Plaintape LLC is registered with the CFTC as a Commodity Trading Advisor and is an NFA Member (NFA ID 0580378). The firm is not accepting client accounts. Registration and NFA membership do not imply that the CFTC or NFA has approved or endorsed the firm, its trading program, or the material on this site.
A live trading record documents decisions that reached an actual account and the results recorded there. A backtest evaluates a specified rule on historical observations under stated assumptions. Both can answer useful questions. Neither proves that future trading will reproduce the sample, and a short live record may contain very little evidence about rare losses.
A carefully constructed historical test can examine more market conditions than a young live account has encountered. It can show how a rule responds to a changing input, expose a sizing error and compare a decision with a deliberately simple alternative. Repeating the same test after a code change can also expose a regression.
That makes a backtest a useful research and engineering tool. Its scope must remain explicit. A test of one historical path cannot establish all possible loss paths, and success under assumed fills cannot establish what a broker would have executed.
A live account pays actual commissions and encounters actual order handling. Its fills contain information about execution prices and timing. Statements contain financing, interest and currency effects included in the broker's account values. Those observations can challenge a simulation's assumptions about trading costs and position changes.
A proprietary account still describes its owner's capital and decisions. Its results do not automatically describe client accounts with different fees, sizes, permissions or available instruments. The track-record evaluation guide explains those differences.
The revised futures-roll operations note uses published historical failures to examine that boundary.
Consider a signal calculated from yesterday's prices. A historical test can calculate a target position from that input. In production, the input may arrive late, the intended contract may be unavailable, an order may remain partially filled, or a restart may occur between submission and acknowledgement. These are conceptual failure cases. This page does not claim that every case occurred in Plaintape or assign a performance effect to one.
The evidence questions become concrete: which price version was used, which position was requested, what the broker accepted, what actually filled, and how the account was reconciled afterwards? A strategy calculation, an order log and a broker statement describe different stages of that chain.
Plaintape's published methodology records a cash-flow-handling correction. Earlier treatment of transfer days removed trading results on those days as well as the external flow. It also identifies limits in the historical account-value record. These are reasons to retain versions and restatements alongside the live tape.
The integrity page helps explain how linked records make later edits detectable. It does not turn an operator's record into an independent broker audit. The methodology describes the return source and the separation between forecasts, positions and execution.
Ask whether the rule, instrument set and start date were chosen after the results were seen. Ask whether the data used information that was unavailable at the time of a simulated decision. Ask whether costs and contract transitions were modeled consistently. Keep unsuccessful research variants visible in the research record when judging how much searching produced the final candidate.
An out-of-sample test uses observations withheld from choosing the method. Repeatedly inspecting that sample can gradually turn it into another selection input. Calling a sample "out of sample" does not establish that it stayed untouched.
A position can be sized using an estimate of volatility, the expected variation in returns. Actual variation can exceed the estimate. Prices can gap, relationships between markets can change and a position reduction can take time to execute. These limits apply to both a research forecast and a live risk-control calculation. No historical risk statistic is a promise about the next loss.
Use historical tests to examine assumptions and behavior outside the short live sample. Use the live proprietary tape to examine what was actually recorded and corrected. Keep their dates, account basis, costs and limitations visible. No simulated performance figures or curves are presented here.
For ownership and trading authority, read how a managed futures SMA works. Plaintape is not accepting client accounts. Futures trading involves substantial risk. PAST PERFORMANCE IS NOT NECESSARILY INDICATIVE OF FUTURE RESULTS. This is educational material, not individual investment advice.
No hypothetical performance figures are presented on this page. The prescribed statement explains the limitations discussed above.
HYPOTHETICAL OR SIMULATED PERFORMANCE RESULTS HAVE CERTAIN LIMITATIONS. UNLIKE AN ACTUAL PERFORMANCE RECORD, SIMULATED RESULTS DO NOT REPRESENT ACTUAL TRADING. ALSO, SINCE THE TRADES HAVE NOT BEEN EXECUTED, THE RESULTS MAY HAVE UNDER-OR-OVER COMPENSATED FOR THE IMPACT, IF ANY, OF CERTAIN MARKET FACTORS, SUCH AS LACK OF LIQUIDITY. SIMULATED 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 PROFIT OR LOSSES SIMILAR TO THOSE SHOWN.