An increment is a fixed number of seconds added to a player's clock after every move they make — the "+2" in a 3+2 time control means each move you play earns you two extra seconds. Invented and patented by Bobby Fischer, the increment guarantees that a player who keeps moving never quite runs out of time, which transforms the character of fast chess: games are decided by technique instead of by who can shuffle pieces faster in the final seconds. Understanding how increment works, how it differs from delay, and how it changes endgame play is basic literacy for anyone who competes online or over the board.
What Is an Increment?
Read any tournament announcement or online lobby and you will see time controls written as two numbers: 3+2, 15+10, 90+30. The first number is the base time in minutes; the second is the increment in seconds, credited to your clock each time you complete a move. In a 3+2 game you start with three minutes, and every move you make adds two seconds to whatever you have left.
Two properties define the increment and drive all of its strategy. First, it accumulates: if you reply in under two seconds in a 3+2 game, your clock actually rises. Play twenty fast moves in a row and you can climb from ten seconds back over forty. Second, it is unconditional: you receive the seconds whether you used two hours or two tenths of a second on the move. Those two facts together create a safety floor. A player with one second left in a 3+2 game is not lost on time — as long as each move takes less than two seconds, they can play forever, and every quick move rebuilds their buffer.
The arithmetic also changes how long games last. A 3+2 blitz game that reaches move 60 has really given each player 3 minutes plus 2 minutes of increment — five minutes of thinking time, not three. If you are budgeting an evening of chess, my guide on how long a chess game takes works through the real durations of every common format.
Fischer's Invention
The increment is one of the few pieces of chess technology with a named inventor and a patent number. Bobby Fischer, retired and famously obsessed with the conditions of play, patented a digital clock in 1988 whose central idea was exactly this: add time after every move, so that a player is never forced to blitz out moves in a position that deserves thought. His complaint was aesthetic as much as sporting — he hated that classical games between great players were being decided by undignified scrambles, pieces flying, flags hanging, scoresheets abandoned.
The chess world resisted briefly and then capitulated completely, because Fischer was right. The "Fischer increment" (the accumulating kind described above) is now the default at nearly every level: world championship matches use a 30-second increment, most classical opens use 90+30, and online blitz's most popular control is arguably 3+2. When you hear "Fischer clock" or "bonus time," it is this mechanism. It is a small irony of chess history that a man who played his last serious game in 1992 shapes the clock of every tournament played today.
Increment vs Delay: Cousins, Not Twins
The increment has a sibling that is constantly confused with it: the delay. With a delay — standard in United States scholastic and club play, typically five seconds — your clock simply waits before it starts counting down. In a G/25 d5 game, each move begins with a five-second grace period; your time only falls once those five seconds pass.
The practical difference is banking. A delay can never increase your clock: move in two seconds and the unused three seconds vanish. An increment credits you the full amount regardless, so fast play accumulates time. With ten seconds left and a five-second delay, you can survive indefinitely but you will always be surviving — your clock can never climb above ten. With ten seconds and a five-second increment, twenty crisp moves later you might have a minute and the freedom to actually think.
There is a third variant, the Bronstein delay (named after the great David Bronstein), which refunds after your move the time you used up to the increment amount — mathematically almost identical to a simple delay in effect: you cannot accumulate. If you remember one sentence: increment can grow your clock; delay can only stop the bleeding. When I coach students who play both US weekend events and FIDE-rated tournaments, this is a real preparation point — a player who has internalized 30-second-increment habits must consciously recalibrate for delay, where "playing fast to build time" earns nothing.
How Increment Changes Endgames
Here is where increment stops being a scheduling detail and becomes a chess topic. Without increment, huge classes of endgames are decided not by their theoretical result but by flagging: whoever moves faster wins, and the defender of a drawn position loses on time while the position on the board smiles at them. With increment, the board's verdict is enforceable — if you have the technique.
Consider the fundamental rook endgames. The Philidor position is a textbook draw and the Lucena position a textbook win, but both verdicts assume accurate play. In a 3+0 scramble, neither player is really demonstrating anything — the defender flags or the attacker flags, and theory is a rumor. In 3+2, two seconds per move is genuinely enough to execute the Philidor's third-rank defense forever, or to build the Lucena's bridge with Rf4, Kd7-c7 and Rc4+ — if the method is automatic for you. The increment converts endgame knowledge from decoration into currency. This is, frankly, why I insist my students drill the basic endings until they can play them at increment speed: in the increment era, the player who knows the drawing mechanism draws, full stop.
The same applies to defensive resources. Holding a fortress, repeating moves toward a threefold repetition, or shepherding a 50-move-rule count to its conclusion all require playing many accurate-enough moves quickly — trivial with an increment, heartbreaking without one. And on the winning side, checkmating with king and queen or king and rook against a bare king on two seconds a move is a solved problem for anyone who has practiced it and a genuine hazard for anyone who has not. I have watched players with a full extra rook agree to draws in blitz because they did not trust themselves to mate on the increment. That is not a time problem; it is a training problem.
