Accuracy of Implied Volatility

Hello, everyone, and welcome. Brett Friedman is back, and he's back with more options-oriented discussions, and he's done some backtesting on implied volatility, and I'd like him to tell us what he came up with. Thank you. Thanks for having me again.

So this all began, I should give you a little background on what I've been doing with implied volatility. This began a couple of weeks ago when I started looking into the SpaceX IPO and how the options were going to be priced without any underlying ever had traded. uh so and i remembered back to when i first started trading options in oil that oil options started but oil had been trading for a while so we knew how to trade we knew how to price the oil options when they first went on the board because we had had years of data but with spacex we had nothing so how do you price an option on the first day of trading which was i think last tuesday without any underlying data that's meaningful.

So I started looking into basically how would you do this. So extrapolating or putting SpaceX aside for a second, one of the issues is, of course, that you're pricing the option with implied volatility, but the question that no one really asks is just how accurate is implied volatility? It's supposed to be the forecast of future variance, but how well does it do that forecast? So what I did was you take the implied volatility today and you compare it to the result, the next, like if you're doing 30-day volatility, you compare it to the next roughly 21 trading days, figure out the variance of the 21 trading days, and then compare it to the volatility that you had and see how well it does.

It's a basic back test, nothing too snappy. And that's essentially how a variance swap works anyway. And what I came up with is that implied volatility is usually overpriced, which isn't a giant shock. But I was a little surprised about the degree to which it was overpriced.

The degree to it was roughly, on average, 3.82 percentage points. And with the VIX trading in the teens, that's roughly a 20% markup on what's been going on. Interestingly, if you go backwards instead of forwards, so you calculate the last 21 trading days variance and compare it to today's volatility, so a backward-looking look versus a forward-looking look, you get roughly the same results, which is interesting.

which means that the premium that's in implied volatility is more or less consistent. But to me, I knew that there was a premium in implied volatility because you have to compensate sellers for left tail risk, et cetera, et cetera, et cetera. But I didn't really appreciate the degree to which implied volatility is overpriced. I mean, 4%, roughly 400 basis points, it's a pretty good premium for what you need to do.

And you also wrote something on how accurate is implied volatility not too long ago. Right, and that was the results of what I said. So almost all of the time, I think it was like I calculated, Yeah, roughly 83.7.

That's not roughly, that's exactly 83.7% of the time. Since 1996, implied volatility has been trading at a premium to what actually happened. And even on a backward-looking volatility, you see the same result.

So almost all the time, it's trading at a premium. When it goes negative, that is the results overshoot implied volatility. implied volatility was underpriced, it has a tendency to either not last very long. And if it does last very long, it gets extremely negative.

So the most negative, I think, was in the 60s during, I think it was 2007 or 2008. And the pandemic also had extremely negative numbers as well. That went on for a long time. But most of the time when it goes negative, it's just it's more of a blip than anything else.

It's a couple of days. But if it goes past a couple of days, it's kind of a signal that the risk regime has changed, that we're entering a completely different, a very volatile risk regime that implied volatility is not picking up yet. But it will. Well, I mean, the VIX was standing around 17 for the longest time recently, and it seems to be picking up now.

Is that I mean I think it was 19 this morning or so Yeah Yeah And you know if you go backwards if you look at the last 21 trading days and you calculate the variance and compare it to the VIX the VIX is actually what I calculated yesterday it was 8 over positive which usually means that the VIX is overpricing volatility that the market has been kind of quiet and yet VIX is holding up Now, why is VIX holding up? Why is VIX supposedly overpriced? My opinion that there's a lot of stuff going on that has the potential to really affect the market, but it hasn't really.

You have the situation in Iran, you got all this other stuff going on that could really affect the market so that there's this premium stuck in all the time to take that into account. And that's not unusual. If you look at the forward volatility, the forward risk premium is 4.2%, which is exactly more or less the median that it's always been.

So the numbers are telling me if you calculate the forward and backward risk premiums, depending on how you want to do it, whatever you want to do, the market, VIX is a little bit overpriced, in my opinion, compared to what's been going on. now maybe it's forecasting that something's going to go on but you know i can't tell the future and if i could tell the future i wouldn't be on this podcast so of course you would be but you're gonna you're gonna be doing nothing just counting and clipping coupons yes you gotta know you gotta do something trust me uh all right so this is interesting because what you're telling me is basically you feel that something is, there's a regime change then coming up or ongoing right now?

Actually not, Patrick. What I'm saying is that the VIX is overpriced. If the forward one became negative, I would say that there's a potential for a regime change. But so far, I'm just not, I'm not seeing that in the cards.

You know, something could, I mean, the regime, when the VIX, excuse me, when the volatility risk premium, i.e. the VIX minus the variance, goes negative, it does it pretty quickly. So something happens, implied volatility doesn't pick it up, and you see it go negative.

