I feel like a lot of discussions I've had here recently are with people who know the future and/or the one true and correct way to make assessments of markets.
I'm not a big believer in knowing the future and I do not know many people with over 5 yrs trading under their belt who is.
I'd like to have a good old fashioned trading discussion, if anyone else is interested in such a thing.
How would you quantify your trading edges? In a way that does not assume you know the future, does not require you to be right all the time and does not mean you get "Rekt" if you're wrong on any one instance?
Here's one of mine. One I've used for a long time.
I like to trade retracement levels of previous swings. I do this as both bull and bear. Use it for intraday trading and longer term trading.
I'm always looking for the same trade where I long/short at the 76 retracement and I put a stop behind the 86 or last high/low.
Here's the maths on why this makes sense and how I think it gives me a reliable edge over enough trades.
When entering 3/4 of the way through a swing, I always have the same RR. I have to risk 1/4 of the swing and if I target a retest of the last high/low I have 3/4 of the swing to gain.
Which means my default min RR is 1:3. Meaning I have a breakeven win rate requirement of 34%.
That covers the first, and most critical, part. I've defined a way I can routinely attempt to make money in the market and have a fair reason to believe I won't go broke. 34% win rate is very attainable.
From here, I can start to work out my edge. To do this I need to work out my expected win rate. I find depending on the conditions I use this I can target a spike out of the high/low, improve my RR to 1:4 and I can expect to win this 40% of the time against a trend and about 60% of the time with the trend.
So let's work out the worst odds and to keep it simple we'll say I risk $1 in a position. Here's the expected value (EV) over 100 trades:
100 trades taken. 60 trades lost (60*$1 = $60). 40 trades won (40* $4 = $160)
$160 profit. Take away $60 losses and I have $100 net profit.
To obtain this, I've had to bet $1 100 times. Meaning my net spend on bets is $100 and my net return on this is 100% of what I bet.
From there, do you think I care what trades win and lose? Get sad when the 60 losses hit and think I'm Einstein of the markets on the 40 that win? Nope!
I've worked out what my expected value is. I know by the law of large numbers I have a fair expectation of getting this and the only requirement of me is to make this bet enough times (Velocity of money) for the mathematic edge to play out.
I am using fib levels which mean my RR is always the exact same. I might be trading a retracement of a 20 pips swing in EURUSD. A retracement of a 100 points swing in SPX. Retracement in a 70% drop in BTC. Doesn't matter. The amount of distance to my stop loss and distance to my target is always the same. It's ratio based. Symmetrical RR.
The only thing left for me to do is make sure I have a uniform bet size by trading larger positions in less volatile assets and trading smaller ones in more volatile assets so all moves against me "Cost" the same.
This is one of my bread and butter strategies. If you've followed me day trading you'll have seen me execute on this literally 100s of times. You'll have seen times it went absolutely perfectly and times it looked utterly stupid because I faded mega momentum and it ran through - but to me, they're all just the same trade.
These are discussions on trading I am more interested in. I'm don't care much for, "My analysis is bigger than your analysis" kinda stuff.
I like to talk about how we can accept we do not know the future but have strategies that give us a good expectation of being profitable whatever it may be.
Comment
⋅
Sadly, the comments are mostly one guy who thinks they know the future and does everything they can to avoid framing their ideas in a testable strategy.
Sad times in the market.
Name of the website should be changed to ClairvoyantView.
@QRVi "last time under 1k .. sure" I didnt say 1k, I said 800$. If It's going bellow 600-800$ im going allin... You guys live in different universe... maybe on Mars. lmao
Couldnt reply thread keeps crashing.
holeyprofit
⋅
@citsvar, Would you like me to make a post titled, "Do you have any random opinion on anything?"
Good read! I enjoyed this thoroughly. my edge is definitely similar to yours. Newbie traders have a hard time understanding that trading is not about getting it right 100% of the time. they think trading is about predicting the market perfectly and they don't understand that trading is actually just "managing your losses."
holeyprofit
⋅
@handyrams8, People coming into markets these days are taught to "Have conviction". Rather than have a viable plan to deal with all market outcomes.
