A year ago a reader wrote me: “I want to start algo trading. Where to begin — TSLab or straight to Python?”

I answered: “Depends on your goals.”

He wrote: “I want to trade with robots. I can’t program. But I’m willing to learn if it gives advantages.”

A classic beginner request. And a classic choice paralysis.

Today, after a year of testing five builders and transitioning to code, I have a clear answer.

Three Paths into Algo Trading

There are three ways to start:

Path 1: Visual Builders

Assemble a strategy from blocks. No code. TSLab, StockSharp Designer, NinjaTrader Strategy Builder.

Pros:

  • Quick start (first strategy in an hour)
  • No programming knowledge needed
  • Visual clarity (see connections between blocks)

Cons:

  • Platform limitations
  • Vendor lock-in
  • License fees (TSLab — 60 thousand rubles per year)

Path 2: Straight to Programming

Learn Python/C#, write code from scratch. Backtrader, LEAN, custom scripts.

Pros:

  • Full control
  • Free
  • Platform independence

Cons:

  • Steep learning curve (3-6 months to first robot)
  • Need motivation to learn code

Path 3: Hybrid (Builder → Code)

Start with a builder, then transition to programming.

Pros:

  • Understand algorithm logic without distraction by syntax
  • Soft transition (builder teaches algorithmic thinking)
  • Can verify ideas quickly before learning code

Cons:

  • Spend time on two ecosystems
  • Pay for builder even though you’ll eventually move to code

Who Suits Each Path

Path 1 (Builders) — if:

  1. You’re a trader with manual trading experience — You have a strategy that works, you want to automate it, you don’t plan to dive into complex logic.

  2. You want to quickly test an idea — You need a prototype over the weekend, scalability doesn’t matter, willing to pay for convenience.

  3. Programming causes aversion — You tried learning code and it didn’t click, willing to accept limitations for simplicity.

Path 2 (Straight to code) — if:

  1. You have basic programming skills — You’re a programmer/analyst/data scientist, know at least one language.

  2. You need complex logic — ML strategies, statistical arbitrage, custom indicators, external API integration.

  3. You plan to seriously pursue algo trading — This isn’t a month-long hobby, willing to invest 3-6 months in learning.

Path 3 (Hybrid) — if:

  1. You’re a beginner in both algo trading AND programming — Don’t know if algo trading is for you, want to understand logic without syntax.

  2. You want a smooth entry — Builder gives quick feedback, then easier to transition to code.

  3. You have a learning budget — Willing to pay for TSLab/NinjaTrader for the first months, then move to free Python/C#.

Traps of Each Path

Builder trap: Illusion of simplicity.

You assembled a robot in an hour. It trades. You think: “Algo trading is easy!” Problem: you don’t understand why the robot does what it does. When the strategy stops working, you won’t be able to fix it.

Straight-to-code trap: Information overload.

You downloaded 10 Python courses, subscribed to 20 YouTube channels. Result: analysis paralysis. You study everything but do nothing. Solution: One course. One goal. “Write an SMA cross on Backtrader in a month.”

Hybrid path trap: Getting stuck in the builder.

You started with TSLab. Built 5 strategies. Paying 60 thousand a year. Keep thinking “should learn Python…” A year passes. You’re still in TSLab. Solution: Set a deadline. “3 months in the builder, then Python. Regardless of results.”

Conclusions

Should you start with builders?

Yes — if: You’re a trader with experience wanting to automate a working strategy, the strategy is simple, programming isn’t your thing, willing to pay for convenience.

No — if: You have basic programming skills, the strategy requires complex logic, you plan to seriously pursue algo trading for years.

Hybrid path (builder → code) — if: You’re a beginner in everything, want a smooth entry, willing to spend a month on a builder then learn code.

My personal opinion:

If in doubt — start with a builder. StockSharp Designer is free. fxDreema works in the browser.

Spend a month. Build an SMA cross. Run on demo.

If you liked it — learn Python. If not — you lost nothing.

Better to try and realize it’s not for you than spend a year wondering whether to start.

Useful links:

Platforms for beginners:

Sources: