How do you measure the real impact of economic news on an EA?

New comment

How do you measure the real impact of economic news on an EA?

I've been researching how automated trading systems behave around economic releases, and I've noticed that simply classifying events as Low, Medium or High impact doesn't tell the whole story.

Two High-impact events can produce completely different market reactions depending on the symbol, volatility and market conditions.

I've been experimenting with measuring things such as:

  • Price movement 1, 5, 15 and 30 minutes after the release
  • Maximum upward/downward excursion
  • Time to peak movement
  • Reversal after the initial reaction
  • Historical median and percentile reactions for the same event
  • Performance of EA trades exposed to news versus normal conditions

I'm curious how other algorithmic traders handle this.

Do you simply block trading around High-impact news, or do you measure the historical reaction of each event before deciding how relevant it actually is?

I'm currently building some tools around this idea for MT5, so I'm especially interested in approaches that can be measured objectively rather than manually classifying every event.


Do you simply block trading around High-impact news, or do you measure the historical reaction of each event before deciding how relevant it actually is?


To block trading because of the news is wrong. I won't enter in details. Stop loss can be easily hit, true but is a small loss. If the news move the price in my favor i will win more. We talk about asymmetry. The fear of losing trades keeps over 90% of the traders unprofitable.

I honestly just remove the EA from the chart before a big news release and then re-attach it after the initial chaos. I let the EA find the next valid signal (not looking for bounces after news reaction). I don't have news checks coded into my EA. It takes very little time for me to log into the VPS. Whether I believe in full automation or not is still unknown, so I work in this manual/automated way.
Tag every trade with the time of the nearest high-impact release (the MT5 economic calendar is accessible from MQL5 via CalendarValueHistory), then split the EA's closed deals into "within N minutes of an event" and "everything else". Compare average R-multiple (profit divided by initial risk), win rate and average slippage between the two groups rather than total profit, since the news group will be a small sample. Use the same split on both the Strategy Tester run and live history so you can see whether news trades behave differently in each. Since news samples are small, bootstrap the news-trade average or look at its confidence interval before concluding anything; with fewer than ~30 trades per bucket, a "difference" is mostly noise. Also record spread and fill price versus requested price at entry, because on news the cost side (spread and slippage) often explains more than direction does. Only after that decide whether to block, reduce size, or leave the EA alone.

Hi Abel,

Abdullah's split is a good start. One thing I'd add: look at what the EA is already holding when the release hits, not only at trades opened near it.

Many EAs, grids especially, don't open anything around news but sit with severalpositions already open. Those never show up as "news trades", but that's where the damage usually comes from.

So I'd log open positions and floating drawdown at release time and look at the worst case, not the average.
On an account with a daily loss limit, one bad release is enough.

Best regards
Chris
[Deleted]
Great question, and I agree with the direction Abdullah and Chris took this. One thing I'd add: measure the cost side separately from the direction side. On news, the damage often comes from spread widening and slippage rather than the price move itself — on gold (XAUUSD), spreads can easily widen 5-10x around NFP or CPI, which quietly kills scalping EAs even when the trade direction is right. I'd also segment by event type rather than lumping all 'high impact' releases together, since NFP, CPI and FOMC decisions produce very different volatility signatures on gold. The cleanest test is an A/B run in the Strategy Tester with the news filter on versus off over the same period, then comparing average R-multiple and max drawdown between the two. That tells you objectively whether the filter earns its keep or just removes profitable trades.
Personally, I would prefer to pause around CPI and FOMC unless there's convincing evidence that the EA handles those conditions well

To measure the impact, you wanna compare the same EA with the news filter enabled and disabled, keeping the settings and risk identical. Look at average profit per trade after costs, maximum equity drawdown and the worst losses across a meaningful number of releases. Then check whether those results hold on data you haven’t used to tune the filter

Spreads and slippage matter too. So you wanna use real ticks and test execution delays in MT5, then compare that with live execution records. Including the trades already open when the news hits

Your measurements are a good starting point. But the deciding factor whether to trade those events or not depends on if the extra profit justify the additional risk. If it doesn't, then you should always sit that one out

I prefer measuring the effect of news instead of assuming that every high-impact event has the same market impact.

For each release, I would compare the same strategy and symbol under identical conditions and record:

• spread before, during and after the release
• slippage and execution failures
• price movement after 1, 5, 15 and 30 minutes
• maximum adverse and favorable excursion
• stop-loss and take-profit outcomes
• results with the news filter enabled and disabled

In an EA, the news filter should normally prevent new entries during the configured time window. Existing positions still remain exposed and should continue to be managed by their stop loss, take profit and trailing rules.

I would also apply a separate maximum-spread check. If the current spread exceeds the strategy’s configured limit, a valid signal is ignored. The slippage limit controls the maximum execution deviation allowed when sending the order.

This comparison gives a more objective view of whether the filter improves risk-adjusted results, because spread expansion and execution quality can matter as much as the direction of the initial price reaction.

There is only one true method. Anything else is meaningless in algorithmic trading.

You don't know whether the news is beneficial or negative for your results. So either you prepare a CSV file with all the relevant news from previous years and have the strategy tester read the CSV file for each trade as if it were a news filter, or you don't know. Therefore, the decision to block certain news becomes entirely discretionary. In fact, most news filters in expert trading products are added simply because they provide a sense of security to traders who buy the products.
Roberto Danilo Riccio #:
There is only one true method. Anything else is meaningless in algorithmic trading.

You don't know whether the news is beneficial or negative for your results. So either you prepare a CSV file with all the relevant news from previous years and have the strategy tester read the CSV file for each trade as if it were a news filter, or you don't know. Therefore, the decision to block certain news becomes entirely discretionary. In fact, most news filters in expert trading products are added simply because they provide a sense of security to traders who buy the products.
Good point @Roberto Danilo Riccio, and from the buyer side it goes one step further.

As far as I know the calendar functions don't work in the MT5 tester. So unless the seller did the CSV thing you describe, the news filter on the product page was never part of the backtest at all.
The backtest shows the EA without it, live it runs with it. Two different EAs really.

Thats one of the questions i'd ask a seller before buying: was the filter active in the test, and how?

Bye
Chris
I measure the real impact of economic news on an EA by comparing backtest and live performance during news events versus normal market conditions, focusing on volatility, spread, slippage, execution latency, drawdown, and profitability. I also use historical tick data and event-based testing to identify which news releases affect the strategy most.
12
New comment