What score do you actually need to pass the POC? Trying to do the math here

Following this because I'm in the same boat, exam booked for early next month. The 70% number matches everything I've seen too, though someone in another thread said not every question is scored — apparently a chunk are unscored pretest items mixed in, so obsessing over the exact raw count might not even be the right math. Wish they'd just publish it plainly.

For those of you who've taken it — how brutal is the OOP section really? I can write classes all day at work, but the practice questions keep hitting me with MRO order in diamond inheritance, when __new__ runs vs __init__, and classmethod vs staticmethod edge cases I'd honestly never thought about. I got wrecked by a super() question yesterday. Twice.

Trying to figure out if I should sink my remaining two weeks into that or spread it around. The data structures stuff feels fine, comprehensions and file handling are fine. It's just the deep OOP internals that scare me. Anyone who passed: did you actually need that level of detail, or is the real thing more practical than the practice material suggests?

Your math is right -- 28 out of 40. But honestly the score stopped stressing me out once I changed how I was studying. Instead of just drilling questions and memorizing which answer was correct, I started asking myself why the other three were wrong. That shift was huge for me.

Like if I missed a question, I didn't just note the right answer and move on. I'd actually think through each wrong choice and figure out what made it wrong -- wrong timeframe, wrong person responsible, wrong step in the process, whatever. You end up understanding the material way deeper, and the exam can't really trick you anymore because you've already seen all the ways something can be wrong. Passed with a comfortable margin and I think that's why.

I'm in the same boat, exam booked for end of the month so I've been digging into this too. Everything I've found says 70% as well, so yeah, 28 out of 40 checks out. What threw me off is that some of the older forum posts mention a 65% cutoff, but I'm pretty sure that was a previous version of the exam. Wouldn't count on it.

Question for anyone who's actually taken it though — how heavy is it on the "what does this code print" style questions? I can write Python fine but the trace-the-output ones are killing me in practice, especially anything with negative step slicing or mutable default arguments. Like I know def f(x, lst=[]) is a trap, I just keep falling into it anyway when they nest it inside something else. Is it worth grinding those specifically or are they only a handful of the 40?

Also curious if partial topics like decorators and generators actually show up or if it stays mostly fundamentals. Twelve days is enough time if I know where to aim.

Yeah, 28/40 is right — that's the cutoff. Pretty sure that's been consistent for a while now. The tricky part isn't the math, it's that the exam doesn't weight all topics evenly, so you can't just grind through random questions and hope for the best.

What actually moved the needle for me was using a poc practice test about two weeks out. I was doing okay on the core Python syntax stuff but kept fumbling the OOP and exception handling questions. The practice tests broke down my results by topic, so I could see exactly where I was bleeding points instead of just getting a vague "you got 24/40, try again." Once I knew OOP was my weak spot I focused there for a few days and my score jumped pretty noticeably on the next attempt.

12 days is enough time if you're targeted about it. Don't just review everything — figure out which two or three areas are costing you the most points and hammer those. Good luck.

I failed my first attempt by two questions, which was brutal. Looking back, I was way too focused on memorizing definitions and not enough on actually understanding how the concepts connect. The second time around I stopped trying to cram everything and just drilled the stuff I kept getting wrong. That shift made a huge difference.

Your math is right by the way, 28 out of 40. But don't just aim for 28. Give yourself a buffer because the wording on some questions is tricky and you'll second-guess yourself on at least a few. I'd say if you're hitting 32-33 consistently on practice tests you're probably ready. Good luck, you've got this.

Your math is right -- 28/40 is the cutoff. But honestly the score stopped stressing me out once I changed how I studied. Instead of drilling answers, I started asking myself why each wrong choice was wrong. Like not just "B is correct" but "A is wrong because it assumes X, C is wrong because it confuses Y with Z." That shift made everything stick way better.

When you actually understand the logic behind the distractors, you stop second-guessing yourself on test day. It's slower at first but you'll notice you're not just recognizing answers anymore -- you're actually reasoning through them. With 12 days you've got plenty of time if you focus on that instead of just running through practice questions on autopilot.

Your math is right, 28/40 to pass. What actually moved the needle for me wasn't just grinding practice questions though. I started forcing myself to figure out why each wrong answer was wrong, not just why the right one was right. Sounds tedious but it's the difference between recognizing a pattern and actually understanding it.

Especially for the technical sections, like poc/questions/concurrency multithreading, the distractors are designed to look almost correct. Once I understood the logic behind why those traps were traps, I stopped second-guessing myself on the real thing. 12 days is enough time if you're studying that way instead of just flipping through answer keys.

Just hit 74% on my last full practice run so I'm feeling cautiously okay about it. I've been grinding the weaker areas all week, especially poc/questions/concurrency multithreading because that section kept wrecking me. Sitting the real thing on Friday.

And yeah, 28/40 is the math. That's what I've been working backward from too. It's tight but not impossible if you're consistently clearing 70% in practice. Good luck to both of us I guess.

Your math checks out — 28/40 is the threshold. But honestly the score stopped stressing me out once I shifted how I was studying. Instead of just drilling questions and memorizing the right answer, I started asking myself why each wrong choice was wrong. Like if you're on a question about locks and you pick the wrong one, don't just move on — figure out what scenario that wrong answer would apply to. That mental model clicked everything into place for me, especially for the tricky stuff. The poc/questions/concurrency multithreading section was where I noticed it the most, because those questions are designed to trap you with answers that are almost right.

With 12 days left you've got enough time if you're intentional about it. Don't skim questions you get right either — sometimes you got lucky and didn't actually understand it. Go back and make sure you can explain the wrong answers too. That's what separates people who pass with room to spare from people who squeak by and aren't really sure how.