How to Evaluate an Investment Strategy Before You Trust It

Any strategy can be described well. Very few can be checked. The difference between the two is the whole of this article.

Evaluating an investment strategy is not a matter of reading a convincing pitch and deciding whether you believe it. Belief is the thing being sold to you. What you want instead is a method you can run yourself, on evidence the seller either publishes or does not. That method has five steps, and every step points at a real artifact – a log, a fill, a dated column – rather than an assertion.

Here is how to evaluate an investment strategy before you trust it, and what to look for before you commit a single dollar.

Start With the Record, Not the Pitch

Skip the performance summary on the landing page. It is the most edited document a manager will ever show you. Go instead to the raw record and ask one question: can I see the reasoning behind each decision, in the order it was made?

A pitch tells you what a strategy did. A record tells you why it did it, when it knew what, and what it gave up. Only the second one is auditable. If the record does not exist, the pitch is unfalsifiable – and an unfalsifiable claim is not evidence, it is marketing.

Step 1: Demand the Decision Log

Ask for the decision log. This is the first and most important artifact.

A trading decision log is the dated account of every decision a strategy made: what was bought or sold, at what size, on what reasoning, and who disagreed. It is not a performance chart. It is the ruled column behind the chart – the entries that produced the number.

What to check in it:

  • Is every entry dated, with the date the decision was made rather than the date it was published?
  • Does each entry state the reasoning at the time, not a rationale written afterward to fit the outcome?
  • Is dissent recorded? A committee that never disagrees is not a committee, it is a rubber stamp.
  • Are entries published on a fixed schedule, or only when they look good?

If a strategy cannot show you a decision log, you cannot evaluate it. You can only take its word. That is a categorically weaker position, and you should price it accordingly.

Step 2: Check Whether Losses Are Published on the Same Schedule as Wins

This is the fastest filter that exists, and almost nothing survives it.

Ask a simple question: when this strategy loses, do you hear about it on the same cadence you hear about the wins? Or does the losing position appear only after it has been quietly closed, while every winner is announced the day it is opened?

Publishing investment losses on the same schedule as wins is the operational proof that a track record is complete. A record with only the wins in it is not a track record; it is a highlight reel. The math is unforgiving here. A strategy that hides its losers is not reporting a lower loss – it is misreporting its risk, and any number built on that record is decoration.

Look for a drawdown column that sits next to the returns column, running the whole time. Look for the losing quarters published in the quarter they happened. If the disclosure only exists in the aggregate, and only after the fact, treat the headline number as unverified.

Step 3: Date Every Number

A number without a date is a rumor.

Investing track record transparency lives or dies on as-of dates. Every figure a strategy shows you should carry the date it was true. “Up 40% since inception” means nothing without knowing when inception was and what happened between then and now. A return reported without its window, its benchmark, and its drawdown is not information.

Demand the as-of date on:

  • performance figures
  • position sizing and current holdings
  • any cash or allocation percentage
  • any claim about “since we started”

A manager who dates every number is telling you that the numbers can be checked. One who does not is telling you something too, if you are listening.

Step 4: Separate Process From Stock Tips

Process versus stock tips is the real fork in the road, and it is easy to spot once you know where to look.

A tip tells you what to buy. A process tells you how the decision was reached, and lets you watch the same method run again on a different name. Tips are unfalsifiable because there is no method to test – only an outcome to hope for. Process is falsifiable, which is exactly why it is worth more.

Here is the test. Ask the person or product to describe the method behind a past decision in enough detail that you could have predicted the next one. If they can, you are looking at a process. If the answer is “we just knew,” or “our intuition,” or a story told only after the fact, you are looking at a tip in a nicer suit.

Quantitative investing methodology is simply process made explicit: rules stated in advance, applied consistently, and logged where you can see them. You do not need to run the math yourself to benefit from it. You only need the rules to have existed before the outcome.

Step 5: Read the Disclosure as Part of the Method

Treat the legal disclosure as a step in the audit, not fine print to scroll past.

A serious strategy states its limitations plainly: what it does not do, what it cannot promise, what could go wrong, and who is responsible for what. That is not a hedge. It is the boundary of the claim, and without it the claim is unbounded – which is another way of saying unverified.

Ask whether the disclosure is specific. “Past performance does not guarantee future results” on its own is boilerplate. A strategy that tells you where its method is weak, and dates its own numbers, has told you more about its honesty than any testimonial could.

What to Look For Before Trusting an Investment Strategy

Run the five steps as a checklist:

  1. Is there a decision log, dated and reasoned, published on a fixed schedule?
  2. Are losses published on the same schedule as wins, with drawdown shown beside returns?
  3. Does every number carry an as-of date, a window, and a benchmark?
  4. Is the method falsifiable – could you predict the next decision from the stated rules?
  5. Is the disclosure specific about limits, or is it boilerplate?

A strategy that passes all five is not guaranteed to make money. Nothing is. But it is falsifiable, it is checkable, and it can be judged on evidence rather than on charm. That is the only honest standard there is.

The point of the audit is not to find a strategy that never loses. It is to find one that shows you the losses, dates its numbers, and lets you watch the method run. You do not have to trust a pitch that survives all five. You only have to read the record. Trust the process that can be checked, and you will never again have to trust a person who cannot be.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *