Pattern Recognition — 800

Week 0 — Setup and baseline

do this first

Three or four days, before any themed drilling. The point isn't admin: Lichess serves puzzles matched to your puzzle rating, so if you start Week 1 without one, you'll be drilling forks at the wrong difficulty and wasting the repetition. Already have a Lichess account? Some of these are already true — go down the list and tick what's done.

Your detailsStored on this device only

Lichess username Phase 1 start Create account With a username saved, this page can read your ratings itself — the same numbers anyone can see on your public profile. No password, no permission, nothing to grant. If Lichess can't be reached, the boxes still work by hand.

Make an account at lichess.orgFree, nonprofit, nothing is paywalled. One catch: your username is permanent. Lichess only ever lets you change its capitalization. Pick one you'll still be happy with at 1600.
Turn on board coordinatesEvery square has a name: columns left-to-right are lettered a to h, rows bottom-to-top are numbered 1 to 8. The square in column e, row 4 is "e4" — the first move of your repertoire. This setting prints those letters and numbers along the edge of the board so you aren't counting squares with your finger.

Where: Preferences → Game display → Board coordinates → Outside the board.

Why it's on this list: everything in this plan is named by square — "the f7 square", "knight lands on d5". If you can't read squares at a glance you're translating on every single puzzle.
Learn what 15+10 means, and always use itA time control is how much clock each player gets. Run out and you lose, whatever the position. 15+10 means 15 minutes each for the whole game, plus 10 seconds added back every time you move. That second number is the increment — it's why you can't bleed out just from thinking.

Where: the grid of buttons on the Lichess home page — 1+0, 3+0, 5+0, 10+0, 15+10, 30+0. Click 15+10 and it finds you an opponent. There's no setting to save; just always click that one.

Why: the pre-move checklist costs about ten seconds a move. In a 3-minute game you can't afford it, so you'd guess instead — training the exact reflex we're trying to remove.

Vocabulary: Lichess sorts games by length — under 3 min is Bullet, 3–8 is Blitz, 8–25 is Rapid, 25+ is Classical. You get a separate rating for each. 15+10 is Rapid, which is the rating this page asks you to record.
Solve 25–30 puzzles without picking a categoryGo to the puzzles page and solve whatever it hands you. Take as long as you want.

What this is doing: Lichess gives you a puzzle rating — a number for how good you are at tactics — and uses it to decide how hard the next puzzle should be. It takes roughly 25–30 attempts before that number means anything.

Why it comes first: if you skip straight to fork puzzles in Week 1 without a settled rating, you'll get forks at random difficulty — some insultingly easy, some impossible — and the repetition won't build anything. Don't log failures yet. This is calibration, not training.

Already solve puzzles regularly? Then your puzzle rating is already settled — tick this and skip the calibration solves.
Play 4 games at 15+10Real games against real people. Play your five opening moves even when they feel wrong — that's the point of a fixed repertoire.

Afterwards, ask one yes-or-no question: by move 10, had I played e4, d4, both knights out, and castled? That single question is the whole opening audit. Log the yes or no in the game log below. Don't analyze deeply yet.
Write down your two starting numbersYour rapid rating is on your profile page — click your username, top right. Your puzzle rating is on the puzzles page. Put both in the "start" boxes below.

In eight weeks it will feel like nothing has changed. Having the honest before-number is the only defence against that feeling.
Optional but recommended: hide all ratingsPreferences → Display → Show player ratings → No. Every rating on the site disappears. You still play rated games, you just can't see the numbers.

This turns "don't check your rating for three weeks" from advice you'll fail into something that isn't available to fail at. Turn it back on when you want to update the baseline.

Glossary

Time control	How much clock each side gets. "15+10" = 15 minutes each, plus 10 seconds back per move.The second number is the increment.
Rapid	Lichess's name for games lasting roughly 8–25 minutes. 15+10 is Rapid.Bullet is under 3 min, Blitz 3–8, Classical 25+. Separate rating for each.
Coordinates	Square names. Columns a–h left to right, rows 1–8 bottom to top. e4 = column e, row 4.
Puzzle rating	A separate score for tactics only. Decides how hard the puzzles you're served are.
Motif	A recurring tactical pattern with a name — fork, pin, skewer. The vocabulary of the whole plan.
Tempo	One move's worth of time. "Gaining a tempo" = forcing them to respond so you effectively move twice.
Repertoire	The fixed set of openings you always play. Yours is the five moves at the top of this page.
Opening audit	One yes/no per game: by move 10, did I get e4, d4, both knights, and castle?

The scoreboard

baseline · deltas · audit

Start numbers, current numbers, and the two leading indicators that move before the rating does.

Rapid — start Rapid — now Puzzle — start Puzzle — now

Rapid

—

no baseline yet

Puzzle

—

no baseline yet

Opening audit

—

no games logged

Drill set

0

puzzles logged

Don't read the rating for three weeks. It's a lagging, noisy indicator and provisional ratings swing 100+ points on nothing. The audit pass rate and the drill-set growth above will move first.

