Paper-Trade Protocol Before Live Size

[NYC88]

Paper trading lies if you let it. Constrain it until it hurts a little.

I papered calm middays for a month once and called myself ready. First live open disagreed. Protocol below exists so that doesn't happen twice.

### Protocol

1. **Same hours** as live — including the open.
2. **Same hotkeys / platform** path.
3. **Realistic size** you actually intend to start with.
4. **Forced slip** — add X cents adverse on every fill in the journal even if sim fills perfect.
5. **20 sessions** minimum across mixed regimes (trend, chop, event).
6. **Pass criteria written in advance** (process adherence ≥ 90%, not "felt good").

### Fail criteria (examples)

Code: Select all

- Skipped journal >2 times
- Broke daily DD rule even on paper
- Changed rules mid-week without version bump
- Only papered calm midday and skipped opens

### Promotion

Paper pass → live **micro** → normal size only after live micro matches paper **process** stats (not PnL fantasy).

If you cannot obey rules on paper, automation will not redeem you. Neither will hope.

What forced slip (¢) do you add to paper fills?
How many mixed-regime sessions before you promote — 20, or do you cheat early?
Anyone failed their own paper pass criteria and still went live? How'd that go?
Process adherence % — do you track it, or only PnL?

[Fairman]

Paper lies if you let it. Constrain it until it hurts a little.

Protocol I respect before live size on stocks/ETFs:

- Same hours as live, including the open.
- Same hotkeys and platform path.
- Realistic starting size you actually intend to use.
- Forced adverse slip in the journal even when the sim fills perfect.
- At least 20 sessions across mixed regimes (trend, chop, event).
- Pass criteria written in advance: process adherence ≥ 90%, not "felt good."

Fail criteria:

- Skipped journal more than twice.
- Broke daily DD even on paper.
- Changed rules mid-week without a version bump.
- Only tested calm middays.

I papered calm middays for a month once and called myself ready. The first live open disagreed. The open is a different animal: spread, speed, and nerves. If your paper protocol never touched 09:30–09:45 ET, you papered a different job.

A worked example of the forced-slip rule. At 09:36 ET the sim fills my buy of 500 shares at $44.10, exactly my limit, on a name where the live spread at that minute is usually two cents and moving. In the journal I record the fill at $44.12, not $44.10. On the exit, sim says $44.22, journal says $44.20. The sim trade made 12 cents a share, $60. The journal trade made 8 cents, $40. Over twenty sessions that adjustment is the difference between a protocol that looks ready and one that is ready. I'd rather find out on paper that the edge is thin after slip than find out live.

If you're scripting this, which is the point of this board, the slip can be a fixed parameter per time bucket applied to every sim fill: something like 2¢ per side at the open and 1¢ midday. Pick the values from your own live fill log if you have one, and keep them conservative if you don't. Version the parameter with the rest of the rules.

Failure modes:

- Quietly dropping the slip adjustment when a week goes badly.
- Using a larger account balance on paper than you'll fund live, which changes how size feels.
- Counting "I would have skipped that" trades as skips after the fact.

When to extend the protocol instead of going live: adherence under 90%, any daily DD breach, or fewer than a handful of sessions that include a genuinely fast open. Another two weeks of paper is cheap.

Diagram: MUST vs FAIL columns.