Direct answer
MT5 indicators are tools inside MetaTrader 5 that compute values from market price data (and sometimes volume or other series) using a defined formula. Their main limitation is that the relationships they display are not guaranteed to persist. Even when an indicator looks consistent in the past, the future can differ due to changing market behavior, execution costs, and the fact that an indicator cannot predict events beyond the data it is built from.
Mechanism and definition
An indicator is an algorithm that takes inputs—such as the chosen timeframe, price fields (for example, open/high/low/close), and indicator-specific parameters—and outputs a new series (such as a moving average line, an oscillator, or bands). In practice, the indicator logic runs repeatedly as new bars form.
This leads to two important mechanics-related uncertainties:
- Dependency on inputs: Change the timeframe, price type, or parameters, and the indicator’s output can change materially.
- Indicator view vs. decision reality: Indicators do not include every element needed for real outcomes, such as bid/ask spreads, order fill behavior, slippage, and how positions are actually managed.
Evidence or example (why “it worked once” can fail)
A common failure mode is assuming that a historical pattern implies future repeatability. For example, a moving average may have acted as a trend reference in a past period, but trend strength, volatility, and participant behavior can shift. When those conditions change, the same moving average rule may produce more false crossovers or late entries.
Another example is parameter sensitivity. Suppose two versions of the same oscillator use different lookback lengths. The shorter lookback often reacts faster but can be noisier; the longer lookback may be smoother but can lag. In both cases, there is no universal parameter choice that works everywhere because market structure and volatility regimes can vary over time.
Limitations and risks to expect
Material limitations often show up in several places:
1) Lag and delayed information
Many indicators are based on smoothing or lookback windows. That smoothing can reduce noise, but it also means the signal reflects what has already happened. In fast-moving conditions, the indicator may update after the most relevant price change.
2) Noise, overfitting, and regime dependence
Indicators can fit a specific historical regime (a specific pattern of volatility and trend behavior). When the market enters a different regime, the indicator may underperform. A system that appears effective in one sample may fail when conditions differ.
3) Non-stationary relationships
Markets are not static systems. The statistical properties that an indicator implicitly relies on (such as volatility patterns or how quickly prices revert) can change. Because the indicator formula does not “know” that change occurred, its output can become less meaningful.
4) Repainting and calculation timing risks (conceptual)
Some indicator designs can produce outputs that change as more data becomes available (for example, if an indicator uses values that are revised when a bar completes, or if it depends on future-looking calculations). Even without naming specific products, the general risk is that what appears on the chart historically may not match what would have been observable at the time.
5) Backtest uncertainty and missing costs/execution details
Historical performance claims are easy to misread if the analysis ignores assumptions. Typical sources of mismatch include:
- Costs and execution effects: Indicators do not simulate real order filling, spread changes, or slippage.
- Data handling assumptions: Different data feeds or bar formation rules can lead to different computed series.
- Selection bias: Tweaking parameters to match a past period can create an illusion of robustness.
Verification and next question
You can verify indicator behavior independently by focusing on stability and observability rather than prediction:
- Define the indicator inputs exactly: record timeframe, price type, and parameters.
- Test across multiple, non-overlapping periods: check whether the same indicator behavior remains consistent when volatility and market behavior change.
- Compare “in-chart” vs. “at-time” availability: look for whether historical outputs would have been known when the bar was forming.
- Include realistic assumptions in evaluation: treat costs and execution uncertainty as part of the analysis, not an afterthought.