Read on Twitter
ok so i want to expand on this. the target audience of this thread is not quants (hopefully) or seasoned participants. this is a thread for people who are maybe new to investing or don’t have a lot of mathematical sophistication or experience with these situations https://twitter.com/NewRiverInvest/status/1356036042659299328
a lot of the time you see people, me included, show price or total-return ratios of two securities a lot of the time its price of A divided by price of B or a chart of the cumulative sum of log(A_0 / A_-1) - log(B_0 / B_-1) where x_0 is today and x_-1 is previous period (yday)
to replicate this return you would have to have equal amounts of each position, usually in dollar-terms, at the beginning of each period. that means for any non-zero move in the pair you need to adjust positions at end of every period by shorting more or buying more of one side
you can do that, but it means that, when the trade goes against you, you have to take some losses on the short leg, add to the long leg, or a mix of both. this introduces differences in return—colloquially, slippage, if reducing position size—or rising exposure if increasing size
there’s also, to a smaller extent, frictions (costs) from those adjustment trades and from changing margin requirements and financing costs. but the main issue is position size adjustment
because these trades are usually aimed at mean-reversion, a lot of newer participants risk falling into loss-aversion-bias and their instinct is to increase the losing leg when the ratio moves against them. this is *super dangerous* behavior. it models like being short an option.
it’s easy to fall into the trap (and unethical people will try to lead you down this path) that your trade is “safer” because you’re not betting on A or B direction, just the relative (out- or under-)performance of the pair. the reality is more nominal exposure = more risk.
this is especially dangerous in pairs that are in a trend: if you bought $1k A + sold $1k B with a $1000 account and it goes against you 1% everyday for 10 days and you hedge end of day by adding to losing leg you are down 10.4% + transaction fees and your position is now $1010.4
so your original trade went from being “neutral” at 100% of your account value to being “neutral” at 123% of your $895 account value and your gross leverage went from 2:1 to 2.26:1. if you don’t watch yourself your account equity could fall as leverage increases & margin-call you
so, these ratio charts are a useful short-hand for seasoned investors that understand these trade-offs, but they should not be used by people who don’t understand those trade-offs to enter into trades where the recent volatility of the ratio is high and it is trending
theres no trading advice to be had, its an open-ended question. i, personally, think a ~generally resilient approach is to keep the ratio of your gross-position size as a portion of your account equity is ~fine, but you may be ultimately right and lose money sometimes.
someone else can QT this thread & explain how this ~kinda mimics the delta-hedging on a short option, but imho, its more complex than the target-audience i’m aiming for in this thread. if someone does that, i’ll RT it, but it’s not what i tried to do here. be careful, make money.
