Ask HN: How to use technical analysis for long term investing?

[kpr21]

I am not a fan of trading. I look at the fundamentals and buy stocks with the intent to hold for long term. One of the mistakes I make is buy without looking at the technical analysis. This can incur considerable opportunity loss or real loss. So I am planning to apply technical analysis to fundamental analysis. Here is my thought process.

1. Shortlist the stocks based on the fundamentals, future potential, moat etc. I will apply technical analysis only to this list. Wont buy anything outside this list, even if the technical analysis says its is a screaming buy.

2. I am very bad at technical analysis. But we get the summary verdict (sell/buy/neutral) based on oscillators and momentum indicators easily in several sites. I will use this as the input. The plan is to use 1 day interval.

3. For sell/hold decisions, I wont rely on technical analysis verdict. ( Unless there are some fundamental based issues).

4. For buy/entry decisions, rely on the technical analysis verdict. This is the toughest part. Because it may not factor in the trend reversal. So we may end up buying at the peak. To overcome this, what are some of the key parameters to give more importance to? ( I mean among the momentum and oscillator parameters used in the verdict, such as moving average, RSI, ADX etc)

Can someone help me with the point 4?

[kjellsbells]

You may not want to hear this, but technical analysis of /price data/ is a fools errand for a number of reasons. There are the folks with PhDs who dream in C++ about martingales but even they dont consistently beat the index. More fundamentally, as a retail investor, you cant get better arbitrage than the pros can since they know 10x more than you do (I dont believe that the protections and firewalls between banking and broking are anything more than useful fictions, which means brokers and market makers always know more than you do).

I do think it is worth using technical skills to audit non price data however. Much less sexy but more valuable. P/E, FCF and all that.

[kpr21]

Thanks. Very well said. I agree with you. I am actually passionate about non price data like cash flows, balance sheet, income statement, PE, EV/EBIDTA, etc. But I have seen a lot of long term fundamental based investors talking about using technical analysis (price data) to get a better entry point. In that context, I wanted to try technical analysis. To be honest, I dont like technical analysis. Find it very boring.

[asr21]

Indian here, so I won't know if this works on NASDAQ or US stock market in general. Also, I have traded only a short time in bear market. Most of my trading is in Bull Market. So take my input with a truck load of salt .

The thing worked for me, is the simple concept of Support and Resistance and Dow theory ( Higher high, Higher low all that shit.) . So let's say there is a certain point or a range after which , it usually bounces back . That's the support and I would buy at that range.

One more logic here was, let's say a stock is trading at 50. It's support is at 48. I can afford to lose 100$ in this trade. So I would get 50 shares at 50$. If it goes below 48. I will take my loss of 100$ . Exit that position. It's more about calculated gambling

[kpr21]

Thanks for the good input. Glad it worked for you. I had read a little bit on support/resistance, higher high/low theory etc. But I find them very boring. ( In fact everything around technical analysis). As I said, I am not looking at trading. The idea was to use technicals for a better entry price.

[YuriNiyazov]

Having gone down a similar path about a decade ago, let me short circuit this for you.

The consensus seems to be that technical analysis doesn't work. However, we don't have to commit to that very categorical statement, and we can work within a much weaker framework:

Even if various aspects of technical analysis could work in theory, they are too hard to get right for a single trader who is trying to wrangle it alone on his laptop. If it worked, then to get it right you still need a team of PhDs, and infrastructure, and computing clusers, etc. etc. So, forget it.

If you are worried about paying too much, or you are worried that you will miss out an opportunity, then learn how to use options to solve those problems for you.

[kpr21]

Thanks for the input. Nope. I never really liked options etc

[ALittleLight]

What evidence have you seen that makes you think "technical analysis" works?

[kpr21]

No evidence. But I have seen a lot of fundamental based investors talking about using technicals to get a better entry point.

[larrykubin]

I've studied many of these indicators. Honestly, my favorite is to look at the new 52 week high list rather than trying to time reversals.

[kpr21]

Thanks. Thats interesting. I have seen a lot of people follow this strategy. The problem is, what if the new high is the top?

[fakeElonMusk]

I did a fair amount of technical analysis trying to learn day trading a few years ago. I've also bought and sold stocks based on fundamentals, trends or hunches. I lost money and I made money. The best returns I had were from buying stocks in 2008/2009 during the crash (duh) and holding.

My approach now for long term investing is to auto invest monthly into 5-6 index funds w low fees, aka Bogelheads. Simple, low stress. Dollar cost average on auto-pilot.

Also, set aside a nice chunk of cash to invest during the next crash.

[deanmoriarty]

> Also, set aside a nice chunk of cash to invest during the next crash.

Do you have any evidence this is actually recommended? If I leave $100k on the sidelines for 5 years and magically invest at the exact bottom of a 50% crash and double my money when it rebounds, how is that any different than having invested those $100k in the first place? At a 10-15% return, that’s basically the same returns (we can play with the 10-15% and 50% crash numbers, but my point stands).

[Cicero22]

My understanding is that time in the market beats timing the market.

somewhat relevant video: https://www.youtube.com/watch?v=w_aOERmUWdA

His videos generally are high quality and offer good advice, although some are tilted towards a Canadian point of view in terms of taxes.

[fakeElonMusk]

I have zero evidence that it's recommended or out performs your scenario. I do have one data point of it working well for me, so I'm planning on doing it again.

[kpr21]

Thanks for the input

[nmhancoc]

I haven’t had (or seen anyone else have) success with TA, but if you’re looking to improve your entry into existing positions why not try selling cash covered puts.

The premium for short-dated puts gives you some bounds about which the options market thinks the underlying will move, and I’ve been able to improve my entries in effected issues by 5-8% in terms of price (before fees for options).

It’s effectively a limit order that you can’t cancel, with potential upside if the puts expire worthless.

[kpr21]

Thanks for input. I want to keep things simple. So not looking at puts etc

[slapslash]

Oh yeah,trying this since at least a decade now. Western indicators,eastern indicators and my own reprensentation… brute force, kind of all machine learning algorithms, genetical algo… it‘s a mess and i might be some kind of obsessed by it. By now,i‘m nearly convinced,that it‘s not possible to predict Stocks for normal persons /with known indicators. BUT at least i got a quite good programmer by trying it aaand… who knows… ;)

[kpr21]

Thanks. You are right. It is quite complicated.

[nmz91]

I buy at any price as long as it is below what I calculated as fair price. However, I wait for it to bottom out if the trend is in the downward direction.

P.S. I have never regretted that I've missed out on 1% return as long as the return is in my range of 18+%

[kpr21]

Thanks. I think fair price is subjective. Gone were the days stocks had reasonable valuations.