streetpips
Publish date: Fri, 19 Dec 2014, 01:59 PM
[adapted from an email to clients]
December has been a tougher market for us to navigate so far. Trading is a probability game and string of losses is unavoidable in any probabilistic game. A game of dice can have a long string of coming up “Big” or “Small” even though the odds are 50-50. Diversification continues to be the most important concept in combining different strategies together and we continue to do so, researching profitable internally developed robots and more importantly, seeking out profitable traders and strategies out there. These are combined together.
We have learnt a few lessons from this drawdown and have tightened selection process of strategies. Let me illustrate below with 2 examples.
Trader X has historically done very well in his live trading over the last year and likes to trade exotic currency pairs occasionally. We were fine with that given his good track record. The Ruble has collapsed 50% recently and hence likely oversold. This trader bet a few trades on the Ruble rebounding. This turned out negatively and Ruble continues to fall. This trade had a huge impact on Trader X’s own account. As we are multi-strategy, this trade does not impact our portfolio as much although it is the single most negative trade we had so far, at -3.8% of portfolio. We have since tightened our selection process and now take a strategy/trader’s tendency to trade exotics as a negative.
Trader Y has also done very well trading manually for at least a few years and has a few positions on the Yen recently. However he started holding on to his losers which we spotted. We then removed this trader from our portfolio and realised his losses. A total of 5 yen trades with combined impact of -3.93% on our portfolio. Today this particular trader is still holding on to these trades in his own account. We moved on. While the Yen could go either way subsequently, the reason we stopped this strategy and moved on was due to the lack of discipline we have seen. The impact of these trades to trader Y’s account is above -20% while the impact to us is -3.93% due to our multi strategy approach. For evaluation of external strategies previously, if the stop loss was indirect, meaning it wasn’t entered at trade entry and trader showed track record with cutting losses, we were fine with it. In some cases some traders do not input stop losses to avoid stop loss hunting by brokers. Going forward we decided to view negatively, all strategies without direct stop losses entered at trade entry. This helps to mitigate the effect of traders/strategies holding on to losses far too long.
We have also tightened up on some other aspects in the strategy selection process apart from the 2 examples illustrated above.
Apart from these we continue the trading process as per plan. Drawdowns are painful but it is part and parcel of ANY investment. Famed fund manager Meb Faber had a good piece that came to my desk recently, addressing the fact that 78% of the time traders and investors are in drawdown mode. (Read at http://mebfaber.com/2014/12/12/be-a-good-loser/) What’s important for us and for clients, is to improve the plan and stick to the plan.
I wish you a Merry Christmas and Happy New Year ahead!
[Addition para to email]
Sometimes I get asked this question in a few different variations: “Since you diversified with multiple different strategies (e.g. trend, reversion, technical-driven etc), why is it still possible to face drawdowns?”
Assuming all sub strategies are profitable strategies, there still exist chances of majority of the strategies facing drawdowns together. The lower the correlations of performance between the strategies the better the effect of combining them. Low correlation tells us that over the mid/long run performance is extracted at different times. However low correlation is not NO correlation hence sometimes due to randomness strategies can still produce losses or wins together. Correlations do change as well and we monitor changing correlations on a monthly basis.
Strategies sometimes might lose money at simultaneous times in a way that is not measured properly by the correlation matrix. We examine using a Coincident Negative Return Matrix to complement correlation analysis. This was also mentioned in a recent book I read, “Market Sense and Nonsense by Jack Schwager, author of Market Wizards series.
Markets are always challenging and we continue to work on shifting some odds in our favour.
Discussions
Be the first to like this. Showing 0 of 0 comments

Post a Comment