Mở khóa ghi chú này

A stock hitting a new high: are you chasing the top, or catching a trend early? We added one check — revenue has to be at a new high too. This Taiwan stock strategy (Taiwan New High + Revenue Check, official and free) does one thing each month — after monthly revenue reports come out, it picks stocks whose price just hit a recent high and whose last three months of revenue are the best in a year, then buys the ones with the fastest revenue growth, trading at the next day's open. Backtested February 2015 to October 2026: +1,315% total return, about 25.4% annualized, Sharpe 1.56, max drawdown −22.0% — about a third shallower than the −34.0% from simply holding the 0050 ETF. Why check revenue when the price is already breaking out? Plenty of price breakouts are hype or short-term money that fades once the move is done. Revenue is the actual business. When price and revenue make new highs together, the market is usually chasing fundamentals that are really improving, not just a story. Both conditions have to hold: First, price: the close is the highest of the last 20 trading days. Second, revenue: average monthly revenue over the last 3 months is the highest it has been in 12 months. Stocks that pass are ranked by year-over-year monthly revenue growth; the top 10 are held equally weighted, capped at 12% each. Stocks under disposition (Taiwan's trading-restriction list) are skipped, and a holding that lands on the list is sold at the next open. Why wait until after the 13th to rebalance? Taiwan-listed companies report monthly revenue by the 10th, later if it falls on a holiday. The strategy waits for the first trading day on or after the 13th of each month, picks at the close and trades at the next open. On every day of the backtest it only sees revenue that had already been published, never numbers that weren't known yet. Backtest performance (2015–2026, daily bars) MetricValueTotal return+1,315% (0050 +794%)Annualized~25.4% (0050 ~20.6%)Sharpe1.56Max drawdown−22.0% (0050 −34.0%)Year by yearAhead of 0050 in 7 of 11 years from 2015 to 2025; −14.3% in the 2022 bear market vs −21.4% for 0050 Includes 0.5% cost per side, covering brokerage, the transaction tax on sells and slippage. Backtest results, not a guarantee of future performance. Six things to know before you run it First, it rarely fills all 10 slots. Few stocks pass both checks at once: only 34 of 142 rebalances found 10 names, the average holding count is about 5.6, about 38% of capital sits in cash on average, and 8 months held nothing at all. Part of the shallow drawdown simply comes from not being fully invested. Second, timing matters a lot. Same rules, rebalanced on the 11th: Sharpe 1.68. On the 13th: 1.56. One day later on the 14th: 1.18. From the 16th on: about 1.0. The days right after revenue reports carry most of the edge, so once it's deployed, rebalance on schedule. Third, it lags when mega-caps carry the market. In 2025 it made 6.5% while 0050 rose 36.9%, and it trailed slightly in 2019, 2020 and 2023. The stock list is a 292-name sample across industries without TSMC, so it can't keep up when the largest names do the lifting. Fourth, the stock list is today's listed companies. Stocks delisted or acquired along the way aren't in it, and that survivorship bias flatters the backtest — expect live results to come in lower. Fifth, there is no out-of-sample period. The parameters were set on the same data. We scanned the new-high window, number of holdings, position cap and rebalance day; the neighborhood-average Sharpe came out at 0.99× and 0.90× the chosen setting — a plateau, not a lucky spike, but still no promise about the future. The deepest drawdown ran from July 2021 to June 2022 and took until December 2023 to recover. Sixth, fill prices are assumed: the backtest fills at the next open, and opening auctions can fill at worse prices. The rules are simple. Running them on time every month is the work. The logic is all above. What takes time is pulling revenue and prices for nearly 300 stocks the moment reports land, checking new highs, ranking growth, dropping restricted names, and rebalancing right after the 13th without missing a month. Blave Agent runs that chain for you: add this free strategy from the library, then tell your Agent to deploy it. See the Taiwan new-high strategy in the library →