A Simple Mean Reversion Strategy | Page 18 | Forex Factory

Options

Go to
Permalinks
Print Thread

Related Instruments

EUR/GBP

Similar Threads

Simple Mean Reversion 621 replies

M5 Mean Reversion Strategy 65 replies

mean reversion strategy 8 replies

Forward test of my new mean reversion strategy 11 replies

Synthetic hedges, cointegration, mean reversion and similar stuff 419 replies

Attachments: A Simple Mean Reversion Strategy
Exit Attachments

A Simple Mean Reversion Strategy

  • #341
  • Oct 22, 2025 8:23pm Oct 22, 2025 8:23pm
@Riskcuit, thanks for sharing your strategy.

I backtested the strategy over past 4 years, and here's what I found (see screenshot).

It assumes risking 1% of equity per trade, executing on the 15m chart time frame.

Are these figures in line with what you've experienced as well?

From these numbers, if you traded a basket of names with high investability (which = return in local currency / drawdown), and risked more than 1% of equity per trade, you could have done quite well.

Similarly, if you traded the ones with negative investability, it would have been a challenging four years.
Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 21 KB
  • #342
  • Oct 22, 2025 8:26pm Oct 22, 2025 8:26pm
Also @Riskcuit, how did you go about developing this strategy?

I've tested adding other filters, to try to filter out the losing trades, but none of my filters improved the performance of the strategy.
  • #343
  • Oct 26, 2025 7:07am Oct 26, 2025 7:07am
Quoting IronPanda
@Riskcuit, thanks for sharing your strategy. I backtested the strategy over past 4 years, and here's what I found (see screenshot). It assumes risking 1% of equity per trade, executing on the 15m chart time frame. Are these figures in line with what you've experienced as well? From these numbers, if you traded a basket of names with high investability (which = return in local currency / drawdown), and risked more than 1% of equity per trade, you could have done quite well. Similarly, if you traded the ones with negative investability, it would have...
Have the commissions/fees been considered? What entry/exit rules did you apply? Do the rules adapt to the asset dynamically?
  • #344
  • Oct 26, 2025 11:17am Oct 26, 2025 11:17am
1. Mostly yes.
2. Mostly the same ones as Riskcuit posted.
3. Mostly yes.

Overall the general direction and relative magnitude of the results I feel are mostly accurate, but on an absolute basis they probably overstate actual trading due to slippage and bad fills (which is true of all backtest results with retail level data and tools).

Quoting Fiox89
{quote} Have the

1. commissions/fees been considered?

2. What entry/exit rules did you apply?

3. Do the rules adapt to the asset dynamically?
  • #345
  • Oct 27, 2025 2:45am Oct 27, 2025 2:45am
Quoting IronPanda
1. Mostly yes. 2. Mostly the same ones as Riskcuit posted. 3. Mostly yes. Overall the general direction and relative magnitude of the results I feel are mostly accurate, but on an absolute basis they probably overstate actual trading due to slippage and bad fills (which is true of all backtest results with retail level data and tools). {quote}
Unfortunately, I don't see any specific rules in this thread. Selling above the 200MA and buying below the 200MA isn't a rule... it's a general guideline. It's unlikely anyone will be able to tell you if your results are consistent, or extend your strategy, without clear rules.

The distribution you've had is still interesting since it's verified on a portfolio, and there could be potential, but too little information at the moment.
  • #346
  • Oct 27, 2025 11:40am Oct 27, 2025 11:40am
Yes, I asked the web to to figure out how far away from the 200MA is a good entry point, but not so far that's its a bad entry point (per what Riskcuit wrote). There are plenty of suggested methdologies to choose from, I just picked one and ran with it.

Quoting Fiox89
{quote} Unfortunately, I don't see any specific rules in this thread. Selling above the 200MA and buying below the 200MA isn't a rule... it's a general guideline. It's unlikely anyone will be able to tell you if your results are consistent, or extend your strategy, without clear rules. The distribution you've had is still interesting since it's verified on a portfolio, and there could be potential, but too little information at the moment.
  • #347
  • Oct 27, 2025 5:37pm Oct 27, 2025 5:37pm
Quoting IronPanda
Also @Riskcuit, how did you go about developing this strategy? I've tested adding other filters, to try to filter out the losing trades, but none of my filters improved the performance of the strategy.
Thanks for sharing your results! They are inline with my findings.