Phase 1 — Motif isolation

8 weeks · 1 motif each

Eight weeks, one motif each. Ordered by frequency in the positions your repertoire produces, cross-checked against which themes actually peak in the 800–1200 band. Every chip is a live drill.

Weeks 1–80 of 8 complete

1	Hanging pieces Universal, and still the number one miss in this band. Everything else is downstream of noticing what's undefended. hangingPiece Week complete
2	Forks Peaks precisely in your band. Knights on f3 and c3 feed the d5, e5 and c7 outposts your center creates. fork Week complete
3	The f7 / f2 square The defining tactic of open e4 games. Your Bc4 and Nf3–g5 point straight at it — and so do theirs. attackingF2F7 Week complete
4	Pins and skewers Bb5 against Nc6, Bg5 against Nf6, and e-file pins once the center opens. pin skewer Week complete
5	Back rank and forced mate You castle every game. So do they. Start at mateIn1 and only move up when it's instant. backRankMate mateIn1 mateIn2 Week complete
6	Discovered attack and double check The e-file opens in nearly every game you play. This is what lives there. discoveredAttack doubleCheck Week complete
7	Removing the defender How you dismantle a defended center — and how yours gets dismantled. Deflection runs slightly ahead of your band, which is the point: it's the next rung. capturingDefender deflection Week complete
8	Kingside attack and the Greek gift Bxh7+ arises naturally from e4 + d4 with a knight on f3. Learn it as attacker and as victim. kingsideAttack sacrifice attraction Week complete

Not yet — these peak above your band

Published theme-frequency data puts x-ray around 1400 and trapped piece around 1600. Drilling them now is wasted repetition; they'll arrive on their own.

xRayAttack trappedPiece interference quietMove zugzwang

Once a week, drill your actual openings

Puzzles by opening All themes

Lichess theme tags are auto-generated and occasionally wrong — a puzzle tagged fork may really be about something else. Note it and move on.

The daily thirty minutes

20 + 10 · checklist

20	Themed puzzles — this week's motif only. Untimed. No hints.Say the motif name out loud, then find where it lives on the board. You're practicing recognition, not calculation.
10	Review one of your own games — find your worst move before the engine goes on.Opening audit first: by move 10, did you get e4, d4, both knights, and castled?

Games: 3–4 per week at 15+10 or 10+5. Not blitz. Blitz at your level trains pattern guessing, the opposite of the objective.

The pre-move checklist

Ten seconds a move. Four weeks and it stops being conscious.

What did their last move actually threaten?
What of mine is undefended right now?
What of theirs is undefended right now?
What checks, captures and threats do I have?

Game log and opening audit

0 logged

One line per game. The audit is the only leading indicator that tells you the repertoire has gone automatic — when the pass rate stops moving because it's pinned at 100%, the opening is no longer costing you anything.

With a username saved, Fetch recent games reads your public rapid games from Lichess and computes the audit from the actual moves — no self-reporting.

Gamesaudit: by move 10 — e4, d4, both knights, castled

No games yet. Log the four from Week 0 to establish where the audit starts.

Phase 2 — Woodpecker your own failures

0 logged

Log every puzzle you miss. By week 8 you'll have 150–250, and that becomes your drill set — better than any book's, because every position is a pattern you specifically don't see. Each ID opens unrated, so replaying never touches your puzzle rating.

Failed puzzlesthe drill set

Connect your puzzle history (optional)

Everything in this box can be done by hand with the paste field below. The token just does the typing for you.

What a token is

A personal access token is a long random string that Lichess generates on request — a spare key cut to open exactly one door. Paste it here and this page can ask Lichess which puzzles you attempted and whether you solved them. Missed puzzles land in your drill set automatically, with their motif tags filled in.

Where to get one

Create the token on Lichess — that link opens the form pre-filled: one permission (Read puzzle activity). Press Create, copy the token, paste it below. If you ever lose it, don't hunt for it — delete it and make a new one.

What it can do

Read your puzzle history. That is the only permission on it.

What it can't do

Play moves, change settings, send or read messages, see your password, or touch your rating. Lichess enforces this on its side — it isn't a promise from this page.

Where it lives

In this browser's storage, next to the rest of your log. It's sent to lichess.org when you press Sync, and nowhere else, ever.

How to undo it

Press Disconnect here to remove it from this browser, and delete it at lichess.org/account/oauth/token to make it stop working everywhere.

Bulk import — paste puzzle IDs or full puzzle links, one per line, straight from your browser's address bar

Nothing logged yet. The ID sits in the top-right of the puzzle screen — four seconds to copy, and it's the input the whole of Phase 2 runs on.

The cycle, once the set is built

Cycle	Target	Roughly
1	Full set, untimed, get them all rightMiserable and slow. Expected.	~2 weeks
2	Full set in half the time of cycle 1	~1 week
3	Half of cycle 2Where it clicks and you start seeing motifs live without hunting.	~3–4 days
4	Half again	~2 days
5	Near-instant recall on every position	~1 day

Why the classical center works

4 mechanisms

In order of how much it's actually doing for you.

