The queen's pawn opening is 1.d4, White's pawn moving two squares in front of the queen. Sources disagree on how far the name stretches. Chess.com uses it for 1.d4 alone. As of October 2026, Wikipedia's Queen's Pawn Game is any opening starting with 1.d4, or in some usage only those where White avoids the Queen's Gambit (1...d5 2.c4), and it calls 1.d4 the second most popular first move after 1.e4. Our page on the best first move in chess puts that against Stockfish, which scores 1.d4 and 1.e4 identically. Lichess narrows it to 1.d4 d5, which Chessiverse calls Queen's Pawn Systems under ECO D00. Wikipedia's wider list runs A40 to A99, D00 to D99 and E00 to E99. This page uses the narrow meaning for its data, the wide one for its maps, and says which is which.
Most pages we read define the term and stop. We ran the maia3-5m checkpoint (see how Maia chess actually works) on the position after 1.d4 d5 at seven rating bands from 1000 to 2200 and scored every legal move with Stockfish 18 at depth 20. We then set the model's predictions next to real games from Chessiverse, which builds on Lichess blitz and rapid play, as of October 2026. The two agree closely, and that agreement is the finding. If you want the sibling page for 1.e4, it is the King's Pawn Opening.
Evaluations here are from White's side unless stated, so +0.31 means White is better by 0.31 pawns and a negative number means Black is. Treat any position under about 0.30 as roughly level, because no side can claim it.
White's second move after 1.d4 d5, by rating
After 1.d4 d5, Stockfish ranks 2.c4 first at +0.31, then 2.Nf3 at +0.29, with 2.Bf4 well back at +0.07. In each cell below, the first figure is the share Maia3 predicts for a player at that rating and the second is the share of real games at that rating on Chessiverse (523,262,183 Lichess blitz and rapid games in the position as of October 2026; the bracket labels are theirs).
| Rating | 2.c4 (+0.31) | 2.Bf4 (+0.07) | 2.Nf3 (+0.29) |
|---|---|---|---|
| 1000 | 25.6 / 25.4 | 23.0 / 22.3 | 16.1 / 15.3 |
| 1200 | 31.8 / 31.8 | 24.9 / 22.7 | 15.6 / 14.9 |
| 1400 | 36.0 / 37.1 | 24.2 / 21.4 | 15.5 / 15.1 |
| 1600 | 40.0 / 41.4 | 22.0 / 19.1 | 15.8 / 16.1 |
| 1800 | 44.1 / 44.6 | 18.5 / 16.8 | 17.0 / 18.5 |
| 2000 | 47.0 / 47.2 | 14.9 / 13.8 | 20.6 / 22.7 |
| 2200 | 49.1 / 50.4 | 11.7 / 10.7 | 25.4 / 26.2 |
Percentages, Maia3 predicted / real games. Chessiverse does not report 2.e3 or 2.Nc3 per band, so those moves appear only in the next section.
Across these 21 cells the mean absolute gap between model and real games is 1.1 percentage points, and the largest is 2.9 (2.Bf4 at 1600). The first headline is that 2.c4 is both the engine's top move, by 0.02 over 2.Nf3, and the most played move at every band from 1000 to 2200, and its share roughly doubles: 25.6 to 49.1 percent in the model, 25.4 to 50.4 in real games. Chessiverse's other brackets as of October 2026 point the same way, with 2.c4 at 56.3 percent at 2500. At 400 it is not even in their top three (2.Bf4 at 19.2, 2.Nc3 at 17.8, 2.e3 at 17.1), a band our model does not cover.
The second headline is the race for second place. 2.Bf4 is second through the club bands and then fades; 2.Nf3 takes over. In real games the change comes at 1800, where 2.Nf3 passes 2.Bf4, 18.5 to 16.8 percent. Maia gets there later: at 1800 it still has 2.Bf4 ahead, 18.5 to 17.0, and 2.Nf3 is ahead of it from 2000, 20.6 to 14.9. That is a real difference between the two sources, and the model is one band late. The model also overstates 2.Bf4 at all seven bands, by up to 2.9 points.
