Stuck at 1200? A 90-Day Plan Before You Buy a Course

Disclosure. No affiliate links are active on this page. How any future ones would be marked.

If your rating has sat near 1200 for months, find out what is going wrong in your games before you pay for anything. Then follow a free, 90-day study plan with a fixed point at which you decide whether a course is worth buying. The plan is a study structure only: it does not promise any rating change, and nobody at this site has run it.

This page is based on published documentation, pricing and reports, such as sellers' pages, platform help pages and research abstracts, and on coaching advice that is conventional rather than measured. The "Original test pending" section at the end says what a real trial would involve.

Ratings here follow the site's default, the Chess.com rapid scale, which is approximate. See the rating bands overview.

Step one: diagnose what is going wrong

Coaches commonly name a handful of causes for a plateau, and the right fix depends on which one is yours. Spend an evening on your last 20 games before touching a course. You need no paid tool: Chess.com's free plan includes one full game review per day, and Lichess analysis is free.

Is it noise?

A rating is an estimate with uncertainty. Chess.com says its system is Glicko-based and uses a rating deviation, and it keeps a separate rating for each time class (Chess.com support). A flat line over a handful of games can be ordinary variation, so judge by your games, not the number. If you have played fewer than a few dozen games at your current rating, keep playing before concluding you are stuck.

A small calculation shows how much luck is involved. If you are exactly as strong as your opponents and there are no draws, 20 games average 10 wins, with a standard deviation of about 2.2 wins (the square root of 20 x 0.25). So 8 wins or 12 wins out of 20 can happen with no change in your strength at all. Also check whether you are looking at the right number: a blitz rating says little about your rapid play, because the two are tracked separately.

Is "stuck" even the right word?

Two named writers offer other ways to read a flat or falling rating, and both sell training. Dalton Perrine argues in Chess Chatter (6 February 2026) that ratings are relative, so a flat rating can mean you are improving at the same pace as the pool. He adds an important limit: grinding games with no review may be real stagnation. He offers no data, and his newsletter has a paid tier. GM Noel Studer, on overcoming plateaus, says you must accept getting worse in the short term, perhaps 50 to 100 points, and commit to a change for one to three months before judging it. If you start this plan and your rating dips in the middle, that is one possible explanation, though neither writer can tell you it is the reason in your case.

ChessFox, which sells an academy, gives three different causes: knowledge without practical skill, a chaotic thinking process, and weak positional understanding. They are worth holding against your tallies below, as the tags show symptoms and those are possible causes.

Tally your losses

For each loss, note the move where the game turned and tag it with one cause:

The largest tag is your priority. If it is "hung piece" or "walked into a tactic", the usual coaching view is that the fix is a habit and a few weeks of easy puzzles, and that a course is unlikely to add much. If the tags are scattered, you may need breadth, which is where a structured curriculum can help. Coaches commonly advise this kind of tallying; we found no study showing it works, so treat it as conventional wisdom.

Tag the first mistake that changed the result, and give each game one tag only. If a loss has two candidates, ask which error would have saved the game had it not happened. Twenty games at around 10 minutes each is about three hours, so spread it over two or three evenings if one is not enough.

Check how you study

One research finding is worth knowing, with its limits. In a survey of tournament-rated players, serious solo study was the strongest predictor of rating among the activities measured, and a combination of the activities measured explained about 40% of the variance in skill (Charness et al., 2005). The study is a self-reported correlation and cannot show that more study causes higher ratings. Separately, a well-known memory experiment found that testing yourself gave better retention a week later than re-reading (Roediger and Karpicke, 2006), although it used prose passages and students, not chess. Taken together, they suggest that solving problems and reviewing your own games is a better use of time than only watching videos, but that is an inference, not a chess finding.

The 90-day plan

This is 13 weeks of study in four phases. Hours are not specified because your time is yours; the plan is built so each phase can be done in short sessions. For weekly time budgets by available hours, see the study plan for 1000 to 1500, which is a different page with a different job.

