Strategy pattern · Momentum
Test a time-series momentum strategy before any money is at risk.
Super Algo Kit grades the time-series momentum strategy A: hold each market long after a positive past return and short after a negative one. That grade covers diversified futures portfolios studied over decades. A one-market or intraday version is capped at B or C. The kit tests your version with costs, walk-forward checks and a paper run before any live step.
strategy facts
Time-series momentum across many futures markets
- Family
- Momentum, from published literature
- Markets
- Futures, forex, commodities, bonds and equities
- Timeframes
- Daily and weekly bars, computed on completed daily closes
- Evidence grade
- A: strong published evidence, but only for broad multi-asset portfolios tested over decades
- Status
- Literature: described in published research, not yet tested by the kit
- Data needed
- Daily prices per contract month, roll calendars, multipliers, margins, cash rates and per-contract costs
- Your variant
- A single-market, short look-back or intraday version is capped at grade B or C
From the kit's own card. Not investment advice.
Paper first
Test it before you trade it.
Super Algo Kit runs inside the coding agent you already use. It interviews you, writes the spec with pass marks set first, then builds, backtests and paper-tests. Real money starts only when you type the authorize command.
The plan, step by step
Tick them off as you go.
How does time-series momentum work?
The idea is that a market's own recent trend tends to persist. In futures on stock indices, bonds, currencies and commodities, moves over the past 1 to 12 months have tended to continue. Over longer spans they have partly faded.
Researchers point to slow reactions to news, later overreaction, and hedgers who trade against trends. The trend follower takes the other side of traders who sell late in falls and buy late in rebounds. The research finds it clearest across a broad basket of liquid futures; any single market is far noisier.
Time-series momentum rules in plain words
This is a summary, not the card's full rule set. Every look-back and rebalance choice you try counts as a trial.
- Hold a broad set of liquid futures from several asset classes, rolled by each pack's rule.
- Check whether each market's excess return over the last 12 months is above or below zero.
- Buy when it is above zero and sell short when it is below.
- Size each market to a similar share of risk, capped by your risk budget.
- Rebalance weekly or monthly.
What the evidence says about trend following
The founding study by Moskowitz, Ooi and Pedersen (2012) is reported to cover 58 liquid instruments. Hurst, Ooi and Pedersen (2017) then looked at a century of trend-following evidence.
Grade A describes that diversified evidence, not your result. Research on published anomalies also suggests that predictability tends to shrink once a strategy is published and crowded.
How Super Algo Kit tests a momentum bot before any money
You set the pass marks, trial budget and minimum sample in the strategy card before the first run. The backtest fills on the actual traded contracts, with roll costs and the pack's costs.
Walk-forward validation deflates the Sharpe ratio for every look-back tried. Once ten or more configurations exist, it also estimates the chance of overfitting. Results are split by trend and volatility regime, and the kit checks the smallest account that can hold the set in whole contracts.
Before you rely on it
Where it fails.
- Choppy, sideways years bring many small losses.
- Sudden V-shaped turns leave positions facing the wrong way.
- Long dull or losing spells are normal, and many people give up at the low point.
- A single market is much noisier than the broad portfolio the research studied.
- A small account cannot buy enough whole contracts to spread risk across many futures.
- Futures leverage can turn a normal drawdown into a margin problem if sizing ignores margin.
FAQ
Time-series momentum: questions.
Is time-series momentum the same as trend following?
Time-series momentum is the core rule behind many trend-following systems: each market is judged on its own past return. The kit's card cites a century-long study of trend-following investing alongside the original 2012 paper. Moving-average filters and breakouts are close relatives with their own cards.
Does time-series momentum still work?
Super Algo Kit makes no promise that it does. The kit grades it A for diversified futures portfolios over decades, and also notes that published, crowded strategies tend to weaken. Your version must clear pass marks you set before testing, on data it was not tuned on.
Can I run time-series momentum on one market or a small account?
You can test it, but the kit grades a one-market version at most B or C. A single market is much noisier than the diversified portfolio in the research. Small accounts often cannot hold enough futures in whole contracts, so the kit checks the minimum account size first.
What data does a time-series momentum backtest need?
You need daily prices for every contract month: a back-adjusted series to compute the trend, and the traded contract's own prices for fills. Add roll and expiry calendars, multipliers, margins, cash rates and per-contract costs. Long histories matter, since a modest Sharpe ratio takes many years to separate from luck.
Can Super Algo Kit build a time-series momentum bot for me?
Super Algo Kit helps your own AI coding agent build and test one as real code in your project. It interviews you, writes a strategy card and spec, then backtests with costs and runs a paper test. It does not trade on your behalf; live starts only when you type the authorize command.
Keep reading
More guides.
United States · Broker APIInteractive BrokersSuper Algo Kit helps you build an Interactive Brokers trading bot inside your own AI coding agent.
United States · Data feedDatabentoSuper Algo Kit lets you use Databento data for backtesting a bot you build in your own AI coding agent.Claude Code · Coding agentClaude CodeSuper Algo Kit turns Claude Code into an algo trading team, so you can build your own Claude Code trading bot.Sources the kit's card cites
- Moskowitz, Ooi and Pedersen (2012). Time series momentum. Journal of Financial Economics.
- Hurst, Ooi and Pedersen (2017). A Century of Evidence on Trend-Following Investing. Journal of Portfolio Management.
- McLean and Pontiff (2016). Does Academic Research Destroy Stock Return Predictability? Journal of Finance.
Build it with Super Algo Kit.
15 specialist roles and 35 skills inside your coding agent. Paper first, live only when you say.
Super Algo Kit is development software. It is not investment advice, a signal service or a promise of performance. Trading involves substantial risk, including the loss of all capital you trade with.