Elephant Gambit

Its origins are unclear, but records show players using it as early as the 1800s. The Elephant Gambit rarely appears at a high level, and White scores better, as it does against most gambits.

The Elephant Gambit in One Minute

Win Rates: What the Numbers Say

Across master databases, White's extra pawn shows up plainly in the score.

Among amateurs, the gap narrows on wins but the draw rate collapses, which tells its own story about how sharp these positions get below master level.

Source: Lichess opening explorer, blitz, rapid and classical games at the position after 1.e4 e5 2.Nf3 d5. Figures are win rates by Lichess rating band; draws are rare at these levels.

Watch: The Gambit in Action

The video below walks through the gambit in about five minutes.

The video covers the basics; read on to learn the deeper lines and the gambit's hidden resources.

▶

Why Is It Called the Elephant Gambit?

No one knows for certain who invented the Elephant Gambit or where its name comes from. Records show it in casual games as far back as the early 1800s, during the Romantic era of attacking chess. Its main branches were later named after the masters who explored them: the Paulsen Countergambit after Louis Paulsen, and the Maróczy Gambit after Géza Maróczy.

The name itself has a few competing explanations. One theory is that the two advanced central pawns resemble an elephant's tusks. Another points to the fact that, in several languages, the bishop is historically called the elephant, and Black's quick bishop development is central to the gambit's spirit.

Study the Elephant Gambit as White or Black

Key Ideas Behind the Elephant Gambit: What Black Is Going For

The main goal is to dislodge White's minor pieces and grab the center. To do that, Black sacrifices a pawn, striking with the thematic ...e4 pawn push. White can stay active, use the queen and the f3-knight, and come out ahead in material or position. Ne5, d4, and Qe2 are typical white moves that put the e-file to work.

Playing the Elephant Gambit as Black: Practical Tips

If you want to use the Elephant Gambit as a surprise weapon, treat it like any gambit: forget about the pawn and play for activity. Black's whole point is to drag White out of preparation, so it works best in faster time controls where your opponent has to solve problems over the board.

A few practical rules help Black get the most out of it. Develop the bishops quickly, castle early, and keep the pieces pointed at White's kingside and the e4-square. Avoid passive moves like ...Be7 that leave you a tempo short. If White plays the critical 3.exd5, the most ambitious try is 3...e4, kicking the f3-knight and grabbing space before White gets organized. The honest truth is that against precise play Black is worse, but in a real game one inaccurate move from White and the initiative is yours.

Elephant Gambit Theory: Main Variations (3.Nxe5, 3.exd5, 3.d4)

Main Variations at a Glance

Here are the branches you are most likely to meet, with the point of each in a single line.

Paulsen Countergambit: 3.Nxe5 line

This variation starts after White grabs the e5-pawn with the f3-knight (3.Nxe5). Black has two main replies: 3...Bd6 (attacking the unprotected e5-knight) and 3...dxe4 (taking the pawn back). Black wants to dislodge the knight, and White wants to attack on the kingside, where f7 is a constant target.

If Black plays 3...Bd6, White needs to protect the e5-knight with 4.d4.

Anything else, such as 4.Nf3, hands Black the initiative because White would drop the e4-pawn. If Black captures the knight with 4...Bxe5, White gets the bishop pair and a real edge into an endgame. For example, 4...Bxe5 5.dxe5 dxe4 (recapturing the pawn) leads to a queen trade with 6.Qxd8+ Kxd8, and White stands clearly better thanks to the bishop pair and the uncastled black king.

So Black usually takes the e4-pawn (4...dxe4) instead. This lets White develop quickly with 5.Bc4, eyeing f7. Black can remove the e5-knight with 5...Bxe5, but White does not have to recapture with the pawn and trade queens, which would cost the right to castle. White plays 6.Qh5, threatening both Qxf7 and Qxe5. If Black defends f7 with 6...Qe7, White takes on e5 with 7.Qxe5, and after 7...Qxe5 8.dxe5 reaches a favorable endgame where White can still castle short and keep the upper hand.

A move like 5...Nh6, trying to cover f7, is poor because 6.Bxh6 wrecks Black's kingside structure.

