We hope that the five hot trading topics that we covered in our debut from the Contrarian Trader - including curve fitting, correlations and retail traders - got a few of you thinking.
To further the conversation, we reached out to previous TraderLife interviewees to ask them to weigh in with their opinion. And, as we expected, they’ve provided plenty of thought-provoking nuggets. Whether you agree with them or not, well, we'll let you consider that...
1. Curve Fitting
…my general view is that you can't really curve fit a strategy, but it is very easy to misunderstand what a back test is really telling you. However, if you fully understand what the in-sample results tell you, it’s possible to work out what is likely to happen in the future with the strategy. The problem is not that people use in-sample data, it’s that they don't understand it or make quick assumptions with it that are frequently wrong.Contrarian Trader
If you do enough backtesting, enough optimizing and employ enough rules, you can fairly easily make a trading strategy “fit” historical data. A well fit backtest does not guarantee good future performance.
I've never seen a formal definition of 'curve fitting' (a phrase I never came across as a hedge fund manager or professional trader, since it only seems to be used by retail traders) so I'm going to make my own one up: "Fitting the parameters of a trading strategy with pure in-sample data". Based on this definition, I'd say curve fitting is really a terrible idea.
Firstly, it's cheating. If your data starts in 1990, but the set of parameters you're using relies on knowing about what happened in 1991,1992...2020, then you wouldn't have been able to earn the returns your backtest says you earned in 1990.
Secondly, it's dangerous. The more closely your strategy is fitted to past data, the more brittle and less robust it will be. You will be tempted to tune your strategy (in the jargon, 'use more degrees of freedom') as closely as possible to the data since it will always improve your backtest performance. Unless the market behaves exactly like it behaved in the past, you will lose money when you trade with real money.
To take an extreme example, if you fitted your strategy to 2019 data with sufficient degrees of freedom then your trading rule might be something like 'Buy the S&P 500 on March 3rd because the market always goes up. Sell on March 4th...'. Clearly, it's unlikely the market will always go up on March 3rd and go down on March 4th!
To be clear, this doesn't mean that all fitting is bad. Fitting a strategy with robust techniques (eg Bayesian or Bootstrapping), using only out of sample data, is fine if done correctly.
Correlations…I think are generally impossible to be accurate on in finance. I understand you can break strategies into underlying parts to get factors to then work out correlations. But the fact is that correlations change so frequently over time and that things can look correlated but aren't.
Part of the reason for this is that finance data is pretty short in terms of history and is also dynamic. In these cases how do you measure correlation mathematically and consistently?
A good example of the problem is that if I toss two coins a million times, at some point their results will be positively correlated and at others they will be negatively correlated. They should, over a long enough time, be zero correlated but on different periods they won't be. But does that mean there is a correlation or not? Consider the same problem on two independent strategies, is it coincidence they are operating together or a market structure?Contrarian Trader
Correlations do change over time, but they are actually fairly predictable. In fact they are orders of magnitude more predictable than expected returns, or risk adjusted returns (volatility also changes over time, but is even more predictable than correlations). Using the last six months of returns you can predict correlations across market returns quite accurately. The correlation of trading strategies is even more stable, and you can use decades of data (if you have it!) to estimate correlations.
Being able to predict correlations makes it easier to construct a diversified portfolio of assets, or diversified set of trading strategies. It is much easier to improve your expected returns through diversifying than through any other method. So it would be a shame to ignore this powerful tool because of a misconceived opinion
3. Technical Analysis
Technical analysis and trading is absolute hogwash, it’s open to interpretation and personal bias, to the point where one trader thinks it’s a sell and another thinks it’s a buy….and just because price has bounced from a certain level before doesn’t mean it will react the same way when it reaches the same point again.Contrarian Trader
It depends on what you mean by technical analysis. If you mean the cornucopia of funny patterns and coloured lines that can be found on many a trading screen, then yes. There is absolutely no reason why something like a Fibbonaci level should be profitable (except through being a self-fulfilling prophecy). But an alternative definition of technical analysis is any trading rule that only uses price as an input.
Under this wider definition there are plenty of technical analysis rules that make a lot of sense; ones that are objective rather than subjective, and which pick up on well-known sources of return such as momentum or carry. There are many successful systematic hedge funds using technical analysis of this type, far too many for it to be hogwash.