This does not prove the model right everywhere. It shows that for this one position, the predictions track what Lichess players actually do. The limits section says where that stops holding.
What it means at your rating
- 1000 to 1600. Expect 2.c4 about one game in four at 1000, rising to two in five at 1600. 2.Bf4 is the common alternative, 19 to 23 percent of real games. It scores +0.07, which is 0.24 of a pawn below 2.c4 and still level, so it is a style choice, not a mistake. The London System page covers that setup.
- 1800 and up. Prepare for 2.c4 in 45 to 50 percent of games and for 2.Nf3 as the main side line. The engine puts 2.Nf3 nearly level with 2.c4, 0.02 apart, so players who choose it are not giving up anything.
- Your own repertoire as White. The two moves with the strongest engine scores are also the two most common at 2000 and 2200. Below that, the popular 2.Bf4 gives up 0.24 by the engine's count and stays level, so choose it or 2.Nf3 on the plans you want to play, not on these numbers.
For how openings change with strength in general, see chess openings by rating.
Black's first replies to 1.d4
We have no reliable frequencies for Black's reply to 1.d4, so we do not quote any. What we have is engine scores from a local Stockfish check at about ten seconds, with Black to move after 1.d4, shown from White's side: 1...Nf6 +0.23, 1...d5 +0.27 and 1...c6 +0.36. The first two are inside 0.05 of each other and the third is 0.13 away. None of the three is a problem for Black.
The names matter for finding more. Wikipedia files 1...d5 as the Closed Games (D00 to D69), the Keres Defense under A40, the Pillsbury Defense under A41, the Dutch Defense under A80 to A99, and the Indian defenses across A45 to A79, D70 to D99 and E00 to E99. Our pages for those branches are linked in the section on Black's answers below.
White's systems after 1...d5: the map
Many d4 openings are setups, not single moves. The defining moves follow Wikipedia, and each line shown is the exact position we ran. The score is from White's side, for the position shown, with the side to move in brackets.
| System | Moves | Score | Favours |
|---|---|---|---|
| Colle-Koltanowski | 1.d4 d5 2.Nf3 Nf6 3.e3 e6 4.Bd3 c5 5.c3 (Black) | -0.05 | Level |
| Colle-Zukertort | same, 5.b3 (Black) | +0.16 | Level |
| Stonewall | 1.d4 d5 2.e3 Nf6 3.Bd3 c5 4.c3 Nc6 5.f4 (Black) | -0.86 | Black |
| Veresov | 1.d4 d5 2.Nc3 Nf6 3.Bg5 (Black) | -0.29 | Level |
| Jobava London | 1.d4 d5 2.Nc3 Nf6 3.Bf4 (Black) | -0.09 | Level |
| Torre | 1.d4 d5 2.Nf3 Nf6 3.Bg5 (Black) | -0.15 | Level |
| Barry | 1.d4 Nf6 2.Nf3 g6 3.Nc3 d5 4.Bf4 (Black) | +0.06 | Level |
| Blackmar-Diemer | 1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 (Black) | -0.72 | Black |
| Catalan | 1.d4 d5 2.c4 e6 3.Nf3 Nf6 4.g3 (Black) | +0.21 | Level |
The Colle and Torre both begin with 2.Nf3, which Wikipedia lists with the Colle's 3.e3 and the Torre's 3.Bg5. The Colle-Koltanowski and Colle-Zukertort differ in the fifth move here, 5.c3 against 5.b3, and the engine sees neither as an edge. The Barry Attack begins 1.d4 Nf6 but reaches a ...d5 structure, so it belongs on this map.
Two entries are lopsided. The Stonewall figure is for one specific order, 4.c3 Nc6 5.f4, where Black replies 5...Bg4 and scores -0.86. It says nothing about the structure reached by other move orders. The Blackmar-Diemer Gambit (-0.72 after 4.f3) fits what Wikibooks says of 2.e4, that it is dubious against best play. If Black accepts further with 4...exf3 5.Nxf3, the best reply 5...c6 holds -0.66, so the gambit does not repair itself in the next move.
Each score is one position at about ten seconds of local Stockfish, depth 22 to 28. It is not a verdict on an opening, only on that position.