Reading Common Time Controls
A quick tour of what the increment does across the popular formats, because its effect scales inversely with the base time:
1+0 versus 2+1. Pure bullet at 1+0 is a typing contest in the final phase. Add a mere one second — 2+1 — and endgames become almost playable; flagging still decides games, but a composed player can convert simple wins. The one-second increment is the smallest dose of civilization available.
3+0 versus 3+2. The great blitz divide. At 3+0, the clock is a weapon of first resort and dead-drawn positions are played out for tricks and flags. At 3+2, the scramble has a floor: two seconds is enough for a trained hand to make a sane move, so wins must generally be earned on the board. Players tend to have a strong preference; ratings between the two pools are not quite interchangeable.
15+10. The standard rapid control of online play and many hybrid events. The ten-second increment means you can always afford a real think at the critical moment and still never lose a trivially drawn ending on time.
90+30 and classical. With thirty seconds guaranteed per move, "losing on time in a winning position" nearly disappears from classical chess, and with it the worst injustices of the pre-Fischer era. What remains is time trouble of a subtler kind: thirty seconds keeps you alive, but it does not make deep decisions for you, and playing a sharp middlegame on increment alone is still miserable.
Armageddon. The tiebreak format has its own twist: many armageddon implementations only introduce an increment after a certain move number (say, move 61), precisely to allow flagging drama for most of the game while preventing infinite grinding at the end. The increment's presence or absence is a deliberate dial for how much chaos the organizer wants.
Strategy: Playing With the Increment
The increment rewards a specific set of habits, and players who adopt them gain fractions of a point everywhere.
Bank in calm positions, spend at critical moments. When the position is quiet and your move is obvious, play within the increment and let your clock grow. You are not "playing fast" — you are saving for the decision that will actually matter. The skill is knowing which is which: the recapture is free, the pawn structure decision is not.
Never resign yourself to perpetual seconds. With an increment, being down to ten seconds is uncomfortable but not terminal. Simplify into positions you can play on autopilot — technical endings you have drilled, structures you know — and let twenty quick moves refill your tank. The worst response to low time on an increment clock is panic-spending your two seconds on moves that deserved none.
Steer the endgame toward your knowledge. Since increment endings are decided by technique, choose simplifications where your technique is the relevant one. Trading into a rook ending you know cold is a clock strategy as much as a chess strategy.
Against a time-pressured opponent, deny the autopilot. The flip side: when your opponent is living on the increment, avoid forcing sequences they can rattle off, and instead pose one new problem per move — quiet moves, small threats, choices between plans. Two seconds is enough to execute a known method; it is not enough to solve a fresh problem. This is the increment-era version of classical time-pressure torture, and it is fully legitimate.
Strategy: When There Is No Increment
In 3+0 and 1+0 pools, everything above inverts. The clock is a piece; material and even mate threats can be secondary to the flag. Keep pieces on the board when ahead on time — every extra piece is another decision for the opponent. Avoid endings that require you to demonstrate precise technique with no safety net. Learn to move with a stable hand and an efficient mouse, because execution speed is literally rating in these pools. And accept the culture: winning on time in a lost position is not sharp practice in a no-increment game; it is the explicit rule set both players chose. If that offends your chess conscience — it sometimes offends mine — the solution is simple and one click away: play the increment version.
Training for the Increment Era at ChessMind AI
Everything the increment rewards is trainable. The endgame positions that increment play converts into must-know currency — Lucena, Philidor, key pawn endings — are drilled to automation in our endgame trainer, which makes you prove the technique against precise resistance rather than nod along to a diagram. The basic checkmates trainer takes the king-and-rook and king-and-queen mates to increment speed, so a two-second clock is never again a reason to draw a won game. Sharpen the fast-and-sane decision habit with our tactics puzzles, and if you want to know which phase of your game actually bleeds points under time pressure, our assessment will tell you where the training time pays best.
Frequently Asked Questions
What does 3+2 mean in chess?
Three minutes of starting time per player, plus two seconds added to your clock after each move you make. The first number is always minutes of base time, the second is the increment in seconds. A 15+10 game is fifteen minutes plus ten seconds per move, and so on.
What is the difference between increment and delay?
An increment adds seconds to your clock after every move, and unused seconds accumulate — your clock can actually grow. A delay merely waits (typically five seconds) before your clock starts counting down, so you can never gain time, only avoid losing it. Fischer increment is standard in FIDE and online play; simple delay is common in US tournaments.
Can your clock go up during a game with increment?
Yes. If you consistently move faster than the increment, your clock rises — this is the defining feature of the Fischer system. Players low on time deliberately play a stretch of quick, safe moves to rebuild a buffer before the next difficult decision.
Why do classical tournaments use a 30-second increment?
Thirty seconds guarantees that no game is decided purely by the flag: a player can always record the moves, execute known technique, and finish a won position. It preserves the sporting result Fischer cared about — the board decides, not the scramble — while still punishing players who mismanage their main thinking time.
Does increment make flagging impossible?
No, it makes flagging fair. You still lose on time if a move takes longer than your remaining clock plus nothing — the increment is credited after you move, not before. What disappears is losing a trivially held position because physically shuffling pieces faster than the opponent was the only thing that mattered.