We're just not seeing that yet. Okay, well, I guess when you're seeing it, it's going to be too late anyway. Yeah, so the good news is the negative ones rarely last all that long, especially in today's market i mean if you go back in time and you look at the various things that people considered to be earth shattering they turned out not to be they turned out to be you know two three days the market forgot it forgot about it and went back to uh ai obsession so i mean that's one yeah did you see how quickly they forgot yeah it was first i mean a couple of days ago it was ai is over yeah and then and i think micron came up it's earning and then it was like okay you know what never mind let's keep on making money because i mean if the war the oil i mean do you believe i mean it's what when i said when i was looking at the vicks not going anywhere and i said but the the straight over moose was closed do you know it was never closed ever it being all the wars that went on down there and yet and yet it was like as if well yeah but it has to open so therefore there's nothing to worry too much about and therefore um so i mean what you're saying basically well what i was thinking was that if i had gone back in time like at two years yeah someone told me that we're going to be at war with around the strait of is going to be closed for for months long periods we're going to have to reserve deficits that we've never seen before.

And oil is going to be trading like around 90 bucks. I'd say you're out of your mind. We're going to see oil at like 150, $175, something insane. And yet that just didn't happen.

And, you know, I wrote something a couple of weeks ago for Option Metrics about, or yeah, Option Metrics, that, you know, there were a lot of people saying at the time, as this often happens, is that the market was wrong. They're arguing with the market, which is always the sign that your fundamentals are screwed up. There's something going on here that you don't understand. And there was another factor going on that the opinion was 100% that oil was underpriced when this was going on.

And 100% opinions always bother me. Something going on. If I had a trader who was long oil during this thing. And every single, you know, he makes money every now and then when it shoots over 100, but it's just not working out.

You know, it's drifting back down to 90. As a risk manager, I'm going up to him and saying, what's going on here? And he's saying, well, you know, it's going to pop, it's going to pop, it's going to pop. And as a risk manager, I say, well, first of all, we're tying up capital and it's not really working out all that well.

And B there something going on here that you don understand It become evident eventually but right now this trade just ain working And the only people who are long are people who are long in the ether They not actually long It's not going to affect their bank account or their personal livelihood if they're wrong. So, you know, these are like market commentators and social media people. They could care less if they're wrong. If you're trading, it's a big deal.

Yeah. But, you know, as it turned out, you know, there was some end-arounds that kind of plugged the hole for a little while, you know, and the supply situation was dire, but it wasn't as dire. If it had gone on for a couple more months, yes, maybe, you know, then we would have seen some real effects, but it didn't. So, anyway.

But yet, but yet, okay, so let's forget the straight, it's open, or it's down again. So oil is down $70 or $72. Yeah, it's in the low mids. In the low 70s now, right?

I think it's below now when this whole thing started. And yet the 10-year is still at 4.5%. And the market is, well, I don't know if it's up anymore because it was down and then today's still down.

It was more or less up. Let's put it this way. When oil was going down and down and down, the 10-year stays up and the stock market goes up. So shouldn't we?

I'll just be very blunt. The market doesn't give a shit about that. They care about AI. Right?

I mean, it's got to be it. AI is everything. and you know when the we have an incredible chip rally going on and that's driving the whole thing so you know i saw an interesting study the other day i'll send it to you if you want sure that they gave a whole bunch of traders the question is if you gave people today's headlines or yet uh today's headlines and without any prices on them but you gave them the headlines in advance how well would they do to trade? So they have the information, they have the news in advance, how well would they do?

They don't have any prices, but they're just basically betting on the headlines. So they're seeing oil, you know, whatever they're saying. And you know something? They didn't do all that well.

Because they didn't, for two reasons, they didn't gauge the importance of each headline relative to each other. So they didn't really gauge like how much is in the market, you know, like oil's down, like oil, like a couple of weeks ago, oil, you know, oil isn't going to go into great deficit, but it was going to go into great deficit yesterday and the day before that and the day before that and the day before that. So it was kind of built into the market and the market was like, yeah, I know this is what, what do you want me to do about it?

Exactly. So, uh, so it's not just the news. It's, it's, It's how are people interpreting that and what's important today? And the market is kind of like a toddler that what they were focusing on yesterday isn't what they're focusing on today for some reason.

There's a new toy. Yesterday's toy, they played with it for 24 hours, but now it's like at the foot of the steps going down to the basement. Today, it's something new. And the toy that they've been playing with for the last two years is AI, consistency.

And don't you think they're going to drop that toy and worry about the new toy, which is the inflation toy and wash and all that? This is one of these things where you get the commentary versus reality. Try being short AI. See how well it works out for you.

Say that, you know, it's going to crash. It's overvalued. It's overvalued. It's overvalued.

Well. Well, you know, it's not that overvalued. Those P ratios that I'm seeing now, when you compare it to 2000, some of the stuff was way bigger. We thought we were way higher than that.

