Flip of a coin..but

[Overnight]

For a day going from low to high - the day low will be found within the first 60 minutes 74 % (!) of the time on ES.

Within the first 90 minutes the day low will be found 85 % (!) of the time.

View attachment 303761

Sample size is the last 8 years. RTH session. ES.

...

Yep, that is the pattern I started seeing in 2018 when I switched to equities. I called it the 10:30AM reversal. It was a great time for bulls.

How does that look if you run the data from Jan 2022 until now?

 

[smallAxe]

I don't think human affairs are random. But, we're specifically talking about the financial markets here so... I'd try to serve the needs of the participants.

Yes, that's what I mean. You're not participating in the game, right?

So if I believe that prices are random, then the only bet I can be sure of is the unlikelyhood of consecutive hits.

In a 50/50 dice, what's the probability of 10 head in a row? 15? 20? The longer the streak, the lesser the probability. ex: 15 streak is about 1.5%

https://www.omnicalculator.com/statistics/coin-flip-streak

But saying you're just going to double your bet like a martingale system after every loss is never going to work out since you have limited capital or exchange limits

So you do the next best thing, you scale in. I don't care what entry rules you pick, you add a scale in process into that strategy, you'll have a higher hit rate (assumming a 1:1 RR).

The trick is to manage your trading size so that your scalein can withstand those long xR streaks (trends). It's all about managing your average entry. You do it too much too soon, you'll blow up.

Smaller incremental scale in, breakeven stops, high volume trading periods should all help minimizing the risk of blow up.

So like I said you can manage your risk like a poker player. You split it into various pools and go all out into each one (you don't just go all in with your stake in one session at a table). Probablity speaking you will have about 40% chance of doubling up if your win rate is 85% (using 2% risk), or 95% chance of doubling up with 90% win rate.

http://www.beatingbonuses.com/calc_streak.htm

You still need a decent amount of capital, but I think with micros you can pull it off. At least I think so, that's what i'm doing

Hilmy83, if I wanted to make a living, I couldn't play in a 50/50 coin toss, unless I could find someone willing to bet 2 for 1.

When I can't find that person, while you're forcing me with a loaded gun, what do I do then?

Alright, I'll bet, but only on the first toss and only as much as I can afford to lose. I can't expect the next toss to come out heads in a row. On the first toss, win or lose, I go home, and life goes on.

That's how the game is to win (lose) on that day. My decision on that day, does not affect tomorrow.

The market is not a game like that. The market is like life. If I lose today, I have a match again later, tomorrow, the day after, and so on, and so forth. Every decision is not independent, price changes are not independent: today, affects tomorrow.

Let's change the perspective a little here.

Can a random entry strategy make money in the market long term?

Yes, it can, of course.

With the condition, the participants are very many. Say with 1,000,000 people, some random winners can come out safe.

Now, imagine only consisting of 10 participants, do you think any random participant can come out safe?

The market is followed by many participants. So, random winners can very well happen, and this is something very normal, common, and usual: in a bull market, everyone is an expert.

 

[Laissez Faire]

Yep, that is the pattern I started seeing in 2018 when I switched to equities. I called it the 10:30AM reversal. It was a great time for bulls.

How does that look if you run the data from Jan 2022 until now?

Here's the same statistics spanning from the entire last year (2022) and until present date (last Friday). In summary, it's basically the same as for the entire 8 year period.

What some people fail to recognize is that every new day in the market is basically a repetition of a similar day from the past. It’s the same patterns repeating themselves over and over again. Just look at last Friday. There's nothing special about last Friday. I wonder how many similar days we've had over the years. I could of course answer that, but asking rhetorically.

Once in a while you get the outliers, but most days are a repetition over a common theme.

So, again, I strongly disagree with the notion that markets are moving randomly. They’re not.

 

[Tradess0610]

Interesting stats.
But I find suspicious that in 8 years only one time the low of the day was in the last hour of trading. It may be the case, but it doesn't sound right. I wouldn't bet 2$ on it.
How come you only have 1065 data points for 8 years? Shouldn't be more like 2000?

Could it come from traders closing their daily profits of a trade with a trend, which can reverse the price slightly?

 

[Tradess0610]

Let's change the perspective a little here.

Can a random entry strategy make money in the market long term?

Would guess yes - in case you are trading in the direction of a predominant trend. It should be more likely than not, that the trade would end up in profit.

With random entry i ment the timing of entry is random - the direction of the trade is up to the trader

 

[newwurldmn]

Yea I agree with that. I did mention trading as the context of the discussion.

But what if, in the grand scheme of things, the stock market is random. Honestly, how long has our stock market existed for? Relative to our existence, it's pretty short history. Just a snippet of a lucky streaks of heads...

wealth is created as a civilization progresses by building things and improving its people’s standard of living (until it blows up). The stock market is a reflection of that wealth creation.

 

[SunTrader]

Survivorship bias in the indices gets a lot of the "credit" too.

 

[hilmy83]

Did some weekend backtests:

Took the 1st minute opening range of the ES. Using neutral entry (first hit) fading vs breakout of that range with 1:1 RR, yields 50% success rate for both.

At the very basic level, there is a 50/50 chance that a price will close either way on the next bar. My current strategy is anticipating the breakout off the opening range (at some point). But the problem is when you get into consecutive losses, you're pretty much will revert back to 50/50 chance. You can do things like smaller/bigger targets, scalin in , martingale, to skew the EV.

There are so many papers that support brownian motion for these financial markets.

In fact just made these using random() function in excel. I'm pretty sure i can fool somebody into creating a technical analysis on these charts.

View attachment 303288

View attachment 303290

View attachment 303291



So where am i going with this. Basically I think trading is mostly bullshit lol. If I'm going to play the random game, I might as well play the mean reversion with scalin in. If anything that these charts show, is that there will be some kind of retracement. No lines jsut go straight up/down, MOST of the time.

But you will have days like this (MES 12/22/22)

View attachment 303292

But if you use b/e stops and scale in, and manage your risk with smaller position (to give room) you can "potentially" avoid gettting fucked in the ass..

What do you think guys?

Just sharing my results of the test that reversion/breakout off the 1st minute bar is 50/50 outcome

Rule: Faded the opening 1 minute range. Target and stop loss open till end of day. Testing during spring dst

some half day holidays might've screwed up the results a bit


2021
3/15-11/5

2022
3/14-11/4

Win rate is 49% over all
avg w/l =1.08
avg W=144.72 avg L=134.02
EV=$2.56

almost 50/50. letting the trades run seem to give it just a little edge overall.

Moral of the story is, let it ride

 

[SunTrader]

Fading the opening one min range?

Why do that?

 

[hilmy83]

Fading the opening one min range?

Why do that?

Just showing the flip side of ORB, which I use. It's 50/50 at that scale.

Incorporating stop and reverse, b/e, and fixed martingale seem to eek out a small edge from my own personal trading.