Blitz chess is chess played on a short clock, usually between 3 and 10 minutes per player, and on Lichess it is the most played speed: 686,648 of the 1,481,695 games in our sample were blitz. The word means different things on FIDE's rules, on Chess.com and on Lichess, so a game called blitz in one place can be rapid in another. We read the first 1,481,695 usable Lichess games of 1 October 2026 to answer the questions people type most often: is blitz harder than rapid, is it good for beginners, and does it make you worse.
The short answer
Blitz is faster, not simply worse. At 2200 and above, players blunder 1.87 times per 100 moves in blitz and 1.42 in rapid, so the clock costs strong players something real. Below 1000, rapid is not cleaner than blitz at 6.15 blunders per 100 moves against 5.71, which suggests the clock is not what holds those players back. For most people the useful question is not blitz or rapid but whether you look at the games afterward.
What counts as blitz chess
Three bodies define it, and as of November 2026 they do not agree at the edges.
- FIDE. Appendix B of the Laws of Chess says a blitz game is one where all the moves must be completed in a fixed time of 10 minutes or less per player. With an increment, the allotted time plus 60 times the increment must be 10 minutes or less.
- Chess.com. Blitz is between three and ten minutes per player. Faster than three minutes is bullet, and longer than ten is rapid. A Chess.com member blog on classification counts total time as the base plus 40 times the bonus.
- Lichess. Each game gets an estimated duration, the starting seconds plus 40 times the increment in seconds. Up to 179 seconds is bullet, up to 479 is blitz, up to 1499 is rapid, and 1500 or more is classical.
The multiplier is the disagreement: FIDE uses 60, the two websites use 40. It matters most for the 10 minute games. A 10+0 game is blitz for FIDE, rapid on Lichess, where 600 seconds is over the 479 limit, and right on the boundary of Chess.com's three to ten minute range. Some third-party pages say blitz is three to five minutes. The official definitions say otherwise, so we ignore that range.
What 3+2 and 5+3 mean in total time
The notation is minutes plus the increment in seconds added after each move. Working both common controls through each rule:
| Control | FIDE count | Lichess count | Chess.com blog count | Label |
|---|---|---|---|---|
| 3+0 | 180 seconds (3 minutes) | 180 seconds | 3 minutes | Blitz |
| 3+2 | 300 seconds (5 minutes) | 260 seconds | about 4.3 minutes | Blitz |
| 5+0 | 300 seconds (5 minutes) | 300 seconds | 5 minutes | Blitz |
| 5+3 | 480 seconds (8 minutes) | 420 seconds | about 7 minutes | Blitz |
| 10+0 | 600 seconds (10 minutes) | 600 seconds | 10 minutes | Blitz on FIDE, rapid on Lichess, boundary on Chess.com |
The arithmetic is ours, from the three formulas above. On Lichess a 3+0 game is on the edge: 180 seconds is the first value that counts as blitz, so a 3+0 game is blitz while 2+1 (2 minutes plus 40 seconds, 160 seconds) is bullet. Chess.com's terms page names 3|0 and 3|2 as the most popular blitz controls, with 5|0 and 5|5 also very popular. We could not source a popularity claim for 5+3.
The rules that change in blitz
FIDE says blitz play follows the Rapid chess Laws in articles A.2, A.3 and A.5, apart from the blitz-specific articles. Two differences matter at the board, and none of them changes how the pieces move.
- Illegal moves. In rapid and blitz, if no arbiter steps in, the opponent may claim an illegal move only before making their own next move. Once they have moved, the illegal move cannot be corrected unless both players agree without the arbiter. We did not retrieve FIDE's blitz penalty for an illegal move, so we do not state one.
- Claiming a win on time. The claimant may pause the clock and call the arbiter. The game is drawn if the claimant could not checkmate the opponent's king by any series of legal moves. The arbiter may also call a flag fall they see themselves.
The time rule has a practical effect. If your opponent's flag falls but you could not checkmate them by any series of legal moves, for example because you have a bare king, the game is drawn, not won. The same rule saves you when your own flag falls against an opponent who has no way to mate.
