Conquering The Physics Gre

How I Survived the Physics GRE (And What Actually Works)

I took the GRE Physics Subject Test three times between 2018 and 2020. Got a 780, then a 820, then stopped caring. Most people asking about Conquering The Physics Gre want a quick fix. There isn't one. But there are things that separate the kids who score in the 700s from the ones who grind for months and plateau at 650. First, a clarification nobody makes: the GRE Physics test isn't about knowing everything. It's about pattern recognition under time pressure. The exam gives you 170 minutes for 100 questions. That means 102 seconds per question on average. You cannot derive Lagrangian mechanics from first principles for five of the classical mechanics problems. You need to know the answer before you read the full question. I learned this the hard way on my first attempt. Scored a 710. Felt like garbage because I'd spent 3 minutes on a thermodynamics question about entropy changes in irreversible processes, then panicked on the last ten questions. The test doesn't reward thoroughness. It rewards speed with acceptable accuracy.

What Actually Matters (And What Doesn't)

The test covers roughly: — Classical Mechanics: 20% — Electromagnetism: 18%

— Optics and Wave Phenomena: 9% — Thermodynamics and Statistical Mechanics: 10% — Quantum Mechanics: 12%

— Atomic Physics: 10% — Special Relativity: 6% — Laboratory Methods: 6%

— Additional Topics: 9% Classical mechanics and electromagnetism alone make up nearly 40% of the exam. If you're solid on these two and have working knowledge of quantum and thermodynamics, you're already in a good position. The rest is memorization and shortcut recognition. Here's the thing that surprised me: special relativity is almost free points if you know your Lorentz transformations cold. The questions are short, the math is straightforward, and they appear 5-6 times on every version. I dedicated about 3 hours total to SR prep and gained roughly 30 points on my second attempt. That's the highest ROI topic on the entire exam.

My Study Strategy (What Worked)

I used three resources: a review book (I liked Schroeder's "Thermodynamics" for the stat mech parts, but for overall prep, theETS official guide plus the Peterson's guide), past exams (there are some floating around online), and a lot of practice problems timed strictly. My routine for about 8 weeks before the second attempt: Weeks 1-3: Rapid review of all topics. Not deep study—just enough to remember what exists. I'd spend 2-3 hours per day going through standard undergrad textbooks (Griffiths for EM, Taylor for classical, Shankar for quantum) and doing example problems. The goal was coverage, not mastery.

Weeks 4-6: Targeted practice. I identified my weak spots (atomic physics and lab methods were brutal for me) and spent extra time there. For atomic physics, I memorized the selection rules and the hydrogen atom energy levels. For lab methods, I learned error analysis, basic circuit theory, and detection methods. These topics are small but the questions are quick to answer if you know the material. Weeks 7-8: Full practice exams under timed conditions. This is where most people skip, but it's the most important part. I took at least five full practice tests, each in one sitting, strictly 170 minutes. The fatigue factor is real—you need to build mental stamina for 3 hours of intense problem-solving. One specific technique that helped: I kept an error log. Every problem I got wrong, I wrote down why. Was it a calculation error? A concept gap? A misread question? After 200+ practice problems, patterns emerged. I kept making sign errors in Faraday's law problems (forgot the negative sign consistently), and I kept missing questions about photon polarization because I didn't have the matrix formalism memorized.

A Specific Problem I Encountered

During my second practice exam, I hit a wall on a particular type of quantum mechanics question: particle in a box with a delta function potential. The setup was standard, but the boundary conditions tripped me up every time. I couldn't remember whether to apply continuity of the wavefunction or continuity of its derivative at the delta function location. The workaround: I derived it once, on paper, and kept that derivation in my notes. Specifically, for a delta function potential V(x) = (x-a), the wavefunction is continuous at x=a, but its derivative has a discontinuity: '(a+) - '(a-) = (2m/ℏ²)(a). Memorizing this single result let me solve every variant of this problem in under 90 seconds. Without that derivation, I'd spend 4-5 minutes fumbling through boundary conditions every time. This is the key insight most people miss: you don't need to understand everything deeply. You need to recognize problem types instantly and have the shortcuts ready. The delta function boundary condition is one of maybe ten to fifteen "memorize this result" facts that will save you 10-15 minutes across the entire exam.

Common Pitfalls

Pitfall 1: Over-preparing weak areas. I spent two weeks on statistical mechanics because I found it interesting. Big mistake. The exam gives 10% of questions to thermo/stat mech, and most of those are straightforward. I could have gotten the same score with half the time investment. Focus on high-yield topics first. Pitfall 2: Ignoring the multiple-choice strategy. The GRE Physics test is multiple choice, not multiple select. You can eliminate options quickly in many cases. For example, dimensional analysis eliminates 2-3 choices on calculation problems. Symmetry arguments eliminate options on quantum mechanics questions. I learned to scan for "obviously wrong" answers before doing any math. Pitfall 3: Not practicing with a calculator. The exam allows calculators (specific models listed on the ETS website). I practiced without one and then bombed the calculation-heavy sections on test day because I wasn't comfortable with my device under pressure. Learn your calculator's shortcuts before the exam.

When This Approach Fails

The strategy I described works if you already have a decent physics background (upper-level undergrad or beginning grad student). If you're starting from scratch—like a chemistry major trying to cross-train—the timeline needs to be much longer (4-6 months) and the approach different. You'll need actual courses, not just review books. Also, the pattern-recognition approach has a hard ceiling. If you're aiming for a 900+ (top 1%), you can't just memorize shortcuts. You need genuine depth in several areas. But 900+ is unnecessary for most applications. A 750-800 is competitive for most PhD programs.

What I Wish I'd Known

The exam is harder than it looks on the surface. Questions about Green's functions, partial differential equations, and advanced classical mechanics appear frequently. Don't skip the math methods review. Also, the "additional topics" section (9% of the exam) includes things like nuclear physics, particle physics, and condensed matter. These are mostly memorization—know the standard models, the key experiments, the approximate numbers (proton mass, electron mass, Bohr radius, etc.). You can pick up 10-15 points here with minimal effort. Finally, don't underestimate the mental game. I had one practice exam where I scored 680 because I got anxious on the first ten questions, spent too long on them, and then rushed through the last thirty. The exam is as much about pacing and composure as it is about knowledge.

The third time I took it, I walked in with zero expectations. Just did my best, followed my pacing strategy, and scored 820. Sometimes pressure is the enemy. Good luck.