Here’s a little snippet concerning the price behaviour of EUR/USD. If price makes a first daily higher close and then next day a second daily higher close, the probability of the next day closing lower is about 75%.
The correlation holds true for lower closes also, and is relevant from at least the start of 2020 right to the present week (I haven’t checked earlier than 2020).
You’re welcome to try to do something with this, please post some details.
Thanks for this interesting fact. Tested this for 2023 only:
Sell at 9am UTC+2, stop loss at the daily high but at least 20 pips. Close the position at 8pm UTC+2
Trades: 26 Profit: 491 pips Wins: 13 Losses: 12 Breakeven: 1
January 10: -20
January 20: -14
February 2: +96
February 15: +42
March 7: +134
March 13: -20 (+75 turned into a stop out)
March 20: -35
March 29: -28
April 4: +39
April 13: -25
April 24: -22
May 5: +40
June 9: +29
June 14: -20
June 28: +16
July 4: +26
July 10: -20
August 7: +40
August 11: +40
August 30: -25
September 12: 0
September 19: -20
October 2: +97
October 6: -20
October 12: +108
October 18: +53
Next monday would be a valid sell, too 
Nice test thanks. Did you use software to run this or work through manually?
No old school, a chart, a pen and a pice of paper. Sorry this is all I can do. Backtesting is a lot of work for me.
My head hurts just thinking about it.
Ok here is the long version, 2023 until now:
Trades: 28 Profit: -5 pips Wins: 14 Losses: 14 Breakeven: 0
Jan 4 +12
Jan 30 -14
Feb 6 -20
Feb 17 -20
Feb 22 -20
Mar 9 +20
Mar 27 +32
Apr 11 +26
Apr 18 +22
May 10 -20
May 15 +6
May 18 -20
May 24 -20
Jun 20 -20
Jun 26 +7
Jun 30 -20
Jul 6 +14
Jul 20 -20
Aug 2 -20
Aug 9 +6
Aug 15 -23
Aug 28 -7
Sep 4 +5
Sep 15 +12
Sep 21 +16
Sep 26 -1
Oct 4 +42
Oct 16 +20
How would you feel about trading this price action without a stop-loss?
We can talk about widening the stop loss, letting positions run for a swing trade, pyramiding and using for example the daily 200 ema as a directional filter. But at one point there has to be a stop loss. at least for me.
According to the info provided by Wokre bee,
The above spreadsheet is how i would organize my data
As you can see from my FREQUENCIES table, the chances of having 3 wins in a row is 14%.
I beg to differ . . .
For the next instances of price making higher DAILY close on 2 consecutive days, i’m inclined to BUY at 9am UTC+2 instead of selling. I would set stoploss at 30pip since chances of that is only 4% , and target 20pip profit which has a 54% chance of occurrence .
Ah I see what you have done there. Nice work and interesting conclusion. Did you also include the data of the short backtest in my first post?
And you have to account for the fact that the minimum SL of 20 pips is the reason why there are so few stops with 30 pips.
OIC . . .
In that case, we set stoploss to be 20pip instead… i guess
Your frequencies table shows 54%, 25%, 14%, 7% and pure random would be 50%, 25%, 12.5%, 6.25%. You are running slightly higher than random. However that could be accounted for by the statistically small number of data points that you have.
Beauty lies in the eye of the beholder . . .
The VALUE of the above colour coded table lies in the way the data was organized, and a way to interpret was developed. How a trader utilizes this information for trading is secondary, as it depends on the trader’s risk appetite, personal discretion, and preferences.
Some additional performances from the current year, to 18/10 only, just for comparison and hopefully to confirm this price action is a real phenomenon.
Probability of 2 consecutive higher closes being followed by a lower close (or vice versa, 2 consecutive lower closes being followed by a higher close) -
key US index, S&P500 = 77%
key UK index, FTSE100 = 68%
least volatile major, EUR/CHF = 76%
most volatile pair, AUD/JPY = 81%
least volatile pair, AUD/NZD = 79%
Higher, higher then lower or lower, lower then higher starts to look like a reliable pattern of price action.
@ayodimej1 - You have shown a section of the D1 chart for EUR/USD.
Look for 2 green candles which follow a red candle - the probability of a red candle being the next candle after the two green candles is about 75%. Also vice versa, when you see two green candles after a red one.
Round-up for Monday 23/10, across the 28 leading forex pairs -
20 signals - 11 winners, 9 losers
Nice test! Keep going on and share more.
Round-up for Tuesday 24/10, across the 28 leading forex pairs -
17 signals - 16 winners, 1 loser
Of course, if price makes a third or subsequent higher daily close, the probability of the next close being lower is even greater.