Is blitz chess harder than rapid? Our data
We streamed the Lichess open database file for October 2026 (CC0 license) and read its first 1,521,984 games, all played on 1 October 2026. After dropping ultrabullet games and games with a rating gap over 400 between the players, 1,481,695 games remained: 535,168 bullet, 686,648 blitz, 251,339 rapid and 8,540 classical. Lichess's own classification decides the speed. Of those, 150,000 carried computer evaluations, and from them we classified 8,350,512 moves.
A blunder here is a move that drops the mover's win probability by 30 points or more, using Lichess's win-probability formula on the engine evaluation from the mover's side. The first 10 plies are skipped as opening theory. Each move is placed in a band by the mover's own rating. The table gives blunders per 100 moves, and every cell has more than 2,000 moves.
| Mover rating | Bullet | Blitz | Rapid | Classical |
|---|---|---|---|---|
| Under 1000 | 5.41 | 5.71 | 6.15 | 4.96 (wide interval) |
| 1000 to 1399 | 4.66 | 4.42 | 4.48 | 4.77 |
| 1400 to 1799 | 3.97 | 3.42 | 3.32 | 3.00 |
| 1800 to 2199 | 3.62 | 2.63 | 2.34 | 1.92 |
| 2200 and above | 2.95 | 1.87 | 1.42 | 0.82 (wide interval) |
Source: Lichess open database, October 2026 file, games of 1 October 2026, CC0. Blunders per 100 moves, played moves only. The 95% interval is under 0.15 for every blitz and rapid cell. Classical is a small sample (4,287 moves at 2200 and above, 2,257 below 1000), with intervals of 0.27 and 0.90, so treat those two cells as rough.
What the table says
Rating is the biggest effect by far. In blitz, a player under 1000 blunders 5.71 times per 100 moves and a player at 2200 and above 1.87, about three times as often. The speed matters less, and it matters differently at each level:
- At 2200 and above, blitz is 1.87 against 1.42 in rapid. That is about 32% more blunders for players in the same band.
- At 1800 to 2199, blitz is 2.63 and rapid 2.34.
- At 1400 to 1799, it is 3.42 against 3.32, a small gap.
- At 1000 to 1399, it is 4.42 against 4.48, which is a tie within the intervals.
- Under 1000, rapid is higher at 6.15 against 5.71.
The blitz-versus-rapid gap widens with rating. We read it this way, and it is an interpretation, not something the data proves. A player under 1000 mostly blunders because they do not see the tactic, a hanging piece or a fork, and extra minutes do not put it on the board. A 2200 player sees most of those tactics given time, so a short clock takes something from them that a long one gives back. Time buys strong players more.
Two limits apply. These are different games and different people in each speed, not the same players switching clocks. And these are moves from games where someone asked for computer analysis, which brings us to the caveats.
How reliable is this data
- It is part of one day. The first 1.5 million games of the October 2026 file, all from Thursday 1 October. Another day, a tournament or a holiday could shift it.
- The evaluated games are self-selected. Someone requested analysis, which skews toward people who want to improve and toward games that went wrong. Rates for all games may differ.
- The thresholds are ours. A blunder is a win-probability drop of 30 points or more. Other cut-offs give other numbers, though the pattern across bands should hold.
- The ratings are Lichess ratings. They are not comparable to Chess.com or FIDE ratings. Our guide to the chess rating system and how ratings differ across sites explains why.
- Engine evals are server-side Stockfish at varying depth, so some marginal calls could change with a deeper search.
As a check, we split the blitz 1400 to 1799 cell by game index into two halves. They gave 3.44 and 3.40 blunders per 100 moves, a difference of 0.04 that falls inside the margin of error. The other internal checks, such as games read equalling games counted plus games skipped, also passed.
