CCommons

Taking apart my last mock: where 211 marks actually went

A worked example with made-up-looking numbers that are, sadly, completely real for me.

Nisha Verma12th Sample

21 Sept 2026 Β· 5 min read

Photo: Cweveng Β· CC BY-SA 3.0 Β· via Wikimedia Commons

My last mock was 509 out of 720. That's a decent number. It also means I lost 211 marks, and I'd like to show where, because the loss is the only part of the paper I can do anything about.

I'm putting my full analysis here because when I asked people how to analyse a mock they said "look at your weak areas", which, thanks, is exactly what I wanted to figure out. So here's how I do it, with my own numbers from last Saturday. (Class 12, boards in a few months, so this is a quick version.)

Step 1: Subject-wise snapshot

The paper is 180 questions: Physics 45, Chemistry 45, Biology 90. Marking is +4 for correct, -1 for wrong, 0 for unattempted.

SubjectAttemptedCorrectWrongMarks
Physics402812112 - 12 = 100
Chemistry43349136 - 9 = 127
Biology887414296 - 14 = 282
Total17113635544 - 35 = 509

Quick sanity check I always do: marks = 4 Γ— correct - wrong. 4 Γ— 136 = 544, minus 35 = 509. If the sum doesn't match, I mis-copied something.

Biology is at 282 out of 360, so about 78 percent, Chemistry about 71 percent of its 180, and Physics 100 out of 180, so 56 percent. I don't need to tell you where the problem is. Physics is the weak one, as always. But the number "56 percent" is only the headline. It doesn't tell me why.

Step 2: Where the 211 went

Each wrong answer costs me 5 marks relative to getting it right (I lose the +4 I could have had and get -1). Each unattempted question costs 4. So:

  • 35 wrong x 5 = 175 marks
  • 9 unattempted x 4 = 36 marks
  • 175 + 36 = 211 βœ“

That's a useful reframe. My 35 wrong answers cost far more than my 9 skipped ones. If I could convert even a third of the wrong ones, I'd gain about 58 marks, which is more than all the skipped questions combined.

Step 3: Sort the 35 wrong answers into types

I go through them one by one with the paper beside me and put each into a category. Mine were:

TypeCountMarks lost (x5)
Concept gap (didn't know the idea)945
Silly (calculation, misread, sign)1260
Guess (no idea, picked one)840
Time pressure (rushed the last 20)630
Total35175

Look at what that says. The thing I'd always blamed, "weak concepts", is only 45 marks. The silly ones are more than that, 60 marks. If I had said "I need to study more Physics" I would've gone and done three more hours of Physics theory. What I actually need is a habit: read the question twice, write units, check the sign.

Step 4: Pick the leaks to fix

I do not try to fix everything. I pick the two biggest and write a rule for each.

Silly errors (60 marks). Of the 12, seven were in Physics numericals: three unit slips (cm vs m), two sign errors in potential/energy, two where I used the wrong value. Rule for the next two weeks: after every numerical, spend 10 seconds writing the unit of the answer in my head and the sign check. Not more time on each question, just a fixed 10-second closing ritual.

Guesses (40 marks). Eight guesses, five of them wrong enough to hurt. Here's the thing about guessing in a paper with negative marking. If you have no clue between 4 options, the expected gain from a guess is:

14(+4)+34(βˆ’1)=+0.25\tfrac{1}{4}(+4) + \tfrac{3}{4}(-1) = +0.25

That's basically nothing. But if I can honestly rule out two options, it becomes:

12(+4)+12(βˆ’1)=+1.5\tfrac{1}{2}(+4) + \tfrac{1}{2}(-1) = +1.5

So my rule is: guess only if I can kill at least two options for a reason I can say in one line. Otherwise leave it. I'd rather lose 0 than gamble for 0.25.

