⚠️ Personal research and trading journal — not investment advice. The author does not provide licensed advisory services. Verify all figures against official SET/SETTRADE filings before acting.
I set out to answer a simple question: over the last 16 years, could any systematic strategy beat the boring act of buying TISCO and reinvesting the dividends?
The answer turned out to be humbling, and the journey to it nearly fooled me with a number that looked too good to be true. That near-miss is the most useful part of the story, so let me start there.
The trap: a backtest that "made" 67% a year
My first run of a momentum strategy on raw Thai price data showed a +67% annual return — a 4,800× gain over 16 years. One year showed +21,000%.
That is, of course, impossible. No long-only equity portfolio does that. So instead of celebrating, I went hunting for the bug — and found two landmines buried in raw Thai price data:
1. Unadjusted reverse splits. When a stock does a 1-for-100 reverse split (price jumps from ฿0.03 to ฿3.00 overnight), raw data reads it as a +9,900% one-day return. A momentum strategy actively chases that fake spike. 2. A dividend-basis mismatch. The price series was already split-adjusted, but the dividend records were not — so an old dividend divided by the adjusted price implied a fake 20%+ yield every period. This alone made PTT look like it returned +26%/year when its real total return was closer to +8%.
The lesson is one every quant learns the hard way: a backtest that looks spectacular is a data error until proven otherwise. Clean the splits, verify the dividends, then believe the number.
The honest result: TISCO is genuinely rare
Once the data was clean, two things were clear.
First, momentum trading didn't beat buying and holding. Over 2010–2026, a diversified momentum system returned about +5%/year — barely matching the index and far below simply owning quality stocks and reinvesting dividends.
Second, TISCO is an exceptional compounder — and hard to replicate. Ranked by risk-adjusted return (Sharpe ratio), TISCO was the single best liquid stock in the Thai market: ~+17%/year total return with surprisingly low volatility. Its secret isn't just a high yield (~7%) — it's the combination of high yield, steady price growth, and a smooth ride. That trio is rare.
The stocks with the closest results (not the same business) were a mix: - Price-growth compounders — KCE, DELTA, AOT, BDMS — that returned 18–25%/year mostly through capital appreciation, with low dividends. - Dividend compounders — TISCO, and to a lesser extent TCAP — that returned via a high, growing, reinvested dividend.
Most of the very best returns came from price growth in quality franchises, not from dividends. But the smoothest compounding — the kind you can actually hold through a decade without panicking — came from the high-yield names.
Where most of the return actually comes from
A striking detail: in Thailand, the dividend often is the return. PTT's share price barely moved in 16 years, but with dividends reinvested it still compounded meaningfully. A strategy that trades in and out — sitting in cash, rotating positions — forfeits that dividend stream. In a high-yield market, that's a structural handicap you hand to the buy-and-hold investor before you even start.
A note on volatility and dollar-cost averaging
It's tempting to think the wild swingers are better for dollar-cost averaging — the dips let you buy cheap. The data half-confirms this: for a 45%-volatility stock like KCE, monthly DCA achieved an average cost 61% below the period's average price. The swings genuinely lower your cost.
But there's a catch. That discount is measured against the average price — and for a stock that rose over time, the starting price was lower still. Buying it all at the start beat dollar-cost averaging in almost every case, because DCA keeps buying higher as the stock climbs.
So the honest rule: volatility helps DCA only if the stock eventually rises. DCA into a volatile quality compounder harvests the swings and rides the trend — a great combination. DCA into a volatile cyclical that never recovers just means you keep buying a loser. The volatility is only your friend if the business is sound.
The takeaway
You cannot easily out-trade a quality dividend compounder in Thailand. The reliable path to ~13–17% a year is to own a basket of quality businesses, reinvest the dividends, and hold — not to time the market. Beating that, to reach 20%+, requires something a screen can't give you: the judgment to identify the next great franchise early. The dividend reinvestment is the engine; the selection is the edge; the trading is mostly a tax.
And whatever you do — clean your data first. The market is hard enough without your spreadsheet lying to you.
Research and educational use only. Figures derived from 2010–2026 Thai price and dividend data, split-adjusted, with dividends cross-checked against public sources where possible. Not a recommendation to buy or sell any security.