Regarding how I came up with the strategy, it’s explained pretty clearly in the thread but the main points/ideas that led me here are:
- edges must exist in distribution (have another thread of this).
- market distribution has persistent excess kurtosis. This is the edge (mentioned here and in other threads).
- ma as an indicator isolates this kurtosis, making it more extreme/easier to see and trade (in some senses)


As stated above, the strategy edge is in its distribution, specifically the excess kurtosis which creates the opportunity. There is a ceiling of edge you can capture. Adding signals doesn’t create the kurtosis effect we are targeting with MA. Nothing but raw market operations create this, indicators can highlight it but they don’t influence.

When it comes to this strategy, the best optimization you can do is reduce volatility of returns to create a smoother curve. This will cost edge overall but will provide more stable results so it’s not so much feast or famine, but instead, always have enough to eat.

to do this your best bet is to essentially go delta (or directionally) neutral (sometimes even reverse exposure) when you know you are in a giant tail move via hedging. I prefer to use beta weighted delta so you can easily manage exposure across multiple pairs

Quoting IronPanda
Yes, I asked the web to to figure out how far away from the 200MA is a good entry point, but not so far that's its a bad entry point (per what Riskcuit wrote). There are plenty of suggested methdologies to choose from, I just picked one and ran with it. {quote}
because of how the math works out, the further away from MA, specially the start of the swing (where ma was when it started) the more “edge” is present and the less risk you are actually taking on.

think about the discussion on previous page of thread where another trader did a deep dive into scaling the entry. It’s be summarized as, the higher up the MA canceled you can have you entry when it closes, the better.

This is also how the math works when you are trading volatility instead of kurtosis. The higher volatility is when you sell it, the less risk you are actually taking on.
ri · skuht
  • #348
  • Nov 2, 2025 12:43pm Nov 2, 2025 12:43pm
I've been playing around this concept for awhile and adding my own flavor for take partials and profit, also applied filtering so I won't just trade into heavy down/up trend(filtering such don't trade if price moved too much away from the MA but allow to trade only if there is enough buffer from the MA) and coded it into EA

while my results on timeframe such as 15m was lower then the average, the result for playing this on 1h chart and optimizing the parameters while avoiding too much overfitting is something I can work with and test on more instruments.

EDIT: attached results for EURGBP 1H chart, results till 3.2025 are backtest and forward test(out of sample data) from 3.2025 till 11.2025
Attached Image(s) (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 17 KB
Click to Enlarge Name: screenshot.png Size: 23 KB
profiting on actionable patterns
  • #349
  • Nov 2, 2025 1:41pm Nov 2, 2025 1:41pm
ok, so in addition to my last post regarding bad results for 15m chart the results of trading this using EURUSD 5m chart were wild over 2025
definitely something I'll tune a bit more and apply to live trading
Attached Image(s) (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 20 KB
Click to Enlarge Name: screenshot.png Size: 23 KB
profiting on actionable patterns
  • #350
  • Nov 6, 2025 2:16am Nov 6, 2025 2:16am
15m crude oil
Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 125 KB
  • #351
  • Nov 9, 2025 8:55am Nov 9, 2025 8:55am
Quoting zkchyo
15m crude oil {image}
my results for crudeoil for 15m are very low for 2025

below average results using 15m and 5m chart(attached screenshot)

but i'm blown away with the 1h chart for this(2nd chart)
84% winrate
1.74 profit factor
over 100 trades
Attached Image(s) (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 14 KB
Click to Enlarge Name: screenshot.png Size: 13 KB
Click to Enlarge Name: screenshot.png Size: 17 KB
Click to Enlarge Name: screenshot.png Size: 24 KB
profiting on actionable patterns
  • #352
  • Nov 9, 2025 11:59am Nov 9, 2025 11:59am
Quoting poap
… while my results on timeframe such as 15m was lower then the average, the result for playing this on 1h chart and optimizing the parameters while avoiding too much overfitting is something I can work with and test on more instruments.
In my backtests I found the lower in time frame you go, the greater the impact of trading costs relative to the gross profits, that may not be offset by higher trade frequency.

15m generated better numbers than 5m and 1h backtests.
  • #353
  • Nov 9, 2025 2:46pm Nov 9, 2025 2:46pm
Quoting IronPanda
{quote} In my backtests I found the lower in time frame you go, the greater the impact of trading costs relative to the gross profits, that may not be offset by higher trade frequency. 15m generated better numbers than 5m and 1h backtests.
could be a typo of your? because 1h chart is not so lower relative to 15m and 5m.. guess you meant 1m tf

if so i agree, i'm not even backtesting 1m tfs, spread and commission eating up profit.
anyway trading lower timeframe should come with strict filtering rules. there are days in my EA that you don't see a single trade on the 5m chart
profiting on actionable patterns
  • #354
  • Nov 9, 2025 9:38pm Nov 9, 2025 9:38pm
Quoting poap
{quote} could be a typo of your? because 1h chart is not so lower relative to 15m and 5m.. guess you meant 1m tf if so i agree, i'm not even backtesting 1m tfs, spread and commission eating up profit. anyway trading lower timeframe should come with strict filtering rules. there are days in my EA that you don't see a single trade on the 5m chart
Not so much a typo, more so not being clear.