1

It denies squares — a passive attack on every piece they own

Pawns on e4 and d4 jointly control c5, d5, e5 and f5: precisely the squares enemy knights and bishops want. A knight that can't use d5 or e5 goes to the rim and does nothing for ten moves. You aren't just occupying the center, you're evicting their army from the good real estate — and it costs you nothing ongoing.

2

Tempo arithmetic beats everything else at this level

Knights reach f3 and c3 in one move each and hit center squares immediately. A typical opponent in your band burns three to five moves on an early queen sortie, re-moving a kicked piece, or a rook pawn. Finish developing on move 8 against their move 14 and you have six moves of free play — more attackers than defenders at the point of contact, which is the literal precondition for every tactic below.

3

Castling is two moves disguised as one

Your king leaves the e-file — exactly the file that opens in every e4 game — and a rook joins the game. Uncastled kings in opened positions are the single largest source of losses in this band.

4

It eliminates decisions where decisions are cheapest

The opening is the phase where the precisely-correct move matters least at your rating and where thinking costs the most clock. Automating it means arriving at move 12 with full attention, most of your time, and a position type you've seen two hundred times.

The one correction. 1.e4 e5 2.d4? runs into 2...exd4, and now you either recapture with the queen — 3.Qxd4 Nc6 develops their knight with tempo — or you gambit the pawn. Fix: knight before pawn. 1.e4 e5 2.Nf3 Nc6 3.d4 arrives supported and with a threat behind it. That's the Scotch.

The repertoire, stated precisely

move orders

Always 1.e4. Against 1…e5 and 1…c5, knight before pawn. Everything else, pawn immediately.

White1.e4 · then d4, Nc3, bishop, castle

They play	You play	Lands you in
1…e5	2.Nf3, then 3.d4	Scotch
1…c5	2.Nf3, then 3.d4	Open Sicilian
1…e6	2.d4 immediately	French
1…c6	2.d4 immediately	Caro-Kann
1…d5	2.exd5, then 3.Nc3Hits the queen — you gain a tempo.	Scandinavian
1…d6 / 1…g6	2.d4, then 3.Nc3	Pirc / Modern
1…Nf6	2.e5 Nd5 3.d4	Alekhine

As Black — mirror the philosophy, not the moves

You're a tempo down and must contest the center rather than claim it. Against 1.e4 play 1…e5, then …Nc6, …Nf6, …Bc5 or …Be7, castle. Against 1.d4 play 1…d5, then …Nf6, …e6, …Be7, castle.

Nc3 caveat. After e4 it's almost always right. In d4 positions without e4, your c-pawn wants c4 and Nc3 blocks it. You're committed to e4, so this won't bite you — but it's why this is an e4 system, not a universal one.

Traps you'll actually meet

5 · all answered

All over the beginner YouTube pipeline, so you'll see them weekly. Every one only works if you get greedy or panic.

FiveYour system already answers them

Trap	Answer
Scholar's MateBc4 and Qh5 or Qf3 aiming at f7	Develop with tempo. …Nf6 hits the queen, …g6 if she's on h5. Count defenders on f7 first.
Wayward Queen2.Qh5 immediately	2…Nc6 3.Bc4 g6 4.Qf3 Nf6 — you develop, they retreat.
Stafford Gambit1.e4 e5 2.Nf3 Nf6 3.Nxe5 Nc6	Decline the tempting grabs. 4.Nxc6 dxc6 5.d3 and just develop.
Blackburne–Shilling2…Nc6 3.Bc4 Nd4	Don't take on e5. 4.Nxd4 exd4 5.O-O and you're simply better.
Legal's MateTheir …Bg4 pins your Nf3	Break the pin early with h3, or don't let queen and king line up on the d1–h5 diagonal.

What to deliberately not do

4 rules

No opening study beyond the repertoire table. You have a repertoire. It's sufficient to 1600. Adding lines is the most enjoyable and least effective thing available to you.
No Puzzle Storm or Puzzle Rush. Speed modes reward guessing plausible-looking moves.
No engine before your own analysis. You learn from the gap between your judgment and the engine's. Open it first and there's no gap.
No daily rating checks. Lagging and noisy. Use the scoreboard instead.

When this expires

~1400–1600

Around 1400–1600. Opponents there start playing the move orders that punish an automatic Nc3, and know theory past move 10. That's the far edge of "short to middle term" — roughly a year of runway. Don't spend it on openings.

Pattern
Recognition

Start with Week 0

Week 0 — Setup and baseline

Glossary

The scoreboard

Phase 1 — Motif isolation

Not yet — these peak above your band

Once a week, drill your actual openings

The daily thirty minutes

The pre-move checklist

Game log and opening audit

Phase 2 — Woodpecker your own failures

The cycle, once the set is built

The thinking behind it

Why the classical center works

It denies squares — a passive attack on every piece they own

Tempo arithmetic beats everything else at this level

Castling is two moves disguised as one

It eliminates decisions where decisions are cheapest

The repertoire, stated precisely

As Black — mirror the philosophy, not the moves

Traps you'll actually meet

What to deliberately not do

When this expires