2026-09-24 · 5 min read
Session edge, AI coach verdict, and running P&L for 2026-09-24.
Trim Review
How much of today's result was carried by a few outliers? The table recomputes the core session stats with the biggest movers removed. Trimmed rows are colored vs baseline: green = holds most of the move, amber = meaningful erosion, red = collapses or flips sign.
| Scenario | Trades | Net P&L | Total R | E[R] / trade | $ WR | PF |
|---|---|---|---|---|---|---|
| Baseline (all trades) | 48 | −$2663.22 | -2.75R | -0.057R | 42.6% | 0.64 |
| − single best + single worst | 46 | −$2372.81 | -2.45R | -0.053R | 42.2% | 0.62 |
| − top 3 + bottom 3 (≈5% each tail) | 42 | −$2162.93 | -2.23R | -0.053R | 41.5% | 0.56 |
Trimmed trades: Single best + single worst — removed APUS +$789.59 (best) · NCPL −$1080.00 (worst). Top 3 + bottom 3 — removed 3 from each tail (by $ P&L).
Broad-based red session: the loss is still there after trimming the biggest movers — this wasn't one bad trade, it was the whole book tilted the wrong way.
Session verdict
{"dayVerdict":"Today's gross P&L was -$2.66k across 48 positions, and the pattern is uneven: real capture in a couple of names, real losses in a couple of the day's biggest movers. PMAX offered +181.4% from its $1.020 open to a $2.870 high at 10:40am across 7 halts, and closed at $1.650, holding 34% of that move; across 10 entries (6 before the high, 4 after) you netted +$668, with your best single entry at 10:30am at +$661.
HCWB offered +84.6% to a $3.120 high at 10:29am, all 9 of your entries came after the high, and you still turned +$625 out of a stock that closed holding only 21% of its move — a stock that gave most of its move back, traded for a profit. Those two names, both multi-halt, were your best work of the day.
The damage came from APUS and GCTK. APUS offered +25% from $6.990 to an $8.740 high at 9:32am, then gave the entire move back and more, closing at $4.920, -29.6% versus the open; across 11 entries (10 before the high, 1 after) you lost -$841, with the single largest loss of the day at -$685 on a 9:30am entry and the largest win of the day, +$790, at 9:31am — a name that whipsawed a full 11-position sequence to a net loss despite one large win inside it.
GCTK offered +107.8% to a $5.050 high at 4:24am, but by the time your 4 entries came (9:39-9:43am, all after the high), it had faded back to a close only 6% above the open, and you lost -$1.27k there, your worst stock of the day. NCPL cost another -$947 on 4 entries that all came after its 9:08am high, in a stock that gave back 39% more than its entire open-to-high move by the close.
GLND was the one name where the move held: +75.4% to a 3:48pm high, still 88% retained at the $5.350 close, and all 5 of your entries came before the high — yet you still finished -$186 there, the one case today where a fully-held mover didn't turn into profit for you. PARA, +77.8% to a 10:29am high across 2 halts, gave the whole move back and closed net negative versus its open; both your entries came after the high and cost -$690.
On the small side, INLF (+3.4%) and CHSN (+41.1% but only 15% retained) were minor, near-breakeven positions.
Of the 13 runners tracked today, you traded 10 — a high participation rate — and only 3 went untouched, of which just 1 moved 30%+ (max +33.7%, not held). So the opportunity cost sitting outside your trading was minimal; today's -$2.66k is almost entirely a story of how you traded the moves you were already in, not moves you missed.
Across the trailing week, the split between holders and bleeders lines up with float and halt count more than anything else: the two multi-day names you touched, GLND and NCPL, show it directly. GLND on 9/21 offered +23.4% and gave nearly all of it back (-19% retained) on a 31.6M float with zero halts, then today, same float, same zero halts, retained 88% into a 3:48pm high — inconsistent even within one low-halt, higher-float name.
NCPL, a 4.4-7.9M float stock, retained just 12% on 9/18, -67% on 9/21, and -39% today — three straight sessions of giving back more than it kept, all with zero or one halt. Meanwhile PMAX and HCWB, both today's better-retained trades relative to what they gave back, carried 7 and 2 halts respectively.
Your picks today skew toward the higher-halt names that this data associates with better retention, but NCPL and APUS — both zero-halt, both faded hard — are the exceptions where you traded into the fader profile rather than the holder one.
Your own halt-based lanes back this up directly: the 21 positions in names with 2+ halts before entry made +$618, against -$3.27k on the 26 positions in names with fewer than 2 halts — PARA (2 halts but scalped under 30 seconds, a DONT lane) lost $691, while PMAX's 10:30am and 11:07am entries, both 4+ halts and one held 30+ seconds (A+ lane, +$1.17k across 5 positions like it), made $662 and $508. Your validated A/A+ halt lanes are earning money in real time; the fewer-than-2-halt, no-lane trades in APUS, GCTK, and NCPL are where the day's losses concentrated."}