When to Change Your Strategy and When to Sit Tight

Change your strategy when the premise it is built on has changed, or when your own circumstances have changed; sit tight when only the result has changed. This is the only distinction that works. A drawdown is not information in itself: it is built into any strategy that is capable of earning anything at all. Information is when you discover that the instrument has become illiquid, that the tax regime has been rewritten, that the issuer is no longer the business you bought, or that your horizon has shrunk from a decade to next year. What follows is about how to tell these apart before the decision is made on emotion.

A strategy is something that can be proven wrong

Most arguments over "change or sit tight" cannot be settled, because the strategy does not exist in a testable form. "Buy good companies for the long term" cannot be proven wrong: whatever the outcome, you can say that the company turned out to be not so good, or that "the long term" has not arrived yet.

A testable formulation looks different. It states which asset class, what the selection criterion is, what the horizon is, what counts as normal behaviour and what counts as abnormal. For example: I buy securities from the list of stocks with stable operating cash flow, value them by 3,78 and compare that with EV/EBITDA in place of P/E, and the horizon runs until the issuer's debt profile changes. A construction like this can be proven wrong: if the cash flow is no longer stable, the premise is dead and there is nothing left to hold.

As long as a strategy has no condition under which it is proven wrong, the decision to "change" is always taken on the only signal available — how painful it is to look at the account. That is the worst possible input.

Reasons that really are reasons

The premise about the instrument has changed. The most common case is liquidity. A security whose order book rested on the interest of a broad range of participants can turn into a security where the price is set by almost one market maker. A strategy that counted on being able to exit at a reasonable price can no longer be carried out in such a security — not because you were wrong about the business, but because the exit mechanism has disappeared. More on this in the piece on liquidity.

Your horizon or your need for the money has changed. This is not a market event, but this is exactly where a change is needed. A strategy with a long horizon and a strategy with money that will be needed on a known date are different constructions, and they must not be mixed.

The tax or legal framework has changed. The deduction for long-term ownership (Article 219.1 of the Russian Tax Code) and the base on which the tax is calculated (13%) affect what it pays to hold and what it pays to close. The basics are in the piece on personal income tax for investors.

The issuer itself has changed. Not the price, but the business: the debt burden, the revenue mix, the dividend policy. Financial statements are published regularly, and they are worth reading in the reports section rather than learning about the changes from a news headline.

Reasons that are not reasons

A drawdown within the range you named in advance as normal. If the range was not named in advance, that is the problem, not the drawdown.

A split or a reverse split. The number of shares changes; the ownership stake does not. It is a technical operation, and responding to it by buying or selling means trading on a non-event: see splits and consolidations.

The record date and the dividend. After the list of holders is fixed, the price behaves in a predictable way, and the payout calendar — {{dividend_calendar|limit=5}} — is no reason to rebuild the portfolio. Who gets onto the register, and when, is covered separately.

Someone else's return. A strategy that outperforms yours over a short stretch may differ in the risk taken, not in quality. A comparison only makes sense over a horizon comparable to your own.

A single move on the instrument page. Looking at {{instrument:SBER}} is useful; drawing a conclusion about the mechanism from one day is not.

How to change, if change is needed after all

Step 1: write down exactly which premise no longer holds, and what shows it. If you cannot finish the sentence without the words "it seems", it is too early to change.

Step 2: work out the cost of switching. Closing a position realises the financial result, and so it includes tax; moving into a less liquid instrument adds the spread; changing the asset class also changes the holding period from which the deduction is counted. Sometimes the cost of switching eats up the entire advantage of the new construction.

Step 3: change one part of the construction, not everything at once. If you have changed the asset class, the selection rule and the horizon all at the same time, you will never know what worked, and next time you will be making the decision just as blindly.

Step 4: write down the new condition under which the strategy is proven wrong before you enter the position. That is the whole difference between a strategy and a series of reactions.

What you will not find here

The platform has market data, but it does not have your portfolio, your horizon or your obligations — and it is precisely on these that the right answer in a particular case depends. Nor are there statistics here on the comparative returns of different approaches: such estimates depend heavily on the period chosen, and presenting them as a benchmark would be dishonest. This piece offers a distinction, not a recommendation to buy or sell.

Definitions of the terms used in the text are collected in the glossary, and the calendar of corporate events is in the events section.