managing correlated exposure across positions - how do experienced traders think about this
-
after several years i've started to realize that my worst drawdowns weren't from single bad trades but from multiple positions all going against me simultaneously because they were all effectively the same bet expressed in different pairs. long eurusd, long gbpusd, short usdchf all went south together.
how do experienced traders think about correlations when building a portfolio of open positions, and what frameworks actually work in practice?
-
the framework that works: think in terms of underlying risk drivers rather than pairs. usd strength/weakness is a driver. risk sentiment is a driver. commodity prices are a driver. when you have long eurusd, long gbpusd, short usdcad simultaneously, you're not three trades - you're a large short usd position. asking 'what's my total usd exposure' is the right question, not 'how many positions do i have'.
-
risk factor aggregation is the right mental model. i run a simple heat map in a spreadsheet: rows are open positions, columns are the main risk factors (usd, risk-on/off, commodity, rates). each cell is +1, -1, or 0. totaling the columns shows my net exposure to each factor. if a column is +4 i'm heavily betting on one direction of that factor without necessarily realizing it.
-
how do you handle the case where two positions are naturally uncorrelated but might become correlated during a risk-off event? they're independent in normal conditions but cluster in stress.
-
stress correlation is a known problem - correlations across assets tend toward 1 or -1 during extreme events. the practical way to handle it: regardless of normal correlations, reduce total position count before major scheduled risk events like fomc, nfp, or during obvious market stress periods. reducing size when correlation risk is elevated is more achievable than modeling dynamic correlations precisely.
-
after my own correlated drawdown experience i implemented a rule: maximum three open positions, and no two can be in the same underlying direction on the usd. simple and not perfectly optimal but it prevented me from ever having that kind of concentrated exposure again. imperfect rules consistently applied beat theoretically perfect rules inconsistently applied.
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register Login