4. Efficient Market Hypothesis
The Efficient Market Hypothesis is nonsense. The market isn’t remotely efficient. The market is not statistical in nature as humans mess it up by playing with it every day. I like the poker analogies for this. If you put six simply programmed efficient computers playing poker, then the outcome is simply the computer with the best hand, as it will base the entire programme on probabilities. Stick six people round a table and typically it is the person who can lie the best. The market has way too many people messing it up to be efficient …Contrarian Trader
My opinion on the Efficient Market Hypothesis is the same as that of every active trader, that it is just a hypothesis, and it has been proven wrong by Renaissance Technologies, D.E. Shaw, Citadel, and countless other hedge funds and high frequency trading firms. Some of these firms didn't have a single losing quarter in the last decade.
The EMH states that share prices reflect all available information. This is probably fairly close to the truth; and indeed in the age of 24 hour news, the internet and social media, is probably truer than ever. A frequently mentioned corollary of this is that it is impossible to beat the market. In fact the EMH says it is impossible to generate excess risk-adjusted returns.
This means it is perfectly possible to 'beat the market' (make more money, on average, than the stock market does every year) even if markets are efficient. We can do this by taking on risks that the market is willing to pay us for, such as liquidity risk (earned through market making), factor risk (earned through exposure to momentum or value), or using leverage (which many investors are unable or unwilling to use).
Let's take Warren Buffet as an example, since he is often offered as a counter example to the efficient markets hypothesis. A 2012 research paper showed that much of his market-beating return was down to the use of leverage and exposure to well-known risk factors.
Overall I'd say that the market is 99.9% efficient, since there are still a few people who can earn exceptional returns which can't be explained by exposure to risks. But the EMH certainly isn't nonsense.
5. Retail Traders
With less information and worse technology than the ‘professional’ institutions who trade, retail traders simply can’t beat the markets. They can only beat other - less focused, less prepared, more naive - retail traders…Contrarian Trader
I agree with most of the points, I’m not sure if that makes me very controversial or not at all! The only one I disagree with is the retail trader one; I think most institutionals are wrong just as often. Institutional investors can have all the tools in the world but the markets don't always make rational sense. The recent outperformance of the Robinhood investors has shown they do surprisingly well. Even if they're more informed they're wrong just as often.
Retail traders can beat the markets (assuming by 'the markets' we mean a long only investment in a diversified equity ETF), but only if they do some simple things well and avoid stupid mistakes. The simple things they need to do well is gain diversification across different asset classes and sources of risk premia. Risk premia such as momentum and value can be harvested using simple trading systems.
This will allow retail traders to beat the market, which only exposes you to a limited set of risk premia, and isn't necessarily diversified. As for mistakes: retail traders pay higher costs, so they should be careful about trading too often. They are also prone to using too much leverage.
None of this means that they will beat the professionals, since their slim advantages (such as limited market impact) are far outweighed by the disadvantages of being a retail trader, but it does mean they should be able to beat a passive investment in the market.
The view that retail traders cannot beat the market is I believe an elitist, misinformed opinion unfortunately held by many. To begin with, it's factually incorrect, because of course there are a large number of retail traders that profit consistently from the market on the 'A Book', else the retail market wouldn't exist.
I think the misconception stems from the wide range of traders that fall under the retail category. You have your punters who've just completed the BabyPips course, deposit £300 into an account and trade for a few months before their account is blown. On the other end of the spectrum you have traders with double digit years of experience, who've spent an enormous amount of time developing strategies and risk management principles. Some traders even come from the professional space and then make their own way in the retail space, this is rather common.
It is absolutely true that as a retail trader, the deck is stacked against you. It is a fact that the vast majority of retail traders lose money over time, but that doesn't mean that you can't be profitable as a retail trader. There are also advantages to retail trading that institutions cannot benefit from, the biggest I believe being anonymity.
When a retail trader buys or sells a market, the world keeps spinning and it seems as if nothing has occurred. But, if an institution wants to buy into EURUSD for example, they can't just 'buy' EURUSD. They face issues with sourcing liquidity at ideal prices, slippage, front running etc etc etc. Retail traders can hop in and out of markets, as the same applies to selling what you've just bought, at a whim without facing anywhere near as many difficulties.
So to conclude, as I could go on forever, not only is it factually incorrect to say that retail traders cannot beat the market, but also shows a misunderstanding of how broad the retail market is. Most retail will never make a penny, but many will, and so as long as those traders exist then retail traders can beat the market.