Paper Trading Before Live: A Promotion Checklist

Charcoal and teal illustration: a candlestick chart inside a dashed sandbox frame connects through a glowing checklist with a padlock to a closed vault door.

Illustrative only, not a trade signal.

Paper trading feels like the safe part. It's also where a lot of traders learn the wrong lesson.

Not because practice is bad. Because a paper fill is a promise the market never made.

In The Benefits of Paper Trading I wrote about why simulation matters: learning without the burn, building discipline, seeing your own patterns. This post is about the exit. What has to be true before a paper process earns real money?

Call it a promotion, not a graduation party. Promotions are earned with evidence, and they can be reversed.

Education and process only: no trade ideas, no results claims, no broker or tool recommendations. Where I describe a platform, I link the doc page I read.

What paper fills ignore

Dark chart with a teal price line where a hollow gray paper-fill marker and a solid teal live-fill marker sit apart, the gap shaded, next to a panel of stacked order-queue bars with one highlighted.

Illustrative only, not a trade signal.

Here's the uncomfortable part. A simulator grades your paper record, and simulators are generous.

Alpaca's paper trading docs are unusually direct about it. They call paper trading only a simulation, say it's not a substitute for real trading, and warn that performance may differ. Per the page, paper trading does not account for:

Market impact of your orders
Information leakage of your orders
Price slippage due to latency
Order queue position (for non-marketable limit orders)
Price improvement received
Regulatory fees
Dividends

Two more lines on the same page are worth reading twice. Order quantity isn't checked against the quoted size at the best bid and offer, so a paper order can fill far larger than the liquidity that was really there. And eligible orders get a partial fill of random size 10% of the time.

Think about what that means for a limit order. On paper, it fills once price becomes marketable. Live, you may be standing behind a line of other orders at that price, and the market can leave before your turn.

Key takeaway: A paper fill tells you your rules fired. It doesn't tell you the market would have filled you at that price, in that size.

Every simulator makes different assumptions

One order card fans out to three dark panels showing three different fill patterns: a single full bar, three uneven partial bars, and a dot inside a shaded slippage band.

Illustrative only, not a trade signal.

Alpaca, LEAN and Freqtrade all document a paper or dry-run mode. They don't simulate the same market.

Alpaca paper trading

Orders aren't routed to a live exchange. Fills are simulated from real-time quotes and matched against the best available market price (NBBO). The same page notes that different simulators can differ on fill prices, fill assumptions, liquidity assumptions and market data sources.

Read the docs →

QuantConnect paper trading (LEAN)

Real-time data runs into your algorithm, but fills are simulated with fictional capital. Under the default brokerage model, orders don't experience slippage in backtests or paper trading, and market orders fill immediately and completely. The docs also suggest paper trading as a way to check that a backtest wasn't overfit.

Read the docs →

Freqtrade dry-run

Dry-run is on by default, and the docs recommend starting there. In dry-run, market orders fill based on order book volume with a maximum slippage of 5%, a stop loss placed on the exchange is assumed to be filled, and open orders stay open after a restart on the assumption they weren't filled while the bot was offline.

Read the docs →

Same order, three different fills. None of them is the market.

I'm not ranking them. The point is that “it worked on paper” always carries a silent footnote: under this simulator's assumptions. Write those assumptions down before you trust the record.

Key takeaway: Know your simulator's fill model the way you know your playbook. If you can't describe it, you can't discount it.

A written graduation checklist

Glowing teal clipboard checklist with four checked items and two open circles, flanked by a teal graduation cap and a small vault door.

Most traders promote themselves on a feeling. A good week. A streak. Boredom.

A written checklist swaps the feeling for evidence. Agentic Trading: The Foundation already has a go-live mini-checklist for the live gate itself. This one sits in front of it and asks a narrower question: is the paper record honest enough to trust?

Paper-to-live promotion checklist (copy this)

☐ One playbook version ran the whole paper stretch, frozen with a version and date. No mid-run rewrites.
☐ My log shows taken and missed setups, not just the trades I liked, plus a KEEP/CUT pass.
☐ I can explain every paper trade by a written rule, win or lose.
☐ I've written down my simulator's fill assumptions and which documented gaps (slippage, queue position, fees, size) touch how I trade.
☐ My paper size matches the size I'd actually use live, not the platform's default demo balance. (Alpaca's paper account starts at $100k by default; Freqtrade's dry-run wallet defaults to 1,000.)
☐ I've rehearsed failure modes without live money: missed fill, late fill, exit stall, local-versus-broker desync.
☐ I've restarted the system mid-session on paper and checked what happened to open orders and saved state.
☐ Live records will start clean, kept separate from paper records.
☐ The first live step is tiny and date-scoped, and the overnight default stays paper or disarmed.

