ATR Band buys a stock after a sharp drop, when the daily close sits more than one ATR(10) below its 3-period SMA. It holds until the close rebounds to more than one ATR above that same average. Each trade commits a fixed 10,000 of capital and only goes long.
Best run of 31. Broadcom, Daily, from run stock2-queue:2026-09-15-Broadcom.tsv:6.
| Period | Up to 2026-09-15 |
| Instrument | Broadcom, Daily |
| Capital per trade | 10,000 |
| Costs | Not recorded for this run. |
| Bars | Daily bars, full available history |
| Last tested | 2026-09-15 |
| capitalPerTrade Money committed to each trade | 10000 |
| basePeriod SMA length the bands are built around | 3 |
| atrPeriod ATR length | 10 |
Every run
| Market | Timeframe | PF | Trades | Win % | Net | Max DD |
|---|---|---|---|---|---|---|
| BroadcomBest run | Daily | 10.31 | 53 | 83.02 | 51,575 USD | -5,068 |
| Visa | Daily | 9.39 | 49 | 83.67 | 32,600 USD | -3,527 |
| Walmart | Daily | 5.23 | 157 | 57.96 | 135,597 USD | -15,060 |
| Microsoft | Daily | 5.15 | 121 | 77.69 | 77,605 USD | -5,054 |
| Mastercard | Daily | 4.91 | 59 | 79.66 | 35,668 USD | -4,536 |
| Cisco | Daily | 4.69 | 87 | 74.71 | 71,241 USD | -10,153 |
| Costco | Daily | 4.20 | 79 | 73.42 | 42,451 USD | -5,320 |
| Johnson & Johnson | Daily | 3.95 | 203 | 75.37 | 61,424 USD | -3,376 |
| AbbVie | Daily | 3.72 | 36 | 75.00 | 14,076 USD | -4,430 |
| Alphabet | Daily | 3.58 | 37 | 83.78 | 18,329 USD | -3,806 |
| Chevron | Daily | 3.33 | 62 | 67.74 | 27,498 USD | -5,817 |
| Lam Research | Daily | 3.19 | 115 | 77.39 | 84,433 USD | -10,865 |
| S&P 500 | Daily | 3.02 | 155 | 76.77 | 31,583 EUR | -5,680 |
| Amazon | Daily | 2.97 | 77 | 76.62 | 42,925 USD | -7,601 |
| ExxonMobil | Daily | 2.94 | 204 | 65.20 | 48,788 USD | -4,653 |
| Coca-Cola | Daily | 2.86 | 158 | 72.78 | 45,959 USD | -6,099 |
| Tesla | Daily | 2.80 | 49 | 65.31 | 28,497 USD | -8,138 |
| Palantir | Daily | 2.78 | 17 | 64.71 | 17,676 USD | -9,607 |
| Berkshire Hathaway | Daily | 2.56 | 89 | 74.16 | 19,334 USD | -5,059 |
| Meta | Daily | 2.45 | 42 | 76.19 | 19,292 USD | -8,016 |
| Oracle | Daily | 2.30 | 98 | 68.37 | 51,842 USD | -8,110 |
| Eli Lilly | Daily | 1.93 | 183 | 66.12 | 32,263 USD | -6,996 |
| Caterpillar | Daily | 1.87 | 198 | 66.16 | 40,636 USD | -7,830 |
| Applied Materials | Daily | 1.80 | 68 | 76.47 | 22,693 USD | -9,503 |
| Bank of America | Daily | 1.42 | 86 | 65.12 | 14,820 USD | -14,817 |
| AMD | Daily | 1.22 | 141 | 63.12 | 22,332 USD | -17,935 |
| JPMorgan | Daily | 0.67 | 76 | 55.26 | -10,137 USD | -11,148 |
| Apple | Daily | 0.52 | 26 | 30.77 | -10,361 USD | -14,078 |
| Nvidia | Daily | 0.20 | 4 | 25.00 | -10,137 USD | -13,684 |
| Intel | Daily | 0.00 | 1 | 0.00 | -10,000 USD | -10,000 |
| Micron | Daily | 0.00 | 5 | 0.00 | -10,083 USD | -10,170 |
Charts
Above 1.0 made money, below 1.0 lost money.
The code
Every file in the library uses the same header, the same section order and the same variable names. The numbers above were produced by the original version before the rewrite, and the rewrite has not yet been compiled in ProRealTime.
// ============================================================ // ATR Band // ProRealAlgos Strategy Library | prorealalgos.com // Type: Mean reversion // Direction: Long only // Timeframe tested: Daily // ============================================================ // --- Settings --- DEFPARAM CumulateOrders = false // --- Parameters --- capitalPerTrade = 10000 // money committed to each trade basePeriod = 3 // short SMA the bands are built around atrPeriod = 10 // ATR length, one ATR each side // --- Indicators --- positionSize = capitalPerTrade / close baseMA = Average[basePeriod](close) atrValue = AverageTrueRange[atrPeriod] lowerBand = baseMA - atrValue upperBand = baseMA + atrValue
15 more lines: the entry and exit rules.
Read this first
Questions
Yes. It is a standard ProOrder system with no manual steps, trading on daily bars. The stats come from the pre-rewrite version and the rewritten code has not been compile checked yet, so run it in demo first.
Broadcom was the best run with a net profit of 51,575.17 USD, a 83.02% win rate and a max drawdown of 5,067.59 USD. Walmart made the most money (135,597.09 USD) but with a 15,059.76 USD drawdown, and five stocks lost money.
Each trade uses 10,000 of capital. On Broadcom the worst drawdown was 5,067.59 USD, and on AMD it reached 17,934.84 USD, so plan for drawdowns larger than half the trade size.
The rule reacts to one-day overshoots measured by a 10-day ATR. Daily bars keep the trade count low (53 on Broadcom) and make holding through noise part of the design.
Related
The ProRealAlgos community gets the full rules, the complete code and the automated portfolio.
