coins at their fair $300k milestone.
- Overlap risk: about 1 in 4 of these coins are also live-gate trades, so live + kit $300k = two tickets on the same coin.

## A1.28 Drop non-kit coins from the live gate?
| half | live all: trades, $ a day, worst day | kit only: trades, $ a day, worst day | non-kit part: trades, $ a day |
|---|---|---|---|
| A | 48, $3,117, -$2,000 | 22, $1,693, -$2,000 | 26, +$1,424 (mean 1.93x, 10 reached 5x) |
| B | 22, $12,038, -$3,161 | 20, $12,276, -$3,161 | 2, -$238 (mean 0.29x) |

- **Keep the non-kit coins.** Dropping them costs $1,424 a day in half A and gains only $238 in half B (2 trades). They still earn at the 5x sale. No new rule.

## First principles score: 85%
- Strong: every kit coin that reached a fair $300k is in, with its features traced the same way (no outcome selection); entries and exits use the real swap path, block-end exits, 1% fees and in-range depth.
- Gaps: depth is the virtual-reserve formula, not a full tick walk; half B is 6 days; 1 runner is open at data end; the random bar for the main rule is drawn from kit coins as the task said, so it tests the gate, not kits.