Most investment strategies are built to be believed. Very few are built to be checked. That difference is the only one that matters to a reader who has already learned the hard way that a confident track record and a verifiable one are not the same object.
This is a working audit. It is not a list of questions to make you feel diligent. Each step points at an artifact you can actually open, read, and date – a logged entry, a published fill, a ruled column. If a strategy cannot produce the artifact, that is your answer. You do not need to disprove it. You only need to notice that nothing was there to check.
Step 1: Ask for the decision log, not the highlight reel
Start with the log. A decision log is the dated record of what was decided, when, and on what basis – every vote, every dissent, every executed order, written down as it happened rather than reconstructed after the fact.
What to demand:
- A timestamp on each entry, not just a month and a year.
- The reasoning recorded at the moment of the decision, not a later explanation fitted to the outcome.
- The full record, including the decisions that went badly.
A highlight reel tells you what a strategy wants you to feel. A log tells you how it thinks. If someone can show you profits but not decisions, you are not looking at a process. You are looking at a result, and results are the one thing that can be manufactured without a method.
Step 2: Demand losses on the same schedule as wins
This is the test most products fail. Ask how losses are published. Not whether they are disclosed somewhere, eventually, in a footnote – how they are published, and on what cadence.
Wins and losses should travel the same road. If a strategy posts every gain within the hour and mentions drawdowns only in a quarterly letter, the schedule is the message. A drawdown is not an embarrassment to be managed; it is a number your decision depends on, and it should carry the same as-of date as everything else.
A record that publishes losses on the same schedule as wins is telling you something structural: the people running it expect to be judged on the whole account, not the flattering rows of it.
Step 3: Check the as-of dates
Every number has a moment. A track record without an as-of date is a rumor with a percentage sign attached.
For any figure you are shown, ask three things: as of when, measured how, and updated how often. “Up 12% since inception” means nothing without the date the pencil stopped. A strategy that commits to as-of dating on its own numbers is different from one that lets the reader assume the most flattering interpretation. One is designed to be audited. The other is designed to be skimmed.
Hold the same standard for the small numbers as the big ones. If the headline is dated and the position table is not, the position table is decoration.
Step 4: Find the dissent
Consensus is cheap. Anyone can agree in public and decide in private. What you want to see is disagreement recorded next to the decision it belongs to.
Ask whether the record shows dissent – a member who argued the other side, the reasons they gave, and what the committee did anyway. A process that logs its own disagreements is doing something no marketing page can fake: it is showing you its reasoning under pressure, before the outcome was known.
Where there is no dissent anywhere in the record, there are two possibilities. Either everyone always agrees, which should worry you, or the dissent was real and went unwritten, which should worry you more.
Step 5: Separate decision inputs from decoration
Most strategies are full of numbers. Few distinguish the ones that drive a decision from the ones that dress a page.
Decision inputs are the figures a position was actually changed by: the entry price, the size, the trigger, the exit, the fill. Decoration is everything designed to be looked at rather than acted on – smoothed equity curves with no drawdown shown, winning percentages with no position sizes, benchmarks chosen after the fact.
Test each number the same way. Ask what decision it would change. If the answer is none, it is decoration, and its job is to make the process feel more rigorous than it is.
Step 6: Trace one fill end to end
Pick a single trade the strategy claims and try to follow it the whole way: the entry in the log, the reasoning behind it, the size, the execution, the exit. Can you walk from the decision to the fill to the result without a gap?
If you can trace one position from vote to outcome, you understand how the whole thing works. If you cannot, ask why – because a verifiable process does not have unverifiable trades. A published fill is the receipt. Everything else is an invoice you wrote to yourself.
What a real record looks like
Put the six steps together and a shape appears. A strategy worth trusting publishes a dated decision log, runs losses and wins on the same schedule, stamps as-of dates on its own numbers, records its dissent, keeps its decision inputs separate from its decoration, and lets you trace a fill from entry to exit.
That is a lot to ask. That is the point.
None of this guarantees a strategy will make money. Nothing does, and anyone who says otherwise is selling something other than a process. What the audit gives you is narrower and more useful: a way to tell whether the thing in front of you is a method you can examine or a story you are being asked to accept. The first can be followed with your eyes open. The second has to be trusted on faith, and faith is not a position you can size.
Start with the log. If there is no log, stop there.
This article is general information about how to evaluate an investment process, not investment advice. It is not a recommendation to buy or sell any security. Diers Capital publishes its public decision log, including losses and dissent, on the same schedule as its wins; past performance does not predict future results, and you should consider your own circumstances or seek independent advice before acting.

Leave a Reply