How often blitz games end on time
Time is also a result in blitz. We counted the share of all games, evaluated or not, that ended on a time forfeit.
| Average rating | Bullet | Blitz | Rapid |
|---|---|---|---|
| Under 1000 | 65.1% | 25.6% | 13.0% |
| 1000 to 1399 | 57.5% | 22.6% | 11.1% |
| 1400 to 1799 | 54.0% | 22.9% | 11.0% |
| 1800 to 2199 | 51.2% | 25.2% | 11.8% |
| 2200 and above | 44.8% | 23.6% | 10.5% |
About a quarter of blitz games end on time at every band, and the share barely moves with rating. Bullet is different: more than half of games under 2200 end on the clock, and it falls to 44.8% only at the top. Our bullet chess guide looks at how often the player who flags was winning. Rapid sits near one in nine. If you play blitz, one game in four will end by the flag, so time management is part of the skill and not a nuisance.
Blunders by remaining clock
We also recorded how much time the mover had when choosing each move, from the game clock. Blunders per 100 moves in bullet and blitz, by clock left:
| Clock left | Bullet | Blitz |
|---|---|---|
| Under 10 seconds | 5.45 | 4.86 |
| 10 to 30 seconds | 4.49 | 4.04 |
| 30 to 60 seconds | 3.57 | 3.74 |
| 60 seconds or more | 2.96 | 3.22 |
In blitz, a move played with under 10 seconds left is a blunder 4.86 times per 100, against 3.22 with a minute or more. In bullet the gap is wider, 5.45 against 2.96. Part of this is the clock itself, and part is that players short of time are often already worse and defending. We cannot separate the two. The practical advice is the same: when you are below about 10 seconds, choose the safe move, not the clever one.
Is blitz chess good for beginners
The experts we found disagree, and the data above does not settle it.
- For blitz. NM Jeremy Kane of Chess.com argues blitz is the best form of training: it builds pattern recognition, exposes weaknesses and helps build an opening repertoire. His caveat is not to play when you are overly tired.
- Against too much of it. NM Bruce Pandolfini wrote in a Chess.com mailbag that if you play too much speed chess, you are apt to start neglecting strategy in slower chess.
- The forums. One poster calls blitz junk food for beginners and says to stick to rapid and slower games. Another says blitz is bad only if you do not review your games. These are opinions, not evidence.
What our data adds is narrow. A beginner under 1000 blunders 5.71 times per 100 moves in blitz, but also 6.15 in rapid, so slowing down alone does not fix it. We think a beginner is better served by tactical pattern work, which our guide to chess tactics and the patterns behind them covers, than by choosing a different clock. Our advice: play some rapid so you get used to thinking, play blitz for fun and pattern practice, and never skip the review. If you are just learning the rules, start with how to play chess for beginners before you worry about clocks.
Does blitz chess make you worse
Our data cannot answer this. It compares the games of different players at different speeds on one day, not one player over months. What the sources do say:
- Pandolfini's warning is about strategy: heavy speed chess can wear down slow, strategic play.
- Community threads blame habits: lucky tactical wins and bad opening habits.
- Kane says blitz exposes weaknesses, which is a benefit only if you look at the weaknesses afterward.
The shared point is that blitz without review repeats the same errors faster. We could not find a controlled study, so treat any claim that blitz definitely harms or definitely helps with caution. The honest position is that a fast clock rewards what you already know and punishes what you have not practiced, and a review is the step that turns losses into learning. Our guide on how to get better at chess places blitz inside a full training week.
Blitz chess tips that follow from the data
- Check for blunders every move. Before you play, ask what the opponent's last move threatens and whether your piece is hanging. It takes one second.
- Bank time early. Play opening moves you know quickly. Players with 60 seconds or more blunder less than players with under 10, so the time you save in the opening is there to spend in the middlegame. Our look at the best first move in chess shows how little the first move matters and how much knowing your setup does.
- Pick an increment control if flags decide your games. With 3+2 or 5+3 you get 2 or 3 seconds back after every move, so a long endgame does not run your clock down the way 3+0 or 5+0 does.
- Know the time rule before you flag. If your flag falls but your opponent could not mate you by any legal sequence, the game is a draw, not a loss.
