The LeetCode Trap: Why 500 Problems Got Me Nowhere

Day 2. Let's talk about LeetCode.

I have nothing against LeetCode. It's an incredible resource. 3000+ problems, great test cases, solid community discussions. I learned a lot there.

But I also wasted a lot of time there.

The Trap

Here's how I used to study:

1. Open LeetCode

2. Pick a problem (usually sorted by "most frequent")

3. Try for 10 minutes

4. Get stuck

5. Look at the solution

6. Think "oh that makes sense"

7. Move to the next problem

8. Repeat

I called this "studying." I tracked my progress in a spreadsheet. I felt productive.

But here's the thing: I wasn't actually learning. I was just exposing myself to solutions without internalizing them.

The Numbers Don't Lie

I went back and tested myself on problems I had "solved" 3 months ago.

Out of 50 problems I tested:

That's a 16% true retention rate.

All those hours grinding, and I retained less than 1 in 5 problems.

Why This Happens

When you look at a solution and think "that makes sense," your brain files it under "understood." But understanding is not the same as knowing.

To actually KNOW something, you need to:

1. Struggle with it first (this creates the mental hooks)

2. Derive it yourself (even partially)

3. Review it multiple times over days/weeks

The 10-minute-then-look-at-solution approach does none of these.

What I Do Now

My new process:

30 minutes minimum before any hints

If I can't solve it in 30 minutes, I get a HINT, not the full solution. "Try using a hash map" is a hint. The full code is not.

Explain it out loud

After solving, I explain the approach as if teaching someone. If I stumble, I don't understand it well enough.

Spaced repetition

I review problems on day 1, day 3, and day 7. This is the science of memory - you need repetition at increasing intervals.

Focus on patterns, not problems

Instead of "I can solve Two Sum," I want "I understand when to use hash maps for O(1) lookup."

The 100 Problem Challenge

This is why DSA 100 Days is focused on 100 problems, not 1000.

100 problems, done right, teaches you more than 500 problems done wrong.

The key is depth over breadth. Understanding over exposure.

Tomorrow: What actually stuck after 100 days of this approach.

— Marcus