Blitz or Long Games — Which Actually Makes You Better at Chess?

It is one of the oldest arguments in online chess, and it is usually framed wrongly: it is not that blitz is good or bad, it is that it trains something else. Knowing what that something is settles the question.

What each one trains

Blitz (3 to 10 minutes) trains recognition speed. With little time you do not calculate: you recognise. You see a structure, you know what to do, you move. That is a real and valuable skill, and it is built by playing a lot.

What it does not train is deep calculation, evaluating a difficult position, or endgame technique — because none of the three fits on the clock.

Long games (30 minutes or more) train the opposite: calculating to the end, evaluating without hurry, holding a plan across twenty moves and converting an advantage without giving it back. That is where the skills you later use in blitz get built.

Put differently: long games build and blitz rehearses. A player who only plays blitz gets better and better at rehearsing a set of skills that never grows.

The ratio that works

For someone whose goal is to climb:

• One long game a week, with analysis. Non-negotiable.
• As much blitz as you like, as long as it does not replace the long one.

There is no magic number of blitz games. There is one condition: the long game has to get played. If the week disappears into blitz and the long game never shows up, the ratio is wrong by definition.

Why the long game is the one that teaches

For a simple reason: it is the only one you can genuinely analyse.

Analysing a game means reconstructing what you thought at each moment and comparing it with what the position was asking for. In a three-minute game you barely thought at all — you moved by recognition — so there is nothing to reconstruct. The analysis collapses into "here the engine prefers the other move", which is information without context.

In a thirty-minute game there were decisions. You chose between two plans, you rejected a move for a reason, you calculated a line and missed a reply. That is material you can learn from.

The real damage blitz does

It is not what people usually claim. Blitz does not "ruin your style". What it does do, when it is the entire diet:

It installs the habit of moving without checking. With three minutes there is no time to look at your opponent's checks, so your brain learns to skip that step. That habit then travels into the long game, where there was time.

It rewards traps over good moves. In blitz, a bad trap that wins on time is profitable; in a long game it is a mistake.

It blurs evaluation. If you never sit down to decide whether a position is better or worse and why, that skill does not develop on its own.

None of the three is blitz's fault: they are the fault of only blitz.

When blitz is the right tool

There are three moments where it is the best thing available:

To try a new opening. Ten quick games show you which positions actually come up, far faster than reading theory.

To practise a pattern you just studied. If you worked on the isolated pawn this week, playing blitz while steering towards that structure makes it familiar.

To keep sharp when there is no time for anything else. Half an hour of blitz beats nothing.

And the middle controls

Fifteen-minute games are a rather good middle ground and badly underused. There is time to check and to calculate a short line, and the game is over in half an hour. For someone who cannot sit down for an hour, fifteen minutes with increment is the best available control.

An example week

The ratio is easy to state and hard to sustain, so it helps to see it laid out in days. This is a realistic week for somebody who can give chess about six hours:

• Monday: twenty minutes of tactics. Nothing else.
• Tuesday: one long game — thirty minutes or more — and its analysis. Two hours in total, and they are the two most profitable hours of the week.
• Wednesday: twenty minutes of tactics and half an hour of material: a course, a book, whatever you are working through.
• Thursday: free, or guilt-free blitz. Half an hour, with a finishing time set in advance.
• Friday: twenty minutes of tactics.
• Saturday: a second long game and its analysis.
• Sunday: an hour of endgames or of reviewing what you analysed during the week.

That is two serious games, four short tactics sessions and an hour and a half of study. Almost nobody sustains more than that, and almost nobody needs more than that.

The thing to notice about that week: blitz appears once and with a time on it. It is not banned, it is bounded. The difference between those two is what separates a plan that gets followed from one abandoned in three weeks.

What to do with the clock inside the game

Playing long games is useless if the time is spent badly, and the commonest error is not thinking too much but thinking at the wrong moment. Four rules that fix most of it:

The first ten moves are cheap. If you know the position, move. The time you spend there does not come back, and the opening is almost never where the game is decided at this level.

The expensive moment is when the character of the position changes. The first significant exchange, the first structural decision, the moment a file opens. There it is worth stopping for three or four minutes even when it "looks clear".

Under five minutes, you play differently and deliberately. No long lines: safe moves, nothing weakening, nothing complicated. A lot of won games are lost trying to calculate a six-move variation with two minutes on the clock.

After a surprising move from your opponent, always stop. It is the moment most games collapse: they surprise you, you answer quickly so as not to lose time, and that answer is the mistake.

The genuinely expensive habit: blitz after a loss

There is one pattern that does more damage than all blitz put together, and it is worth naming separately: losing a game and immediately starting another.

What happens there is concrete. You play angry, you play fast, you play worse, you lose again, and you repeat. Three or four games later you are playing well below your level, and those games — the worst you play all week — are the ones that stick, because they are the ones you played with the most emotional load.

It also poisons the diagnosis: if you review the week and half the losses come from tilt streaks, you will not be able to see what you are actually missing.

The rule is simple and it works: after a loss, no immediate rematch. Get up, look at which move the lost game turned on, and only then decide whether you want to play more. Most of the time the answer will be no, and that is information too.

How to come back to long games after years away

For a lot of people the problem is not that they do not want to play long: it is that an hour-long game has become uncomfortable. After years of three minutes, sitting and thinking for five minutes over one move feels wrong, and the discomfort gets resolved by moving fast, which is exactly what ruins the exercise.

Three things that help the transition:

Start at fifteen minutes, not ninety. A jump from 3 to 90 does not hold. From 3 to 15 does, and it already changes almost everything: fifteen minutes is enough to calculate a full variation.

Write the moves down, even playing online. It sounds odd and it works: writing forces a minimal pause before each move, and that pause is what gives you back the habit of looking before moving.

Give yourself one concrete task for the game. "This game I play checking my opponent's checks before every move." One task, not five. It gives you something to do with the time, which is what was missing.

After five or six fifteen-minute games, the thirty-minute one stops feeling long. And the thirty-minute one is where chess starts genuinely teaching.

In short

If you had to keep one rule:

Play as much blitz as you like, as long as this week you played and analysed one long game.

Whoever follows that rule improves even while playing plenty of blitz. Whoever does not, does not improve even if they never play any.

Related questions

Is blitz making me a worse player? On its own, no. What does harm is blitz instead of long games, because it trains you to decide by intuition without verifying, and that habit carries over. With two long games a week as a base, blitz is harmless and even useful for your repertoire.

What is the minimum time control for a game to teach anything? Fifteen minutes with increment. Below that there is no time to calculate a complete line, and without complete calculation the game does not exercise what you are studying.

And bullet? It is a different game, entertaining, and it trains almost nothing transferable to chess at a normal time control. There is no need to ban it; there is a need not to count it as practice.

How many long games a week? Two are enough if you analyse them. Four unanalysed are worth less than two analysed, which is the most consistent relationship in all of chess training.