Seasonality in Bursa Stocks: How to Read Monthly Patterns

Last reviewed: 28 October 2026 · Author: Finworld Research Team · Prepared by the Finworld team, who have passed SIDC examination modules 6 (Stock Market and Securities Law) and 7 (Financial Statement Analysis and Asset Valuation)

Seasonality analysis counts how often a stock or index rose in each calendar month across many years. The result is usually shown as an average return and a probability of an up period. It is useful context, but it is purely historical. It cannot know that this year has a new government policy, a commodity price shock or a different interest rate backdrop.

A pattern with a clear cause is more credible than one without. If a stock has risen every August for eight years and nobody can say why, treat it with suspicion. Humans find patterns in random data very easily.

Step 2: Collect enough history. Aim for ten or more years of monthly returns. Check the sample size shown, because a pattern from five observations is weak.

Step 3: Read both numbers. Look at the average return and the probability of an up period. A high average with a low probability means one or two big years carry the number. A moderate average with a high probability is more consistent.

Step 4: Look at the spread. Note the best and worst years for the month. A wide range means the average hides large swings.

Step 5: Look for a cause. Match the pattern to a real calendar event, such as results timing or a seasonal business cycle.

Step 6: Check the current context. Compare the trend, sector direction, valuation and recent news with the historical pattern. A favourable season in a falling trend carries less weight.

Step 7: Combine with other evidence. Use seasonality as a tiebreaker. If fundamentals, trend and volume are supportive and the month is historically favourable, that adds a small amount of confidence.

Step 8: Keep position size sensible. Because probability is not certainty, never size a position on the strength of a seasonal pattern alone.

January looks attractive: 80% of years were up, and the worst year lost only 3.5%. November is moderate at 70%. December averages almost nothing and has a 50% chance, so it has no useful pattern. February tilts negative at 40% up.

Now consider caveats. Ten data points per month is a small sample. A 70% reading is seven positives and three negatives, and two more bad years would move it to 58%. In addition, if Company A’s January strength reflects festive demand carrying through into results, the cause is plausible. But if this year’s festive sales were weak, as shown by a poor quarterly report, the historical pattern deserves less weight. The sensible way to use the table is as a prompt for questions: why was January strong, and is that cause still present? It is not a signal to buy in December.

• At least ten years of monthly data.
• Both average return and probability of an up period reviewed.
• Best and worst years noted.
• A plausible cause for the pattern identified.
• Index, sector and stock patterns compared.
• Current trend and fundamentals consistent with the pattern.
• Seasonality used as supporting evidence only.
• Position size unaffected by a seasonal view alone.

• Treating probability as certainty. Seventy per cent still fails three years in ten.
• Data mining. Scanning hundreds of stocks will always produce a few perfect-looking patterns by chance.
• Ignoring regime changes. A pattern formed before a major change in the business or the market may no longer apply.
• Overlooking costs. Brokerage, stamp duty and clearing fees reduce small seasonal gains.
• Forgetting dividends and entitlement dates. A price drop on the ex-date is not a seasonal weakness.
• Using it alone. A favourable month does not make a weak stock a good entry.