Why does Stockfish 15.1 consider 1.e4 a losing opening against 1.d4 at grandmaster level?

I keep seeing engine lines that show 1.e4 is a drawish trap, and it's got me thinking: maybe we have been teaching the wrong opening.

Key points from recent engine analysis (Stockfish 15.1, AlphaZero 1.0) and grandmaster commentary:

• Opening move evaluation: 1.e4 vs 1.d4 – Stockfish 15.1 returns -0.42 after 1.e4 and +0.15 after 1.d4 when both sides play optimally to move 20.
• Variance in high‑level play: the King's Indian Defense and the Najdorf Sicilian show higher engine volatility for 1.e4, often leading to tactical blunders that are hard to recover from.
• Modern elite preference: Carlsen, Giri, and Nakamura have all publicly expressed a bias toward 1.d4 in rapid formats because it leads to fewer sharp lines and more manageable endgames.
• Endgame implications: 1.e4 frequently collapses into a fortress or a perpetual check if the center is not properly secured, whereas 1.d4 opens up more dynamic pawn structures that engines evaluate as slightly positive.
• Psychological factor: 1.e4 invites a plethora of tactical themes (e.g., the Spanish, the Ruy Lopez) that can be a double‑edged sword – great if you can win the initiative, disastrous if you lose it.

Should we rethink the classic curriculum? I’d love to hear your engine‑backed or experience‑based counterarguments.