Black's sharper tries after 2.c4
Two second-move replies to the Queen's Gambit leave the main roads. In the Chigorin Defense (1.d4 d5 2.c4 Nc6), after 3.Nf3 the best reply 3...e6 leaves White ahead by 0.62. In the Albin Countergambit (1.d4 d5 2.c4 e5 3.dxe5 d4), White to move after 3...d4 has the best move 4.Nf3 and stands at +0.82. These are single positions, not refutations, but both give White a clearer edge than any other line in this article.
How Black answers: the queen pawn defense question
"Queen pawn defense" is not a standard opening name, so we read it as the question of how Black meets 1.d4. We can answer only part of it. The sources support two families.
The first is 1...d5, the symmetrical reply, which Chessiverse files as D00. After 2.c4 it splits into the structures on our Queen's Gambit page, the Slav Defense page, and the Queen's Gambit Accepted page. The second is 1...Nf6, which keeps Black's central choice open. Setups that start there are covered in the King's Indian Defense, Nimzo-Indian Defense and Grunfeld Defense guides, which Wikipedia's ECO ranges place under the Indian defenses. Against 1...f5, see our Dutch Defense guide.
As of October 2026, Chessiverse reports the overall results in the whole 1.d4 d5 position, blitz and rapid only: White 51.3 percent, draws 4.6 and Black 44. That is the whole position, not any one second move, so it does not tell you how 2.c4 performs against 2.Bf4.
The real frequency pass: every second move
The Lichess page for 1.d4 d5, pool 1000 to 2500 across all speeds with the game count not shown, lists White's second moves as of October 2026. 2.c4 39 percent, 2.Bf4 19, 2.Nf3 18, 2.e3 10, 2.Nc3 7, 2.e4 3, 2.Bg5 2, 2.c3 2, 2.g3 1 and 2.f4 1. The shares sum to 102 because of rounding. This is where the minor moves show up, and the overall 2.c4 figure matches Chessiverse: weighting its per-band figures by game counts gives 38.7 percent against Lichess's 39. That cross-check assumes each bracket label covers the page's bracket count.
- 2.e3. Lichess calls it a Queen's Pawn Game and it takes 10 percent. Maia3 puts it at 12.9 percent at 1000, falling to 3.0 at 2200, with an engine score of +0.20. The Stonewall starts here, and the Colle gets there after 2.Nf3.
- 2.Nc3. Lichess names it the Chigorin Variation, not to be confused with the Chigorin Defense above, at 7 percent. Maia3 has it at 11.6 percent at 1000 and 4.1 at 2000, with a score of +0.07. The Veresov and the Jobava London in the table above both start with it.
- 2.e4. The Blackmar-Diemer Gambit, 3 percent. Wikibooks says it is dubious against best play, and our check agrees.
- 2.Bg5, 2.c3, 2.g3, 2.f4. One to two percent each. Our dataset gives 2.Bg5 -0.24, 2.c3 +0.20 and 2.g3 +0.18.
Wikibooks calls 2.Nf3 the main sideline in tournament play and 2.Bf4 the main club-level sideline. The Lichess figures agree for clubs: 2.Bf4 is second through 1600 in both of our sources. Overall it still ranks above 2.Nf3 on Lichess, 19 percent to 18.
If you face 2.Bf4, Lichess shows Black's replies, in the same pool and as of October 2026, as 2...Nf6 at 27 percent, 2...Nc6 at 24, 2...e6 at 19, 2...Bf5 at 13, 2...c6 at 6 and 2...c5 (the Steinitz Countergambit) at 6. Our London System page goes through the setups that follow.
Where other resources are better
This page is a map, not a repertoire. Wikipedia is the better source for history, ECO ranges and the full list of named systems, and for how far each name stretches. Lichess's explorer gives live counts that you can filter by rating and speed, which we have not reproduced. Chess.com has a plain description of 1.d4 with its pros and cons, central space and slower build-up against fewer chances of a direct attack. Chessiverse covers more brackets than we do, 400 to 2500, plus win rates. What none of the pages we read does is put engine scores, model predictions and real-game frequencies in one table for the same position, which is what the section above tries to do.