Step 5: Concept gaps, tiny list

Nine questions, but when I grouped them there were only three chapters involved: two in Physics (a modern physics one and a rotational motion one), one in Chemistry (a coordination compounds one). So the "study plan" for the week isn't "do Physics". It's three specific chapters.

Step 6: What I'm NOT doing

I'm not comparing this to anyone's score. I'm not predicting a rank from it. One paper has its own difficulty level and its own bad luck. I'm treating the 509 as a reading on a thermometer, not a verdict on my future.

I'm also not retaking the same paper in the next two weeks. That'd just be memory.

What I'd tell you to copy

If you're doing this for the first time, it takes me about 90 minutes to do properly. The order is: count, compute the loss, sort wrong answers by type, pick two leaks, write one rule each, list the specific chapters. Write the rules on a sticky note and put it on the desk next to the pen for the next paper. In the next mock, I check whether the sticky note rules were followed first, before looking at the score. If I followed them, a drop in marks is variance. If I didn't, that's the real problem.

Last thing. 211 marks sounds scary. But 175 of those are wrong answers on questions I attempted, which means I was close. Every one of those was a question where I sat and thought I knew something. That's much easier to fix than not knowing the syllabus. That's why I'm weirdly cheerful about this mock.

WhatsAppPost on X

11 responses

Sign in to join the conversation.

  • [+]Kabir Joshi11thSampleΒ· 28d ago

    "Marks = 4 Γ— correct - wrong" as a sanity check is nice and nerdy. I'm going to make a tiny script that takes attempted/correct and spits out the loss breakdown. Not for me yet, but for fun.

  • [+]Siddharth RaoDropperSampleΒ· 28d ago

    The reframe that 35 wrong answers cost way more than the 9 skipped is the right one. I used to chase the unattempted ones and push attempts up. Wrong move. Accuracy first, attempts later.

  • [+]Harsh Gupta12thSampleΒ· 28d ago

    guess expected value of +0.25 is basically a lottery ticket with extra steps 😭 but ok i get the "kill 2 options" rule. also 12 silly errors x 4 = 48 marks, that's like nothing, why the drama

    • [+]Dev Agarwal12thSampleΒ· 28d ago

      Simple version: random guess = tiny profit, 2 options gone = decent profit. But the variance is the thing nobody mentions. +1.5 is an average, any single question can still go -1. Do it a lot and it works out, do it once and it's a coin.

  • [+]Vivek Mishra12thSampleΒ· 28d ago

    Thank you for sharing real numbers. Many people share only the good score. I also lose most marks in Physics silly errors. I will use the 10-second ritual. I think my units mistake is from rushing.

  • [+]Mehak AroraDropperSampleΒ· 28d ago

    the part about not predicting a rank from one paper... I needed to read that tonight. I'm still opening my last scorecard like it owes me money.

    • [+]Nisha Verma12thSampleΒ· 28d agoAuthor

      we'll both survive it. Close the scorecard until tomorrow πŸ™‚

  • [+]Zoya Khan12thSampleΒ· 27d ago

    What I like most is that the analysis takes you from a feeling ("I did badly") to a list with numbers. Anxiety is mostly vagueness, and a list takes some of the vagueness out of it. I also think the cheerful ending is not fake, if the errors are mostly on attempted questions it does mean you are close.

  • [+]Ishita GoyalDropperSampleΒ· 27d ago

    Can I add: the silly errors are 12 x 5 = 60 marks. Someone above said 12 x 4 is 48 but each fixed silly error saves the -1 as well as getting the +4, so it's 5 each. That's why I tell everyone these are the cheapest marks to win back.

    • [+]Harsh Gupta12thSampleΒ· 27d ago

      ok that was me, I did 12 x 4. fine fine, 60 it is. even cheaper marks, even sadder me

  • [+]Neel Chatterjee12thSampleΒ· 27d ago

    Nice that you separated "time pressure" from silly. In a physics paper they feel the same in the moment but need totally different fixes. One needs a pacing rule, the other needs a closing ritual.

More from Nisha Verma