Spreads are relatively fixed during active trading periods, so as you go lower in time frame, the spreads as % of gross profit goes up.

But at the same time, lower time frames gives you more trades, so increase in # of trades can offset the increate in spreads as % of gross profits.

In my backtests, the 15m gave me best investability (per the definition I gave above) vs 5m, 30m, and 1h time frames.

I wouldn't take my backtest (or anyone else's) as 100% certain. The tools and data sources I use are far from institutional grade.
  • #355
  • Nov 10, 2025 1:41am Nov 10, 2025 1:41am
Quoting poap
{quote} my results for crudeoil for 15m are very low for 2025 below average results using 15m and 5m chart(attached screenshot) but i'm blown away with the 1h chart for this(2nd chart) 84% winrate 1.74 profit factor over 100 trades {image} {image} {image} {image}
I take signals from crude and brent spread. The same reason why OP trades EURGBP. High correlation.
  • #356
  • Nov 10, 2025 3:42am Nov 10, 2025 3:42am
Quoting IronPanda
{quote} Not so much a typo, more so not being clear. Spreads are relatively fixed during active trading periods, so as you go lower in time frame, the spreads as % of gross profit goes up. But at the same time, lower time frames gives you more trades, so increase in # of trades can offset the increate in spreads as % of gross profits. In my backtests, the 15m gave me best investability (per the definition I gave above) vs 5m, 30m, and 1h time frames. I wouldn't take my backtest (or anyone else's) as 100% certain. The tools and data sources I use...
I wouldn't underestimate a retail trader capabilities or tools those days.
I'm far from institutional grade but I manage to develop decent EA and with few more $ a month you can either set mt5 server agent on some vps(or old unused laptop) or even use the mt5 cloud agent to help you accumulate backtesting/forward testing results to understand what might work and where to invest your resources

having said that 5m indeed create a lot of noisy signals which in the spread can eat the profits, but also create some quality signals, while the 15m filtered also good signals that the 5m produced
which is why it 5m MUST be optimized using some filtering and once in awhile re-optimize.
profiting on actionable patterns
  • #357
  • Nov 10, 2025 4:45am Nov 10, 2025 4:45am
Trade small and manage positions well. Here is my EURUSD on 1m after commission fees.

Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 6 KB
  • #358
  • Nov 13, 2025 8:17am Nov 13, 2025 8:17am
First of all thanks to Riskcuit for posting this method. I have just stated trying it out on the 15 min using the highest performing pairs from IronPanda's investability list (Thanks IP). Works well!
I'm having a problem hitting all the entries because I can't be awake 24 hours per day. So I'm placing orders at prices that could be hit in future by linear extrapolation from current conditions (just before I go to bed, west coast of NA). The orders always are beyond SL of any open positions so they are anticipating further price movement in same direction. It has worked well in several cases so far.
Does anyone here have comments or suggestions to improve/change this approach?
  • #359
  • Nov 13, 2025 1:51pm Nov 13, 2025 1:51pm
I use the same approach also with great success since long time. Anyway I found out, that sometimes on every market long and strong trends happen, which could generate a higher than usual draw down. I think about to find a way to implement a trend following strategy which kicks in, when such a strong trend occurs and which relative's the draw down a bit… any ideas how to do this?

i tested also some hedging approaches, but they work semi good in my opinion, because you never know when to stop the hedge…
  • #360
  • Nov 13, 2025 4:45pm Nov 13, 2025 4:45pm
Quoting Pip_Monk
First of all thanks to Riskcuit for posting this method. I have just stated trying it out on the 15 min using the highest performing pairs from IronPanda's investability list (Thanks IP). Works well! I'm having a problem hitting all the entries because I can't be awake 24 hours per day. So I'm placing orders at prices that could be hit in future by linear extrapolation from current conditions (just before I go to bed, west coast of NA). The orders always are beyond SL of any open positions so they are anticipating further price movement in same...
You can probably get 80% of the results by limiting new entries to only NY session hours, and trading a small set of names that have no or low direct correlation to each other.

That will make the system easier to trade.
0 traders viewing now