Most of these 3...Bd6 lines steer into an endgame. To handle them well, you need a feel for piece placement and the resulting pawn structure.

Black can also skip the endgame and take on e4 directly with 3...dxe4. White can simplify with 4.d4 (4...exd3 5.Bxd3 gives a clean, pleasant position for both sides).

White can also play the natural 4.Bc4, piling pressure on f7. Here Black has only one sensible reply to keep things under control: 4...Nh6.

4...Qg5 is a try to muddy the waters, but after 5.d4 (defending the e5-knight) and 5...Qxg2, the shot 6.Bxf7+ strips Black of the right to castle, and White can play Rf1 without much worry about king safety, intending to castle long later.

4...Qd4 is a similar idea, and White can grab f7 with 5.Bxf7+ and then guard the knight with 6.f4.

After 4...Nh6 (guarding f7), White plays 5.d4 to open the c1-bishop's diagonal and hit the h6-knight. The game can drift into an endgame if Black finds the fine resource 5...Nd7; one sample continuation runs 6.O-O Nxe5 7.dxe5 with queens coming off on d1.

Maróczy Gambit: 3.exd5 variation

This line starts after White captures on d5 with the e-pawn (3.exd5). It is objectively better than 3.Nxe5 because White usually keeps a clean extra pawn with accurate play.

Black usually pushes 3...e4 to chase the f3-knight. Black can also defend e5 with 3...Bd6. The recapture 3...Qxd5 is playable too, but it lets White develop with tempo via Nc3.

One sample line on 3...e4 runs 4.Qe2 (pinning the pawn) 4...Nf6 (defending it, and note that 4...Qxd5 walks into 5.Nc3) 5.d3 (hitting the pinned e4-pawn) 5...Qxd5 6.Nbd2 (with the d-pawn already advanced, Nc3 would allow the ...Bb4 pin) 6...Nc6 7.dxe4. White is a pawn up, though the f1-bishop still needs the queen to step aside (for instance Qe3 and then Bc4 or Bb5) before castling short.

In the 3...Qxd5 line, Black regains the pawn but allows 4.Nc3 with tempo. After 4...Qa5 (protecting e5 and tucking the queen onto a safe square), 5.Bc4 gives White an excellent position. White wants to castle short and swing the f1-rook to e1, pressing the advanced e5-pawn.

In the 3...Bd6 line, White is already a pawn up and just needs to untangle the doubled pawns; 4.d4 does the job well. If Black plays 4...exd4, White recaptures either way (5.Qxd4 or 5.Nxd4). More testing is 4...e4.

Here it is important not to play 5.Qe2, because after 5...Nf6 Black castles next move and lands the f8-rook on the e-file. So 5.Nfd2 is the safe square for the knight. White then aims for Be2 and short castling to consolidate, while Black looks to hit White's central majority and push ...e3 at some point.

The 3.d4 Sideline

This line starts with 3.d4, challenging the center directly. It is a sideline because Black can often equalize and avoid losing a pawn.

If Black captures on e4 with 3...dxe4, White takes the e5-pawn with 4.Nxe5.

This is the same position we saw in the 3.Nxe5 line (3.Nxe5 dxe4 4.d4), with one difference: Black can no longer recover the pawn with the ...exd3 en passant resource. That quietly improves White's chances.

Black can also take the d-pawn with 3...exd4, and White recaptures with 4.Qxd4. The game can head to an endgame if Black grabs the e-pawn: 4...dxe4 5.Qxd8+ Kxd8. Then 6.Ng5 hits f7 and e4 at once and wins the e4-pawn back with a small edge.

Black can also try 4...Nc6 to hit the queen, met by 5.Bb5 (pinning the knight). Queens come off much as before, with Bxc6 and Ng5 ideas to follow.

These positions tend to reach an endgame where Black often loses castling rights, since the exchanges keep landing on d8.