Limits
The model is Maia3 5m, a small network, and it was checked against real games in one position. The same model, one move later (1.d4 d5 2.c4 dxc4), predicts 3.Nc3 as the likeliest move from 1600 to 2200, while the chess.com sample on our Queen's Gambit Accepted page had 3.Nc3 at 0.4 percent as of September 2026. So we left that table out, and that is why the second-move figures are shown next to real games rather than alone.
Other limits: Chessiverse's pool is blitz and rapid Lichess games and Lichess's own page includes all speeds, so the two real-game columns are not a single pool. The Stockfish numbers are single positions at about ten seconds, and a separate check of the 1.d4 d5 position moves the second-move scores by up to 0.08 of a pawn, with 2.c4 and 2.Nf3 still first and second. We have no results by rating for any move, so we cannot say which second move scores best at 1200. We have no Jobava London, Catalan, Chigorin or Albin game counts, only the move orders and one engine line each.
What to take from it
The Queen's Pawn Opening means 1.d4, with the wider and narrower usages kept apart. After 1.d4 d5, 2.c4 is the highest-scored move, by 0.02 over 2.Nf3, and the likeliest at every band from 1000 to 2200, and the model and real games agree on it to within a few points. The rating story is 2.Bf4 giving way to 2.Nf3 as second choice, at 1800 in real games and 2000 in the model. Of the named systems we scored, only the Stonewall order shown and the Blackmar-Diemer Gambit left the engine clearly favouring Black. For the 1.e4 side of the same comparison, see the King's Pawn Opening.
Frequently asked
- What is the Queen's Pawn Opening?
- Chess.com uses the name for 1.d4 alone. Wikipedia's Queen's Pawn Game is any opening that starts with 1.d4, or only those where White avoids the Queen's Gambit, depending on the author. Lichess applies the name to 1.d4 d5, which Chessiverse calls Queen's Pawn Systems under ECO D00. Wikipedia lists the whole family across A40 to A99, D00 to D99 and E00 to E99.
- Is the Queen's Pawn Opening the same as the Queen's Gambit?
- No. The Queen's Gambit is one second move, 2.c4, after 1.d4 d5. Lichess's explorer shows it in 39 percent of games in that position, as of October 2026. The Queen's Pawn Opening is the wider family that also holds 2.Bf4, 2.Nf3, 2.e3 and the rest.
- What is the most common second move after 1.d4 d5?
- 2.c4. Lichess's page for the position, pool 1000 to 2500, shows 2.c4 at 39 percent, 2.Bf4 at 19 and 2.Nf3 at 18, as of October 2026. By band, as of October 2026, Chessiverse has 2.c4 rising from 25.4 percent at 1000 to 50.4 percent at 2200.
- Is 1.d4 better than 1.e4?
- Our engine data sees no gap. In local Stockfish checks, Black's three best replies leave White ahead by 0.23 to 0.36 pawns after 1.d4 and by 0.27 to 0.40 after 1.e4. Those ranges overlap, and each number is a single position at about ten seconds, so it is not a verdict on either move.
- Is 2.Bf4 or 2.Nf3 the better second move for a club player?
- Both are sound. The engine scores 2.Nf3 at +0.29 for White and 2.Bf4 at +0.07, and 2.c4 at +0.31. At 1000 to 1600 more players choose 2.Bf4, and from 1800 in real games more choose 2.Nf3. Pick the one whose plans you will actually study.
- How does Black answer 1.d4?
- The two replies the engine scores closest to equal are 1...Nf6 (+0.23 for White) and 1...d5 (+0.27), with 1...c6 at +0.36. We have no frequency data for Black's first move, so we cannot say which is more popular. The Indian defenses start with 1...Nf6, and the Dutch Defense is 1...f5.
Want to play 1.d4 against a model that answers like your rating?
Chessdrive runs the real maia3-5m checkpoint on-device as your opponent at any rating you set from 600 to 2600, so you can open with 1.d4 and meet the replies a player at your band actually chooses, then see where the game turned with a separate Stockfish review. Available now on iOS.
Keep reading
Updated