This becomes very obvious when you present a strategy based plan and people act like if it's anything other than perfect you're a complete failure and they're not interested in discussing the core theories as to why that trade is expected to be profitable over enough samples.
molbioinfo
⋅
Very interesting, thanks! Genuine question, I hope, how does this strategy compare with a trading bot? I looks simple to program, I'd believe. Do you think that you'd perform better than a bot, or if not, would you replace it by one? I have only gone for crypto, since 2019, but it is a side thing, and I do not consider myself a trader. Perhaps investor, if currently we can call crypto "investing". My most successful strategy in crypto, and what I do now, is buy low, sell high. If it goes lower, unless it is a critical level, I let go as my order sizes are 0.25-1% of portfolio, so I build a position by averaging. Then when it goes higher, I start selling, and always in week+ timeframes. I define the "low" by oversold RSI and MFI, and the "high" by either hitting a fib level or a previous high in a downtrend, and by looking at overbought RSI and MFI, especially when there are divergences. I also stay out of leverage, with the exception being one isolated account with 0.5% of my portfolio. So an example of my take in BTC is this. So far it's gone further up than I thought it would go (I did not predict the future spot-on!), but I think that structures in the RSI are a good message of what can be going on alongside price.
> compare with a trading bot? I looks simple to program, I'd believe.
The issue is the anchor swing for the fibs. Using default indicators (Zig zag etc) will pick up most of the swings I'd trade but they also pick up some noise ones I'd consider -EV.
>Do you think that you'd perform better than a bot,
I have this strat automated but manual performance is better to date. I've got a hybrid semi automated version of this where I draw the fibs and tell it to execute the 76 trade plan.
This is a general theme with my fib based strategies (Of which I have many). They lend themselves to automation well but there's issues isolating the correct anchor swings. A lot of them I have hybrid bots that I manually draw the fib and they're programmed to execute a set of trading rules.
>by either hitting a fib level or a previous high in a downtrend,
You can use extensions to project into the future, too. For example here for bullish forward looking levels in BTC you could draw a extension fib from the high to the low of the BTC drop and these levels are good future resistance zones. Works really well for longer term trades. Trail stops as we get close to the resistance fibs in case we reverse off them and if they break trail stops behind them so you stay in if they act as support.
molbioinfo
⋅
@holeyprofit, ahh, I thought that pinpointing the anchors would be the key, after I posted. You can smooth the anchors using EMAs and then find minima, but still I agree, especially if you focus on intraday and lower, where volatility rises so much. I liked the idea of trailing stops, used it for a while, but found the wild price swings would burn those stops. That was frustrating, and with time I realised that frustration is not the way to go... now my mindset is to grow portfolio without frustration, because frustration usually made the capital smaller. 2019 was a year of lessons, 2021-2022 more lessons, and now I'm building on those lessons step by step. Thanks also for the suggestion on the fib retracements, I was looking at your analyses too to see how they matched, and like the strategy. Also, if it means anything, I find it amazing how many people respond to posts just with "you are wrong, you are right, you will lose, yes we'll win" and stuff like that. I leave that aside and discuss the data, happy to be proven wrong.
holeyprofit
⋅
@molbioinfo, Writing in the exceptions is the issue for bots because they have no common sense at all.
For eg, I know if price randomly gaps intra day that's anomalous and I probably want to wait and see what happens next. But the bot, doesn't care at all. So I need to write in a rule "If this bar opens more than X points from the last bar closed, ignore signals for X minutes".
This is the thing standing in the way of fully automating these in true set and forget form. They'll generally do well but then maybe get hammered in funky conditions I'd not trade if I was doing it.
>was frustrating
Yeah trailing stops is frustrating because by default we always seem to want to place them right where the spike zone is lol. It's viable but a lot of work and I understand your perspective of avoiding the frustration.
>I find it amazing how many people respond to posts just with "you are wrong, you are right
Usually newbies. Generally in the 1 - 3rd yr people are like this. They think they know everything. Have had a tiny bit of success and think that extrapolates to being able to tell anyone anything.
You tend to find it's been kicked out of them by yr 6 or they've left the market.
There's a lot of it right now because in the last yrs we got a lot of new people into the market. Most of them thinking they're very clever.
> aside and discuss the data, happy to be proven wrong.
This is the way. All ideas should be preliminary and subject to ongoing review. Opinion are for the dinner table. Some sort of quantifiable data is for trading. For me, that's using fibs to define math edges. Everyone has their own thing.
molbioinfo
⋅
@holeyprofit, just realized that 76 and 86% are not fibonacci ratios, how did you come across their significance for your strategy?
Couldnt reply thread keeps crashing.