Declined and Sideline Responses: 3.d4, 3.Nc3, 3.d3 and 3.Bb5+

White does not have to enter the sharp captures at all. The most respectable sideline is 3.d4, challenging the center directly. After 3...dxe4 4.Nxe5 the position transposes into the 3.Nxe5 lines, with one difference: Black can no longer recover the pawn with the ...exd3 en passant resource, which quietly improves White's chances. The line 3...exd4 4.Qxd4 is also pleasant, since the queen trade on d8 wrecks Black's castling. Declining with 3.d3 keeps a small, safe edge and sidesteps theory. The move 3.Nc3 invites a transposition where the d5-pawn becomes awkward to hold. The check 3.Bb5+ is the rarest try, usually met by 3...c6 or 3...Bd7. None of these refute the gambit outright, but each leaves White comfortable without memorizing the critical move orders.

How to Counter the Elephant Gambit: Refuting It as White

The simplest way to counter the Elephant Gambit is to accept it. Black is hoping White will panic and play passively, but a few accurate moves leave Black a pawn down with little to show for it. The two main refutations are 3.exd5 and 3.Nxe5, and both are good for White.

The cleanest refutation is 3.exd5. After 3...e4 (the Paulsen Countergambit), White does not need to retreat the f3-knight. Instead, 4.Qe2 pins the e4-pawn against Black's king. A typical continuation is 4...Nf6 5.d3 Qxd5 6.Nbd2, when Black has no comfortable way to hold the e4-pawn and White stays a pawn up with a healthy position.

If you prefer to keep the structure simple, 3.Nxe5 is also fully sound. The key practical rule for White is the same in every line: develop, target f7, and avoid greedy moves that hand Black the initiative. As long as White returns material only on its own terms, Black does not get enough for the sacrificed pawn.

What the Engine Actually Says About the Soundness

Strip away the romance and the engine verdict is blunt. After the cleanest refutation 3.exd5, White's evaluation settles around +1, the value of a healthy extra pawn with no real compensation in sight. That is why every objective assessment files the Elephant Gambit under "dubious." Black gets open lines and quick bishop development, but no forcing initiative that an engine fears, and the typical endgames after the queens come off only deepen White's edge. The practical caveat matters, though. That +1 evaluation assumes White finds the precise continuation every move, and over the board Black's threats against f7 and g2 are easy to mishandle. So the honest summary runs like this: theoretically near-losing for Black, yet a real practical try at faster time controls, where the defender has to solve those problems on the clock with no help.

How the Elephant Gambit Compares to Other Offbeat Gambits

The Elephant Gambit belongs to a family of rare, aggressive openings that trade soundness for surprise value. If you enjoy this style, it helps to know how it stacks up against its cousins.

Compared with these, the Elephant Gambit sits in the middle: less reckless than the Jerome Gambit, but less reliable than the Urusov. Its strength is that it appears as early as move two, so White almost always has to think independently right away.

The 4.Qh5+ Tragedy: Why 3...f6 Loses on the Spot

This is the move that quietly decides more Elephant Gambit games than any line in the theory section. After 3.Nxe5, the beginner's instinct is to kick the centralized knight with a pawn: 3...f6??. It looks like a refutation. It loses immediately. White replies 4.Qh5+, and the only way to meet the check is 4...g6, because 4...Ke7 walks into disaster. Then 5.Nxg6 exploits the pin along the h-file. Black must take with 5...hxg6, and 6.Qxh8 scoops the rook. The correct third moves are entirely different: 3...Bd6, hitting the knight with a piece, or 3...dxe4, simply recovering the pawn. This single mistake is also the hidden reason behind that anomalous 4 percent amateur draw rate. Below roughly 1500, a large share of games never reach a middlegame at all, because Black plays the natural-but-fatal ...f6. This exact sequence keeps appearing against me, and the queen check ends things before move ten.

Master Games with the Elephant Gambit

Facing the Elephant Gambit as White: One of My Own Games

I have had this one from the White side, rated 1864 against an opponent on 1858. It went 1.e4 e5 2.Nf3 d5 3.Nxe5 Bd6 4.d4 dxe4 5.Bc4 Bxe5 6.dxe5 Qxd1+ 7.Kxd1. The engine liked White by about +0.8 in the opening, but once the queens came off and my king was sitting on d1, it read around -0.5.

