Forem: Rahul Sukumaran

The Illusion of Understanding: Building Real Systems in an Age of “Fake Thinking”

Rahul Sukumaran — Sat, 18 Apr 2026 07:49:07 +0000

Over my years moving from the IT world at IBM to handling the equity portfolio for a bank, I’ve realized something profound about the intersection of machine learning, trading, and the way we process information.

Recently, I was reflecting on the modern information environment, and it struck me:

We are living through the largest scale production of "fake thinking" in human history.

The Rise of Fast Content

Social media and AI have flooded our world with what we can call fast content:

Pre packaged conclusions
AI generated summaries
Engagement optimized alerts

All designed to give you the illusion of understanding in 30 seconds.

In trading and system building, this is equivalent to blindly trusting a mathematically perfect algorithm's output without asking:

Does this actually give me a strategic edge?

Fast content skips the vital process of wrestling with complexity, leaving you feeling informed but ultimately empty.

More dangerously, when we continuously consume and contribute to this kind of thoughtless content, we poison what can be called the:

Epistemic commons our shared intellectual water supply.

Why AI Cannot Replace the True System Builder

As I build machine learning frameworks to navigate market psychology, I am constantly reminded of what AI cannot do.

AI can:

Mimic structure
Generate content
Build predictive models

But it fundamentally lacks a situated point of view.

A robot cannot truly understand risk or resilience because it does not have direct human experience.

My own expected value framework wasn't just built on textbook math. It was forged through real-world struggles.

AI didn’t:

Experience the fear of a recession-driven layoff
Handle real financial distress situations on the ground
Go through deeply personal loss and responsibility

Every hardship, every moment, every honest conversation shapes perspective.

And perspective is where real systems are born.

Writing as an Act of “Ordering Consciousness”

This is why I believe deeply in documenting this journey through slow content and essays.

Writing is not about packaging knowledge.

It is about:

Thinking
Arguing with yourself
Discovering what isn’t already there

In both markets and life, we operate in chaos:

Noisy data
Constant inputs
Emotional volatility

This leads to:

Anxiety
Fragmentation
Lack of clarity

But when you take that chaos and force it into structure:

Writing a thoughtful essay
Updating trading rules through feedback loops
Refining decision frameworks

You are doing something critical:

You are ordering consciousness.

This deliberate wrestling with complexity is how:

Meaning is created
Clarity is formed
Real trading edges emerge

What the World Actually Needs

The world doesn’t need more:

Fast-food content
Viral outrage
Engagement-driven noise

It needs:

People who choose a positive trajectory
Individuals who deeply learn their craft
Builders who think in public and help others navigate complexity

Not perfection.

But honest thinking.

The Personal Anchor

I’ve had to fight for my path.

From:

Studying relentlessly to secure my first job
Navigating corporate systems while staying honest
Learning to think independently

Through it all, one thing anchored me:

Documenting my thinking.

Closing Thought

There is something humbling about the journey of becoming:

More aware
More thoughtful
More grounded

Thanks for reading.

Question for You

How do you take the chaotic information from:

Markets
Work
Life

…and turn it into meaning?

Would love to hear your thoughts.

The Intelligent Trader: Building a Supervised Learning System to Remove Emotion and Maximize Expected Value.

Rahul Sukumaran — Tue, 07 Apr 2026 13:06:43 +0000

I've watched good traders leave ₹30K, ₹50K, even ₹10L on the table and it's not because they picked the wrong stock. They didn't know when to leave.

The entry gets all the attention. But The exit does all the damage.

After 18 years in equity markets, I'm convinced that fixing exit decisions alone without changing your strategy, your position size, or your risk can improve your P&L by 20 to 40%.

The bottleneck is how you've framed the problem.

The Framing Trap

I see traders usually ask: "Will the price go up or down?"

That question is a trap. It's binary. It feeds anxiety. And the moment the price wiggles the wrong way, it triggers panic.

The right question is: "I'm currently holding a profitable swing trade. Should I hold for 5 more days or exit now to protect what I've made?"

These sound similar. They are completely different.

The first question asks you to predict price. The second asks you to evaluate expected return over a fixed window. One invites emotion. The other invites math.

That single reframe is worth more than any indicator you'll ever add to your chart.

What You Actually Need to Track

It's not price or moving averages or any other popular indicatos you use. These three:

Distance from peak profit. How much have you given back from the top? This is your real pain trigger not the entry price.

Momentum decay. Is the move slowing or still accelerating? A trade that was making ₹50K/day and is now making ₹5000/day is already telling you something.

ATR expansion. When volatility spikes on a trade that's been quiet, that's not noise. That's often the exit signal everyone ignores.

Your Errors Are Not Symmetric

Normally decision frameworks treat errors equally a false positive and a false negative are weighted the same.

That's fine for algorithms. But It's wrong for trading.

Overstaying a dying trade costs you money and corrodes your discipline. You start overriding your own rules. You start hoping. Exiting a day early costs you an opportunity painful, but clean.

These are not the same mistake. Your system must treat them differently.

Build in asymmetric penalties. Punish false optimism harder than early exits. That one adjustment changes how your whole system behaves.

The Math Most Traders Don't Run

Say your remaining upside is ₹50K. Your maximum acceptable giveback is ₹30K.

The break even probability for holding is 37.5%.

You don't need 70% conviction to hold. At 40% continuation probability, holding is the mathematically correct choice — even though it feels uncertain.

Many traders exit at 40% confidence. The math says hold.

Structure your exits around this:

Tier	Signal	Action
Hold	Expected return above +2%	Stay. Let it run. Noise is not a reason.
Defend	Expected return 0–2%	Tighten trailing stop to 10–15% below peak profit.
Exit	Expected return below 0%	Leave. Not later. Now.

The tiers remove the decision in the moment — which is exactly when you shouldn't be making one.

The Part Nobody Does

Predicting without learning is just expensive guessing.

After every trade, log four numbers: peak profit reached, actual exit profit, giveback percentage, and what happened 5 days after you left.

Review every 20,30 trades. The pattern will be obvious you'll either see that you're exiting too early consistently, or that you're holding through clear Tier 3 signals and pretending not to see them.

Then adjust your thresholds. Tighten or loosen based on what your own data says not what feels right.

This is the feedback loop. Without it, the rest of the system just repeats its mistakes in a more structured way.

What This Actually Changes

This isn't a trading system in the traditional sense.

It's a decision operating system one where you define what you believe about markets, what errors you're willing to accept, and how you correct over time.

The traders I've watched stay consistently profitable over years weren't smarter. They just stopped improvising at the exit.

Are you tracking what happens in the 5 days after you leave a position? That number will tell you more about your trading than anything on your entry chart.