Stuck at 1200? Four kinds of plateau, and how to tell which one you have
By Nikolas Burk11 min read
A chess plateau usually has one dominant cause: tactics you see too late, openings you memorised but don't understand, endgames you avoid, or a time control that hides all three. Diagnose which one you have from your last ten losses before changing how you study. The fix for each is different; generic advice fails because it treats them as one.
My own plateau
I sat around 1600 on Lichess for a long time. Not a dramatic collapse, just a number that stopped moving while I kept playing. I was doing what everyone says to do: puzzles, an opening I knew by heart, a lot of games.
In hindsight the routine was narrower than it sounds: mostly blitz, because blitz fits into a lunch break; puzzles solved on autopilot, with the little "white to move, there is a tactic" label doing half the work; an opening that was memory followed by a shrug; and no endgame I could avoid by trading into something else and hoping.
None of that is unusual. It is roughly what a 1200 does, and a 1400, and a 1600. The number differs; the pattern is the same. The thing that eventually moved mine was not more of any of it. It was finding out which part was losing the games.
Why doesn't "do more tactics" fix a plateau?
Ask a forum what to do about a plateau and you get the same three answers: do tactics, play longer games, analyse your losses. All three are good advice. All three are also a coin flip, because they are prescribed before anyone has looked at your games.
If your problem is that you drop pieces to two-move forks, more puzzles will help. If your problem is that you reach move 12 with a book position and no plan, puzzles will not touch it. You will solve a thousand of them, keep losing the same way, and conclude that you have hit your ceiling. You haven't. You have been treating the wrong problem.
A plateau is a diagnosis problem first and a study problem second. So here are the four diagnoses I see most, with the symptom you would notice in your own games, a self-check, and what to do about it this week.
Which of the four plateaus do you have?
1. The tactics leak
Symptom: you lose material to short tactics you would solve in seconds as a puzzle. A fork, a hanging piece, a back-rank threat. You see it one move after it happens.
Self-check: open your last ten losses and find the move where the evaluation fell off a cliff. If in five or more of them the losing move allowed a tactic of two moves or fewer, this is your plateau.
This is the position from our fork lesson, and it is the whole plateau in one picture. Nothing is hidden. The king and the rook are both sitting on squares a knight can hit from d4. In a puzzle, with the label "white to move and win," most 1200s find ♘︎d4+ in under ten seconds. In a game, with the clock running and no label, the same player plays something sensible like ♖︎a1 and the chance is gone.
The leak is not pattern knowledge. It is the habit of looking. Puzzles train you to find a tactic when told there is one; games require you to check for one on every move, including the boring ones.
2. Memorised openings
Symptom: your games go fine until the moves you know run out, then you drift. The position is objectively equal and you have no idea what to do with it. Three moves later it is objectively worse.
Self-check: in your last ten losses, find the first move where you were "out of book." If the evaluation starts sliding within five moves of that point in most of them, this is your plateau.
Every move Black has played here is theory. The question is whether the player knows why. The pawn on c5 attacks d4, the base of the e5–d4 chain, because a chain is attacked at the base, not the tip. The bishop went to f5 before ...e6 so that it would not be locked behind its own pawns like the French bishop is. Those two facts are the plan for the next twenty moves: pressure d4, keep the bishop active, decide where the king goes.
A player who memorised the moves gets this position and sees a bunch of pieces. A player who understood them sees a pawn chain with a base and a bishop with a job. The first one plays ...♘︎c6, ...♗︎e7 and ...O-O because that is what development looks like, then wonders why White's kingside space turns into an attack. We walk through the ideas, not just the moves, in the Caro-Kann Advance guide, and the general method is the subject of building an opening repertoire you actually understand.
3. Endgame avoidance
Symptom: you trade into an equal endgame to "simplify" and then lose it, or you decline every trade and get outplayed in a middlegame you were ahead in, because you are afraid of what comes after.
Self-check: count how many of your last ten losses went past move 40 with fewer than three pieces per side. If it is three or more, and you lost or drew positions the engine calls level or better, this is your plateau. A second tell: you have never deliberately studied king-and-pawn endings and could not say, on sight, whether a given one is won.
The endgame plateau is the one most players at 1200 do not know they have, because the games that expose it are long and nobody analyses long games. It is also the cheapest to fix. There are about six positions that cover most of what you will actually reach; we laid them out in six endgame positions worth studying. If you only learn one thing, learn the opposition, because it decides the king-and-pawn endings that every other ending can simplify into.
4. The wrong time control
Symptom: you play almost entirely 3+0 or 5+0. Your losses do not have one bad move; they have four or five, each a little worse, with no time to think about any of them.
Self-check: this one you can answer without opening a game. What share of your rated games in the last month were under ten minutes per side? If the answer is "most of them," this is your plateau, and it is the one that hides the other three. You cannot tell whether you have a tactics leak or an opening problem when every game is a flurry of premoves.
