Skip to content
All library documents

Why Proprietary Trading Signals Are Hard to Reverse Engineer

Article Quant Q&A · Author: rook
Read the original

Summary

The discussion considers how an opaque trading service might produce buy and sell signals. It offers no verified description of the service’s algorithm: its own marketing claims only that the signals come from proprietary mathematical logic. The answer cautions that the method may be too complex to infer externally, or that its claimed sophistication may be overstated.

One proposed possibility is that signals are optimized against historical chart data and then reused on live prices. Such a process could fit past bars, with signal placement changing when the chart’s time span or data changes. The explanation raises look-ahead bias and instability as risks, but presents them as hypotheses rather than findings about the service. Without code, data, or controlled testing, the actual signal logic and its predictive value remain unknown.

Key ideas

  • The service does not disclose enough information to identify its signal-generation method.
  • Marketing claims alone do not establish that a proprietary algorithm has predictive value.
  • Historical optimization could create signals that change when the chart data or time span changes.
  • Look-ahead bias and overfitting are possible explanations, not confirmed facts about the service.
  • Evaluating the signals requires evidence beyond a visual impression of historical performance.

Tags

BacktestingMachine learningStatistics
Full text
# How does trndbot generate buy/sell signals?


# How does trndbot generate buy/sell signals?












I was looking at trndbot and it's buy/sell signals are pretty good - what algorithm are they using to generate these signals? Although it maybe similar to bollinger bands, it algorithmically generates these live signals for any trading graph:

## Answer by Bob Jansen (score 2)

https://quant.stackexchange.com/a/61714

Although it’s their own marketing and should be taken with a grain of salt they claim:

> Buy and Sell Signals. Our proprietary algorithm bot uses complex logic and mathematical equations to predict profitable trades, and to notify you when its time to sell.

Either it will be hard to reverse engineer because of the complexity of their method, or the properties of the algorithm are overstated and it’s not really worth your time to reverse it. It might for example be more simple or have look ahead bias.

## Answer by Sergei Rodionov (score 1)

https://quant.stackexchange.com/a/61713

I don't know how this TradingView script works, but I wouldn't be surprised if it was implemented as follows. It calculates where to place buy/sell signals to maximize the hypothetical trading profit for the historical data on the chart. The number of signals is controlled with 'Sensitivity' parameter. Higher sensitivity translates into more signals, hence more profit. Then the script applies the same parameters that worked for historical data to real-time updates. If the chart is reloaded with a different timespan, or different bars, the signals would be placed at different times, or even reverse.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.