And then they didn't have a business. Well, we have a business here. No, they have an actual business. They have incredible revenues.

They have some profits, some of them at least. And the stuff works. Yeah, it works. It works.

It works very well. Whereas you go back to 2000, which a lot of the people trading right now weren't trading then, or some of them weren't even alive, I think. uh you know there was a lot of incredible bullshit out in the market that made you know that was just you know vaporware so it's different uh and as low as i am to say it's different now but the ai boom is different than the than the dot-com boom it's different so we're gonna worry about i still remember a couple of weeks ago we had we were down three or four percent in one day because we had 172,000 jobs instead of the 85 that we were concerned about.

We say, wow, inflation. You're right. It's like a child. It's like a new toy.

I don't like it. But ultimately, and right now, I guess the question with Walsh and the Fed is basically saying that he's going to raise rates. He's going to raise rates. Okay, well, okay, fine.

And then what do you want him to, lower rates? If he were to lower rates today, you would worry that, wow, the economy is slowing down. This is not good either So you right It a it a it a child It a it right And the simplest trade to make right now is to buy the chips and hold on Right or wrong. Until they realize that Walsh is going to do what he's supposed to be doing.

And they're going to say, well, actually, he's holding his own. He's independent from the president. So this is a good thing now. You can't predict this stuff.

No, you can't. And, you know, what I said at the conclusion of my implied volatility article, that it does not backtest well, you know, that it's overvalued. And it's, thank God it is overvalued. If it could predict perfectly, why would we have an options market?

So, you know, sellers have to be compensated for their risk. For the uncertainty, of course. Simple as that. Otherwise, we'd have no volume whatsoever.

Getting back to SpaceX, SpaceX was sort of an interesting option IPO in that when options IPOs open the first day, generally they're radically overpriced. And options traders that don't keep track of implied volatility, I don't think they realize that. So usually they go off at like 150% or something like that the first day because you have to build in so much to protect yourself because you just don't know which way this thing's going to jump. And then the second day it comes down to like 90 and then it kind of settles down into a trading range of implied volatility.

In the case of SpaceX, it's still kind of like jumping around. We still have like 90%, mid-80% implied volatility in SpaceX, which is a serious number. I mean, that's a commodity-like volatility. It's not, you know, that's, so what is it, like 7% a day or so is what the implied volatility is forecasting.

That's a pretty big variance on a single day, and yet it's still holding up. If it finds a trading range, you know, then you'll see implied volatility coming down. I think what a lot of people don't really understand about implied volatility, a lot of quants, I think, don't really understand this, is how it's priced from the standpoint of a market maker. That, you know, they come up with all these models and stuff, which are all great, and they guide you as to, you know, how should I be pricing this thing according to what's going on.

But really what it comes down to is I'm going to sell this option to retail or to whomever. I'm going to do something in options. And then I'm going to hedge it. I'm not going to be just like hanging out there.

I'm going to hedge it because that's what the business I'm in. I'm trying to make bid offer spreads or I'm trying to make a slight change in implied volatility. So I need to hedge it. And the question always comes down to how difficult is this thing to hedge?

Can I hedge this thing quickly? Do I have to pay a giant bid offer or am I going to get screwed on this thing? So things that are thin trade a really high volatility because I can't hedge it. Or things that are racing around and I can't predict the underlying for one minute to the next, I'm going to jump implied volatility because my cost of doing business is getting stratospheric.

And that's really what's driving implied volatility. It's like when I was a market maker, I would do the option and then I'd hand it over to my hedge clerk who would hedge it. And if she started telling me, you know, this is becoming a problem. We're in a fast market.

I can't hedge this damn thing. I would jump up and play volatility. You know, I'd shade it to the right. And that's really what's driving it on an intraday basis.

you know and for for things that are thinly traded or haven't traded before it's some weird instrument or something you know we'd stick it into all the models and things but in the back of my head what I would always be thinking is all right if I sell this thing how much am I going to need to feel comfortable you know how much am I going to need and so I figured it out multiplied by probably 1.5. And here we are. And, but I was, I remember I once did that.

I had a client, which was one of the more interesting client conversations I've ever had. He asked me to price a deep out of the money option and I priced it. And he said, that's really expensive. And I said, well, do you, do you want, do you want to sell it at that price?

and he said no I wouldn't sell it at that price so I said well then I guess it's fairly priced you know if you don't want to sell it well I'm making you a price if you don't want to buy it or sell it then I guess it's you know you can't complain about this thing so but that's really how options are priced at the end of the day in my opinion great well let's see what happens next let's see what this market takes us because I think that it's going to be an interesting summer. I'd like to have you back at the end of the summer because I want to reply back on this conversation and say, remember, this is you saying whatever.

Thanks for your time today. Of course. Good morning.