- Stop when you are tired. Kane's one caveat is not to play when overly tired.
- Review a few games. You do not need all of them. Pick your losses where you were winning and find the move where the evaluation turned.
Where to play blitz, and where Chessdrive fits
Lichess and Chess.com are better places to play blitz than Chessdrive. They have large player pools and rating lists for each time control, and Chessdrive is an improvement and analysis app, not a blitz server. Play your games there. Our guide to playing chess online for free compares the main options.
Chessdrive fits after the game. A Stockfish review checks every move, marks the turning points and explains the ones that mattered at your level, which is the review step blitz players tend to skip. If you have not tried a review yet, our free chess game review guide shows what a useful one looks like. Playing blitz and then reviewing three of your losses a week is a more reliable routine than playing more games.
Method and sources
Data: Lichess open database, standard rated games, October 2026 file (CC0), read from the first 1,521,984 games in the file, all from 1 October 2026. We streamed it and kept no local copy. Time controls use Lichess's classification (base plus 40 times the increment). Ratings bands are under 1000, 1000 to 1399, 1400 to 1799, 1800 to 2199 and 2200 and above, using the mover's rating for moves and the average of both players for game-level shares. Games with a rating gap over 400 were skipped. Blunder: win-probability drop of 30 points or more, with the first 10 plies skipped. Definitions: FIDE Laws of Chess Appendix B, the Lichess FAQ and the Chess.com time controls page. Coach views: Chess.com articles by Jeremy Kane and Bruce Pandolfini. All numbers in the tables come from our run of that file.
Frequently asked
- What is blitz chess?
- Blitz chess is chess played with a short clock. FIDE calls a game blitz when each player has 10 minutes or less, counting 60 seconds for every second of increment. Chess.com says three to ten minutes per player. Lichess says an estimated duration under 480 seconds, counting 40 seconds per second of increment.
- Is blitz chess harder than rapid?
- In our data it is harder for strong players and about the same for weak ones. At 2200 and above, players blunder 1.87 times per 100 moves in blitz against 1.42 in rapid. Below 1000, rapid is no cleaner at 6.15 against 5.71. The data is one day of Lichess games and shows a rate, not a cause.
- Is blitz chess good for beginners?
- It is fine in moderation if you review your games. Coaches disagree: one Chess.com instructor calls blitz the best training, and others warn it can erode strategy or breed bad habits. Below 1000, our data shows 5.71 blunders per 100 moves in blitz, and 6.15 in rapid, so a slower clock alone does not cut the blunders.
- Does blitz chess make you worse?
- Our data cannot say, because it compares different games, not the same players over time. The coaches we found warn that heavy speed chess can make you neglect strategy in slower games. The common forum view is that blitz hurts when you never review the games, and helps when you do.
- How long is a game of 3+2 or 5+3?
- On Lichess's count, 3+2 is 180 plus 80, or 260 seconds, and 5+3 is 300 plus 120, or 420 seconds. FIDE's count gives 300 and 480 seconds, which is 5 and 8 minutes. A Chess.com member blog's formula gives about 4.3 and 7 minutes. All three call both blitz.
- What are the best blitz chess tips?
- Keep a blunder check on every move, since players blunder more once they drop under 10 seconds. Play familiar opening lines quickly to bank time for the middlegame. Try an increment control such as 3+2 or 5+3, which adds time after every move. Then review the games afterward, because that is where the learning happens.
- How many blitz games end on time?
- About a quarter. In our Lichess data 22.6% to 25.6% of blitz games end on time at every rating band, against 44.8% to 65.1% of bullet games and 10.5% to 13.0% of rapid games.
Play blitz where it is fast, review it where it is slow
Chessdrive is not a blitz server, so play your games on Lichess or Chess.com. When you want to know which moves lost them, a Stockfish review in Chessdrive checks every move and explains the ones that mattered at your level. On iOS and Android, $5.99/month or $39.99/year on the US storefront with a free trial. Prices vary by region.
Keep reading
Updated