Management summary:
“Navel gazing” alert: This post doesn’t contain any actionable investment ideas but rather explores how I can enrich my own investment process in the future by incoprorating some measures of Stock price and fundamental momentum.
Excursion: My secret hobby
First I have to admit that for a few months now, I do have a secret hobby: I am watching on a regular basis a Wikifolio (Wikifolio is a German/Austrian platform where everyone can set up a “fund” and other investors can participate) from an Austrian trader with the name Richard “”Ritschy” Dobensberger.
Not only has he managed to attract 160 mn EUR in investments into his portfolio but he has averaged 33% CAGR over the last 13 years, resulting in an overall performance of around 4000% which is really really remarkable and puts him into the top of any trader I know.
Ritschy’s strategy is relatively simple: He has a universe of a few dozen well known, relatively volatile/high beta stocks and buys them when they seem go up. If they continue to go up, he keeps them or even adds, if they go down he sells them extremely disciplined.
Once in a Podcast he said something along the lines: “It’s like in football. A football coach selects the players that are currently in great shape, not the ones who are out of shape”.
Not every trade works, but those that work well (like Rheinmetall) move the portfolio big time.
To give Ritschy some credit, although it sounds simple, it is clearly not that easy to execute, but it clearly shows one thing: Momentum as a factor works quite well, especially since around Covid.
Don’t worry, I won’t turn into a momentum investor anytime soon because I think I don’t have the mental set up to run such a strategy, but I think I have ignored stock price momentum in my investment process for too long.
Ignoring momentum so far despite some noble intents
For the longest path of my career I have either ignored momentum or actually invested against (negative) momentum. In the past, this has overall worked quite well, but I think I left a lot of return on the way.
I had pondered introducing momentum into my investing process several times. Here for instance is a dedicated post from 2012 (13 years ago !!!).
This was my summary back then:
That was a decent insight, but unfortunately I never followed up. I rather did the opposite, such as documented in this post from 2016:
So after pondering that I was always selling too early, I sold the GTT position which became a multi-bagger (~5x) and reinvested into a stock that turned out to be a value trap.
Why didn’t I follow up on it ? To be honest, I do not know for sure but the main reason is most likely that I outperformed my benchmark anyway for another 6 years until 2019. Why change a system that works ?
However, including 2025 YTD, I have now underperformed in 4 out of the last 7 years.
The current market seems to be extremely momentum driven, which clearly is one factor of the recent underperformance of my portfolio as I have ignored it maybe for too long.
Weaknesses in my current process:
Looking at my more recent actions, I identified the following issues:
- risk of ending up in value traps
- adding mostly to positions on the way down
- not adding to position that work well
- selling too early
- wrong prioritization of watchlist by only focusing on “cheapness based on historic numbers)
- missing out on a diversification angle
.
What does academia / statistics say
There is a lot of evidence that momentum is a strong “factor” in explaining stock returns and especially “alpha”. i.e. positive outperformance.
Here is a summary table generated form ChatGPT when I asked about the 10 most important studies:
What measure exactly is usually used as a proxy for momentum ?
The “Quant literature” usually mentions 6 month or 12 month momentum, often in the form of “6 month -1 momentum” or “12 month -1 momentum” which excludes the most recent month, i.e. looking at the 6 or 12 Month performance 1 month ago.
Why is this ? It seems that the most recent month is statistically “noise” or even negatively correlated with subsequent stock returns. So ignoring the returns of the last month in determining momentum seems to improve results in these studies.
There was a recent interesting post on Klement on Investing that showed that using both time periods, i.e. 6 & 12 months momentum seem to be even better.
Depending on the study, positive momentum is then confirmed if the 6M or 12M price return is either positive or positive AND greater than the risk free rate of return. Most studies than invest into the best decile momentum stocks and short the bottom decile of the stocks with the worst momentum.
Time horizons
Typical momentum strategies require quite frequent rebalancing in order to achieve their alpha, which is clearly not my goal.
Under German Tax law, frequently realizing gains is also not the best strategy to maximise after tax returns. This aspect is often not covered in academic studies.
To be honest, even if a mechanical system would yield better results, I still enjoy being a stock picker and I am actually prepared to sacrifice some performance for the joy of analyzing single companies.
Nevertheless I think I can improve my process by including some aspects of stock price momentum.
How to include stock price momentum into my investment process going forward
As mentioned in my Q2 Performance review, I want to include stock price momentum on a more systematic basis into my investment process.
My main tool for this is a spreadsheet which around ~100 most interesting stocks (including all my portfolio holdings) that I will compare to each other based on quality (measured by some criteria), valuation (i.e. discount to my “fair value) and momentum.
The factors quality and valuation can reach a maximum of 14 points. Momentum gets accounted for in the following way:
I defined a momentum score that will be added to or subtracted from the total score.. The Momentum Score can go from -2 to +2 in increments of 1.
As a first step, I will assign a score of +1 if the 6-1 month performance is >5%, -1 if performance is <-5%, else 0
In a second step, I will add +1 if the 12-1 month performance is >5%, -1 if <5%, else zero.
So depending on momentum, a stock can get a max of +2 added or max -2 deducted. This score is clearly not based on rigorous back testing, it is more a kind of “gut feeling” and it serves one main purpose:
Assuming that all other aspects are equal (Quality, valuation) I will prioritize higher momentum stocks to lower momentum stocks. This applies to both, stocks I want to analyze more deeply and stocks I want to add to or that I want to sell for increasing liquidity.
Fundamental momentum
In addition, I also reflect the fundamental momentum in my qualitative score. I will give a point if the last observable EPS number went up, zero points otherwise (no negative values here).
This is how it looks in practice:
This is an abbreviated snapshot form my sheet that shows how this works in principles with just my portfolio companies:
I do not use this sheet to slavishly follow the ranking but rather as a starting point for further analysis. For instance, some people asked me if I would add to TFF. If I look at my model, adding to TFF is clearly not my top priority. Rather the opposite.
GESCO for instance doesn’t score that well at the moment, but I see some potential for future improvement. But still, I sized the position small as the score is not that good.
But I also would still add (cautiously) to a negative momentum stock like Novo Nordisk if quality and valuation seem to be attractive enough.
Am I 100% sure if this will improve my (relative) results ? I do not know, but I do think that adding this additional perspective could help me in the mid- to long term.
What ChatGPT thinks of my approach
I uploaded this document to ChatGPT (5) and asked it what it thinks about my approach. This is the result:
The overlay strategy that it suggested in the subsequent step however is too complicated for my simple stock picker mind. But the criticism as such is clearly valid.
Therefore I asked it for a simpler set of rules manage the issues which it provided:
To be honest, I found these rules quite helpful and will try to implement them going forward as well as it mirrors my own thinking quite nicely.
When asked about the risks of adding a momentum overlay to fundamental stock picking however, ChatGPT came up with a few points that are also worth considering:
Final thought:
While I don’t like to use LLM’s to create content, I find the conversation with these models often helpful if I ask them about the opposite case or risks. This really enriches an article in my opinion.
To be continued…..