Idiot Question - Help me understand risk
Afaik risk is determined by s.d. I don't understand this - surely momentary swings aren't important when taking concentrated bets?
An (unrealistic) example: I know AAPL is up 10% by EOM. It goes down 50%, back up 120% for a 10% net. The recorded risk (using s.d.) is insane, but in reality, there wasn't any risk because I knew AAPL was going up.
When evaluating risk/reward (say via Sharpe), particularly as an LP, surely this way of thinking is flawed?
Based on the most helpful WSO content, here's what you need to know:
Risk and Standard Deviation:
Criticism of Standard Deviation:
Alternative Views on Risk:
Real-World Examples:
Sharpe Ratio and Risk/Reward:
In summary, while standard deviation is a common measure of risk, it has significant limitations. A more comprehensive approach considers business fundamentals, economic developments, and the potential for unforeseen events.
Sources: Is the common view of financial risk completely wrong?, PE professional, what's your process while judging an investment?, Understanding Concensus & Differentiated View, Difficulty Check, How various professionals see the world
You're correct that standard deviation is commonly used as a measure of risk in finance. It quantifies the dispersion of returns around the mean, giving an idea of volatility. Your example highlights a key limitation of using SD as the sole risk measure. SD doesn't distinguish between upside and downside volatility, and it doesn't account for an investor's specific time horizon or knowledge. That's why we use sharpe to define risk adjusted returns. So we value stable low vol returns more then the same return under a more volatile holding. High volatility, even if temporary, can lead to poor decision-making by investors who may panic and sell at the wrong time.
In your AAPL example, if you genuinely knew the EOM outcome with certainty, then you're right - the interim volatility wouldn't matter to you. However, this scenario is highly unrealistic in real-world investing. We never know future outcomes with certainty. Even if we're confident in a long-term trend, short-term volatility can force us out of positions (due to margin calls, fund redemptions, etc.).
You gotta remember that risk can mean different things to different investors:
That's why we need to use additional risk measures alongside SD:
LPs often care more about downside protection and consistent returns rather than short-term volatility. This is why many hedge funds and PE firms report metrics like maximum drawdown alongside Sharpe ratios.
If you're a value based fund with long-term holding who "arbs" volatility to buy whilst others are being stopped out / selling (increasingly common phenomenon given the rise of multi-manager flows in the market), then you can argue that risk is not synonymous with volatility, it's the possibility of permanent loss.
Markets are increasingly driven by non-stock specific factors. As such LPs care about how you earned your alpha. So vol matters. It's not the perfect way to measure risk, but it's 'a way'. Risk mgmt is an uncompleted subject where we're still innovating and discovering new ways to perfect it to changing market regimes.
Suggest reading Risk Revisited and Risk Revisted Again by Howard Marks for the value based approach to risk management.
Different sectors and stocks have different market participants. As an investor it's important to have a dynamic view not wedded to academic theory to understand how various market participants react to risk, vol and probabilities of outcome.
The answer above me is great and obviously from someone with much better understanding than me but I thought I would add a more human reason why we use volatility as an approximation of risk. In your question you said you knew apple would go up so there was no risk, but in reality you will never have a guarantee of an outcome. Imagine you’ve remortgaged your house and made a huge levered bet on Apple. Day one you’re up 5% you’re golden, you’re king of the world, buy the misses a Porsche. The next day it tanks and the price is down 50% and it trades flat for a week. You are now hyperventilating in the bathroom thinking about telling your daughter you had to sell her horse for dog food and you’re all moving to the Bronx. A month later it’s traded back up a percent and you may have aged 10 years but you’re safe. Uncertainty causes a physiological reaction. Most people are risk minimizing economic actors and will actively avoid situations like this.
So when people decide on investments, let’s say two stocks - one undervalued by 10% and second undervalued by 40%. We should take into account volatility so, should calculate ex ante sharpe ratios before figuring out which to buy ? Maybe second stock is extremely volatile.
The return need to make up for the risk, yes. I don't think many think of sharpe on a individual position basis (at least I don't). I look at it in aggregate for my entire book. You'll have a risk budget. If you break VaR you usually need to let the risk team know why and how you're mitigating it. So if you're existing book is low vol, you could in theory afford a more risk. So you need to assess if that risk/reward is worth it versus other risk/reward opportunities in your idea pipeline.
You usually size for Vol and VaR. So if you have a very high vol stock, your sizing prohibits you from making it a big position (under the tight risk management frameworks). Vol is by definition backwards looking. So you need to differentiate between theoretical sharpe and realized sharpe / vol. But then we're getting theoretical. Some don't look at it this way at all. Some people just chase 4-10x baggers and don't size for vol. But they're a dying breed for a reason.
Worth noting that there's a stark difference between multimanager's perspective of risk vs your single managers. At MMs your payout ratio could have kickers from improved sharpe. Some platforms will give you extra bps on various sharpe intervals. E.g. +25 bps on your 20% payout for 1.5-1.75 sharpe. 50 bps 1.75-2.00 sharpe, etc.
Theoretical vol- is unobservable so you only have the historical to look at - couldn’t you look at option vol or is that still backward looking ?
Dolorem illo qui et libero repudiandae veniam et velit. Qui quam nemo ut quasi sed officiis a non. Tempore cumque ullam eum atque atque. Nulla ea et reiciendis laborum ut.
Nihil aspernatur aut molestiae dolorem molestiae sit. Quod et consequatur animi. Molestiae asperiores nihil sit non.
Officia eaque quibusdam voluptates ab. Veniam eum ullam porro et occaecati ipsam optio et. Rerum repellendus blanditiis eaque sunt consequatur. Ut aliquid earum sit asperiores.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...