Notice what's missing: a target number of sessions or a win rate. How much evidence is enough is your call, and it belongs in your playbook, written before the paper run starts. This list only checks whether the evidence is clean.

Key takeaway: Promotion is earned by a clean record, not a good mood.

Review cadence

Dark calendar grid of teal and gray daily dots with the last weekly row highlighted, beside a teal progress ring around a checkmark.

A checklist you check once is a ceremony. The cadence is what keeps it honest.

Here's the rhythm my published posts describe. It works the same on paper as it does live:

After every session

Taken versus missed, fit to the playbook, honest notes on overrides. In Part 2, review happens after the session, separate from live placing.

Weekly

Keep-or-kill from evidence. A rule that keeps failing on contact with reality gets cut, not defended.

Monthly

A KEEP/CUT pass, then one of three decisions from the Foundation's 30-day plan: stay in simulation, stay tiny, or revise the playbook.

Journal the why, not just the what. The Hidden Truth in Your Wins and Losses makes the case that logging entries, exits and P&L is surface level. The real value is why you made the call and who you were when you made it. Paper is where that habit is cheapest to build.

Once you're live, add one line to every review: how did the real fill compare with what paper would have given you? Paper can't produce that comparison on its own.

Key takeaway: Daily for fidelity, weekly for rules, monthly for the promotion decision.

When not to go live

A dashed sandbox chart on the left, a closed teal-striped barrier arm and a pause symbol in the middle, and a closed vault door on the right.

Illustrative only, not a trade signal.

Sometimes the right promotion decision is “not yet.” The Foundation says it plainly: skipping a live day is success if it isn't earned.

Stay on paper if any of these are true:

You changed the playbook during the paper run and haven't re-run it.
Your log is mostly wins, and the misses were never recorded.
You can't say which simulator assumptions flatter your record.
You haven't verified that day-done actually pauses the system.
Your local records and the broker's records disagree, and you haven't reconciled.
You want to go live to make something back, or because paper feels boring.

That last one deserves a beat. The Benefits post warned that paper lacks real emotion. The answer isn't to rush toward the emotion. It's to walk in with rules strong enough to hold when it shows up.

In my own build, paper isn't a phase to escape. Part 2 describes picking paper or live before the open, with overnight and day-done defaulting to paper and disarmed. Live is a choice made each morning, not a status you keep. The Approval Gates checklist covers the controls that sit around that choice.

Key takeaway: “Not yet” is a decision, not a failure. Write it down with the reason, and revisit it next cycle.

Automation and AI: From the site Disclaimer: “Posts about agentic or automated trading describe my own process and experiments. Software and AI tools can fail, lag or act on bad data. Nothing here is a recommendation to automate trading or to use any broker or tool.”

As the Foundation puts it, simulation proves process shape. Live adds fill reality, emotion and broker truth.

The promotion checklist is how you walk from one to the other without lying to yourself.

Robinhood vs Fidelity Options Trading: Which Broker Wins for Day Traders in 2025?

As a day trader guiding small accounts through the volatile world of options trading on Robinhood’s Legend platform, I’ve discovered that the right broker can transform your journey from survival to success. Robinhood and Fidelity are giants in the brokerage space, each offering unique tools for day traders in 2025. With markets shifting and small-account growth becoming my focus, I’ve gravitated toward Robinhood for its simplicity, access to 0-DTE options, and synergy with my streamlined system. In this deep dive, I’ll compare their platforms, fees, and features—blending my experience with current trends—to help you decide which broker aligns with your goals, whether you’re nurturing a small account or scaling as a seasoned trader. Let’s uncover what drives my choice and how it might shape yours. Trading Platforms: Simplicity vs. Sophistication Robinhood: My Go-To for Simplicity and Growth Robinhood has become my preferred platform, especially for growing small ...

Day Trading Journal

Search This Blog