Funded accounts don't fail on bad picks.
They fail on arithmetic.

Most evaluations are not lost to a wrong read. They are lost to a stop sized for confidence instead of the buffer, a position still open at 17:00, a trade held through CPI, or a contract that was never on the approved list. None of that is prediction. All of it is discipline, and all of it can be checked before the order goes in.

We run a futures research desk. We test mechanical entry rules net of costs and publish what fails, 7 families so far, with the files for every study. What is left is the part of trading that actually decides whether a funded account survives. It is below, free.

1. Size for the losing streak you will meet

Take a trader who wins 40% of the time. Any particular run of ten trades is all losses only 0.6% of the time, so a ten-loss streak feels impossible. It is not. Across a year of daily trading, about 250 trades, the chance of hitting ten in a row somewhere is 45%. Across 500 trades it is 71%.

So ten straight losses is not a pessimistic case. It is the floor. Your risk per trade is your drawdown divided by ten.

The same $450 stop that is disciplined on a 150K account survives three losses on a 25K. Size is a property of the account, not of the setup.

Tiers and drawdowns are TakeProfitTrader's published figures. Other firms differ; the arithmetic does not.

2. The give-back tax

On an account whose drawdown trails your peak balance including open profit, a winner that runs $1,000 and comes back to flat closes at zero — and still spends 22.2% of a 150K buffer. Give back $2,000 and it is 44.4%, on a trade that made nothing.

Taking partial profit on this kind of account is not timidity. It is the arithmetic. Check whether your firm trails intraday or end-of-day — the answer changes this entire section.

3. The rules that end accounts

Flat across the daily halt. Globex stops 17:00–18:00 ET. A funded account must be flat through it. Overnight holds are fine; a hold that is still open at 17:00 is not.

Flat through the news. On TakeProfitTrader PRO and PRO+ accounts you must have no open position and no working order from one minute before to one minute after these releases:

The trap: CPI, payrolls and FOMC bind every product. Moving from NQ to gold to trade around CPI does not escape the rule. TEST accounts may trade news under the firm's published policy. These releases do not print every day — check the calendar for the session you are actually trading.

Trade the approved contract, not the family. These are the ones we see traders reach for:

Rules shown reflect TakeProfitTrader's published policies as we read them on September 16, 2026. Your firm's current rulebook and your account dashboard always govern. Futures Playbook is not affiliated with TakeProfitTrader.

4. Where the range lives

We tested whether any hour of the NQ day has a directional edge. None does. What the hours do differ in is how far price travels — the busiest hour carries 59% more range than the quietest.

The first ninety minutes are where you are paid for attention. Midday is where you pay commissions for noise. A typical full NQ day moves $1,450 over the whole sample and $2,776 since 2020 — that is the number to size against, not your target.

5. Release days are not what you were told

On the 2020+ tape, CPI, PCE and payrolls mornings average a $2,880 NQ move against $2,776 on an ordinary day — about 4% wider. Quiet days run $2,601. Release days are not wild. Their risk is a directional surprise, which is precisely why the blackout exists.

6. Six questions before every order

If any answer is “I don't know,” the answer is no. On a trailing-drawdown account, information you do not have is risk you cannot size.

7. What we tested

Most rules here were tested under a design written down before the result existed: hypothesis, costs, sample and pass mark. Three were not (level breakouts, release days and the last half hour), and the table says so. The pre-registrations, results and code are on the research page. If you have seen one of the refuted rules sold as an edge, this is what a proper test returned.

These results are based on simulated or hypothetical performance results that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under- or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown.

We also tested the thirteen setups most often sold for passing evaluations — opening-range breakouts, ICT Silver Bullet, VWAP fades, EMA crosses, news straddles — and what the account rules do to them. Read the prop-firm strategy test →

The caveat that matters most if you are funded

The pre-FOMC night was measured 16:00 → 09:30 ET. That hold sits straight through the 17:00–18:00 ET daily flat window, so a funded account cannot take it as studied. The prop-legal version enters after the 18:00 reopen. That version was measured too (FOMC nights from 18:00 to 09:30: n = 85, t = 3.61 in the August 2026 study) and is the one the forward test grades, but it has not been confirmed out of sample. On the plain overnight baseline an 18:00 entry averages $59.19 a night against $74.36 for the 16:00 entry; they are not interchangeable.

We would rather tell you that than sell you a rule you cannot legally hold.

In the terminal

Members get this as a working desk: a pre-trade gate that checks the approved list, the flat window, the day's news blackouts and your size before you place the order — and says no when it does not know. See the membership →

Education about risk, rules and evidence — not a recommendation to buy or sell any contract, and not advice about your account or circumstances. Futures involve substantial risk of loss. Hypothetical results have inherent limitations. Read the full risk disclosure.