Weeks 1 to 2: Safety. Before each move, ask what the last move changed, what is attacked and what is undefended. Play longer games, preferably with an increment, and use easy puzzles on Lichess to build the checking habit. After each game, find your first mistake. Your only goal for these two weeks is completing the check on every move, even when it costs you on the clock. ChessMood's 1,000-1,500 roadmap names 10+5, 15+10 and 30+20 among its recommended controls and says to decide how many games you will play before you start.

Weeks 3 to 6: Patterns. Solve short tactics puzzles daily, and repeat missed ones a few days later. This is simple spaced practice. Add the basic checkmates from Lichess Practice. Keep a short list of missed puzzles, because we found no automatic repetition schedule on the Lichess pages we read. Choose your motif from your tally rather than taking a random mix every day: Lichess's Puzzle Themes has separate sets for forks, pins, hanging pieces and mate in 2.

Weeks 7 to 10: The biggest tag. Take your largest loss category from the diagnosis and work on it directly:

If you change your openings or your thinking routine, Studer's advice above applies: judge it after a month or more, and expect it to feel worse at first. Lichess's FAQ says Zen mode hides ratings if the number is distracting you.

Weeks 11 to 13: Re-diagnose. Tally 20 new games and compare with the first set. Count how often each tag appears. Do not use your rating to judge the plan; use the tag counts. Small changes are noise on 20 games. As a made-up example, "hung piece" falling from 8 to 7 means nothing, while 8 to 3 is worth taking seriously. Treat that as a rule of thumb, not a statistical test, and tag the new games by the same rules as the old ones.

The decision point

At the end of week 13, decide whether to pay, using these questions:

1. Did the biggest tag shrink? If not, ask whether you did the work or whether the material was not enough.
2. Is there a topic you cannot find free material for?
3. Do you need an order to follow, or accountability?
4. Does a course you are considering show a free preview or sample you can judge?

If the answer to one or more is yes, a paid course is a reasonable next step. If the biggest tag did shrink and you are still enjoying the free material, you do not need to pay yet.

If you decide to pay

Choose by the tag, not by a seller's label. Examples published by sellers:

• Blunders: ChessMood's BlunderProof is listed at 4 h 35 min with a stated range of 800-2700; $300 as a standalone purchase as of 8 November 2026, or included in a ChessMood plan. Only its introduction is free to preview. Confirm current plans on ChessMood's own pages.
• Tactics volume: Chessable's 1001 Chess Exercises For Club Players lists at $24.99 as of 8 November 2026, about 1,006 puzzles (approximate). At 20 puzzles a day that is about 50 days, so it would fill the whole pattern phase of this plan and then some.

Start with the cheaper fix. A $24.99 set and a $300 course are not alternatives for the same problem: the first gives you volume, and the second, by its section list, is largely about the habits and psychology behind blunders. If ChessMood's yearly plan tempts you instead, its refund policy gives 30 days from the first payment on a first subscription only, which is a much shorter window than this plan.

For the wider picture at this band, see 1000 to 1400 and how to choose a chess course.

Who should skip paying

• You have not completed the diagnosis. It costs nothing and changes what you buy.
• You are choosing based on a seller's claim of a rating gain. Treat such claims as marketing.
• Your tallies point at habits (hung pieces, time) rather than missing knowledge.

The free resources that cover the same ground

Everything in the plan can be done free: Lichess for puzzles and Practice, free game review on Chess.com or Lichess, and your own games. The free versus paid page explains where free runs out.

What the seller's pages do not tell you

• BlunderProof's preview is thin. The page shows only its introduction (9 episodes, 1 puzzle) unlocked, so you cannot judge most of the course before paying.
• ChessMood's range is wide. "800-2700" does not name a rating system and says little about fit at 1200.
• Puzzle counts are approximate. Chessable's 1,006 figure for the club-players course was derived from section counts on the page.