← nuc.grChess · auto-written weekly
The correction 2026-05-23 → 2026-05-30 chess.com/nucos

The wall pushed back.

After the opening stretch's +121, the bill came due. A month in the four-figure foothills: 90–103 over 200 games, a peak of 1062 clawed back to 963 by the same 1000s wall I flagged last time. The Sicilian that carried me cratered to 31%; the Caro-Kann took over as my main Black answer and the London (54%) became the one line I could trust. But watch the last week — 30–27, +31, the first up-week since the slide started. The wall's still standing. So am I.

Record
30–27–2
59 games
Rating
963
+31 · peaked 995
Best streak
4
in a row
Accuracy
86.2
mean
Win rate
51%
this stretch
30 won 2 drawn 27 lost
Rating · the arc · hover to read it
start 932 peak 995 now 963

Opened the week at 932, bled to 898, then a late surge to 995 before settling at 963 — net +31. After a month of giving it back, the first week that clawed some of it back.

Openings — what's working
London Systemwhite · 70g · 54.3%
The floor under a rough month. 54% over 70 games while everything else wobbled — when nothing else works, I reach for the London. model game ↗
Caro-Kann Defenseblack · 48g · 45.8%
Last month's experiment is now the main event: 48 games, up to 46%. Not a winner yet, but it's taken the Black starting job off the Sicilian. model game ↗
Sicilian Defenseblack · 16g · 31.2%
Last edition's hero, this edition's cautionary tale — 67% then, 31% now over 16 games. I stopped trusting it, and it stopped delivering. model game ↗
Queen's Pawnwhite · 12g · 41.7%
A White sideline that hasn't clicked — 42% over 12. Filed under 'needs work.' model game ↗
Games worth seeing
Win rate by opponent strength
800–899 · 13g 92%
900–999 · 117g 48%
1000–1099 · 68g 31%
1100–1199 · 2g 50%
Tendencies the coach noticed
Both colours

White still pays the bills at 50%, but Black sprang a leak — 39% over 99 games. The Sicilian's collapse and the Caro-Kann's growing pains both show up on this side.

The 1000 wall

Still the ceiling: 31% against 1000–1099 over 68 games, down from 38%. Every climb into four figures got me dragged back — this bracket is the whole story of the slide.

How I lose

By resignation (49%) and checkmate (37%), almost never by flag (8%). I'm not getting flagged — I'm getting outplayed. The fix is cleaner moves, not a faster mouse.

Time of day

The afternoon edge flipped: mornings and afternoons sit at 49%, evenings sag to 38% over 69 games. Tired evening chess is losing chess.

Season to date
2026-04-30 → 2026-05-30 200 games 90–103–7 peak 1062 streak 5 scalp 1143

A month of correction: after the surge to 1062, the 1000s wall won the rematch and walked me back to 963 — 90–103 across 200 games at 45%. The encouraging part is the last week, where the bleeding finally stopped.

Past editions
The plateau · "Two green weeks. One stubborn wall."The opening stretch · "The Sicilian is carrying me."