Blitz is not bad for you. It is a bad diagnostic. Everything above depends on being able to look at a loss and say "this is where it went wrong," and in a blitz game there is no single place, just a slow slide. At least some of your games have to be slow enough to carry a lesson.
The five-minute diagnosis
Open your last ten rated losses in your game review tool of choice. For each one, find the move where the game turned (the engine graph makes this obvious) and write one letter in a table: T for a short tactic you missed, O for a position you had no plan in shortly after leaving book, E for an endgame you should have held or won, B for a blitz game with no single turning point.
| Game | Turning move | T / O / E / B |
|---|---|---|
| 1 | ||
| 2 | ||
| … | ||
| 10 |
Some games get two letters. Give both. Then count. The letter that appears most is your plateau; if B appears more than three times, start there, because it is masking the others.
If the count is flat, with two or three of everything, you do not have a plateau yet. You have a sample-size problem. Play ten more slow games and count again.
What's the best way to improve at chess as an adult?
Thirty minutes a day, aimed at the letter you counted. Not an hour spread across everything. Here is one routine per plateau.
T, the tactics leak: fifteen minutes of puzzles, but with a rule: before entering a move, name the threat you are stopping or creating, out loud or in your head. Then fifteen minutes on one rated game at 15+10, where before every move you ask the same question you asked in the puzzles. The goal is to move the habit from the puzzle screen to the board, which is the part that puzzles alone do not do.
O, memorised openings: pick the one opening you play most and spend thirty minutes on one line of it, not learning new moves but asking, for each move you already know, what it does. Which pawn does it attack, which square does it control, which piece does it free. Write one sentence per move. When you reach the end of what you know, write down the plan in two sentences. Next day, play one slow game in that opening and see if the plan survives contact.
E, endgame avoidance: one endgame position per session, played out against an engine until you can win or hold it from both sides. Start with the opposition, then the Lucena and Philidor positions. The point is not to memorise the solution but to be able to say, on sight, "this is won" or "this is a draw," which turns the trade-or-don't decision from fear into a calculation.
B, the wrong time control: replace half your blitz with 15+10 for two weeks. Do not change anything else. Then redo the diagnosis table. You will find out which of the other three you actually have, and most of the thirty minutes will be the game itself.
None of this promises a rating. What it promises is that in two weeks you will know what is costing you games, which is the thing you do not know now.
Stuck at 1200: what app or website should I use to improve?
Probably not a new one, yet. The honest answer to "which app" depends on which letter you counted, and three of the four most common reactions to a plateau make it worse.
A new opening. The plateau is almost never the opening you chose. If you counted O, it is the way you learned it, and the next one will be learned the same way. Switching from the Italian to the London buys you two weeks of novelty and the same drift at move 12.
A new app. If you counted T, you already have a puzzle app, and the problem is the transfer from puzzles to games, which no puzzle app fixes on its own. Our list of the best chess apps in 2026 is sorted by the job you need done, and for most plateaus the job is "look at my games," not "give me more puzzles." Disclosure: we make one of the apps on that list. It is not ranked first overall.
More blitz. This is the most common reaction and the most reliably wrong. A plateau is a signal that something in your play needs looking at, and blitz is the time control that makes looking impossible.
The right tool is the one that shows you your own losses with the turning point marked, so the five-minute diagnosis takes five minutes. Everything else is secondary.
If you want the diagnosis done for you
The counting exercise above is, more or less, what the LearnChess Coach does from the games you import from Lichess or Chess.com. It reads the moves, results and clocks, and ranks the habits that are costing you points, with the number of games behind each one: an opening you score badly with, a share of losses on the clock, a slump in the game after a loss. The stats are free for every account; the ranked leaks with their example games need a paid tier. It will not find a tactics leak for you, because that needs engine analysis of individual games, which is what game review is for. But it will tell you whether you are in plateau 2 or 4 before you have opened a single game, and it will show you the games that prove it.
Frequently asked questions

Nikolas Burk
Founder of LearnChess
I plateaued around 1600 on Lichess and built LearnChess to understand my own games more deeply: engine lines in plain English, questions about the board in natural language, and a structured path of lessons, practice and tests. On the blog I share what I learn along the way.
More from the blog
- Tools & comparisons
Chess.com vs Lichess vs Chessable for improving: which one, for what
What each site is actually good for if you want to improve, in one table: playing, puzzles, analysis, openings, explanations. Prices checked November 2026.
12 min read
- Improver's playbook
How to stop blundering in chess: a check that survives the clock
Blunders are a looking problem. A three-question check you can run on every move, two positions to practice it, and why blitz keeps it from sticking.
10 min read
- Improver's playbook
How to analyze your own chess games in 20 minutes (without letting the engine do the thinking)
A three-pass routine for analyzing your own chess games: notes first, engine as grader, one sentence per mistake. For 1200–1800 players short on time.
10 min read