Black handed it straight back with 9...Bg4+, which moved the reading from -0.06 to +2.05. I answered 10.f3, and after 10...O-O-O+ the engine had White at +1.41. Then I threw it away myself. My 11.Bd2 turned that +1.41 into -0.60, and after 11...Nxc4 12.Rxc4 Be6 13.Ra4 my rook was touring the board while the evaluation slid past -3 by move 15. I lost.

So if you are deciding whether to meet the Elephant Gambit with the critical 3.Nxe5, take it, but take it knowing what it gives you. Usually a pull, not a refutation, and a queenless middlegame with a king that has lost the right to castle. That is the part worth a look before the game, not the first six moves.

Should You Add It to Your Repertoire?

Pros

• White usually has a material advantage with a proper Elephant Gambit response.
• The endgames are often advantageous for White.
• White can decline the complications and still gain a slight edge.
• Objectively, this gambit is not sound.

Cons

• If White doesn’t know the Elephant Gambit counter, the f3-Knight can be misplaced.
• Black can gain central space by advancing the e- and f-pawns forward in many lines.
• In some cases, Black can exploit White’s lack of development on the Kingside.
• Black can still fight after they are down a pawn if the Queens are not swapped.

Elephant Gambit Traps: The Wasp Variation and the 3...f6 Trap

The Elephant Gambit also sets traps in the other direction, so White has to stay alert. After 3.Nxe5 dxe4 4.Bc4, Black can play the tricky 4...Qg5, hitting both the e5-knight and the g2-pawn. This is the start of the Wasp Variation.

The greedy 5.Nxf7 looks tempting because it forks the queen and rook, but it walks straight into the trap. Black replies 5...Qxg2, attacking the h1-rook, and after 6.Rf1 Bg4 7.Be2 Bxe2 8.Qxe2 Kxf7 Black simply scoops up the stranded knight and ends up better. Instead, White should keep things solid, for example the 5.d4 idea shown earlier in the 3.Nxe5 line, and not chase material into Black's counterplay.

This trap arises in the 3.Nxe5 line when Black tries to kick the knight with the careless 3...f6. Now 4.Qh5+ threatens Qxf7 next move, so 4...g6 is forced to block the check. Then 5.Nxg6 exploits the pin along the h-file. Black has to take the knight with 5...hxg6 and give up the rook with 6.Qxh8 to avoid getting mated.

Case Study: White's Model Refutation, Iagoupov vs. Kilic

The cleanest way to see the gambit punished is a full game where White does nothing fancy: accept the pawn, develop, and aim at f7. This 2025 game is a model of that plan, all the way to mate.

White takes on d5 with 3.exd5 and answers 3...e4 with 4.Qe2, pinning the e4-pawn against the king. Keep the focus on development and calculate the pressure on e4 before claiming a forced material gain.

The turning point is the king walk that never gets punished. Black throws pieces forward with 13...Qh7 and 14...cxd5, but White calmly castles long (14.O-O-O) and plants the knight on d6 (15.Nd6), a monster outpost deep in Black's camp. Notice how every white piece points at the kingside and the soft f7-square, exactly the target the theory keeps flagging. Black's 19...Nb4 and 20...Nxa2+ look threatening, but they are lunges without support, and White's king strolls to d2 and c1 unharmed while the extra material tells.

The finish is pure f7. After 30.Bxf7, the bishop pries open Black's king, 32.Nxg6+ rips away the last shelter, and 33.Be6+ Kxg6 34.Qe4# ends it. The practical takeaway for a club player is exactly what the engine verdict predicts: against the Elephant Gambit you do not need a prepared knockout. Accept the pawn, develop toward f7, refuse to grab loose material when it gives Black tempo, and let Black's loose attack burn itself out. The extra pawn does the rest.

Final Verdict on the Elephant Gambit

The Elephant Gambit hands over a pawn early to grab the initiative. It is not sound: White can consolidate and keep the extra pawn in most lines. Black tries to expand in the center, but with precise play White flips the script and takes the center instead. For all those reasons it stays off the elite stage, though it remains a genuine surprise weapon in fast games where one slip by White can swing the position.

Test Yourself: Elephant Gambit Positions

Make the move on the board. Hint gives you a nudge, Answer shows the move and why it works.

Frequently Asked Questions