Thank you for this post. I'm going to break it down in segments. Firstly, I found this comment to be interesting. Don't think of vol as vol. I know what you mean. So if the premium of an option is pretty much the "vol" number this sparks a question: In regards to pricing models the one variable that is "unknown" is implied volatility, no? This is the one input that can't be pinned down. Maybe the reason being is because the entire premium price IS volatility itself!? Options are called options for a reason. They can be used as insurance, hedging, speculation, etc.. but the main "hidden" function of these derivatives is that options are pure vol. Options price the future expected magnitude of a stocks price. This includes probabilities and statistics as tools to measure the expected range. This is very interesting. I'll add on more soon.
Very similar? Isn't VIX just the calculation of OTM SPX options 30 days out? Correct me if I'm wrong. This is why I said that the VIX is just one facet of vol. I notice the finance industry zooms and focuses in on the VIX as a market fear gauge a little too much. Vol is much deeper than the VIX. Plus there are many more CBOE products that measure different vols such as the RVX, VVIX, VIN, VIF etc and a host of vol ETF/ETN's. Regarding your questions. I'm going to randomly pick $TLT. JULY Monthly (19 days) ATM $133 Call: $1.12 1. What is the implied vol of this option? The IV is 9.81%. This is where I get a lil confused. Every individual strike has its own IV, but the entire term (weekly/monthly) also has its own IV. Not only that, the entire option chain/product has its own IV number. So theres several implied vols within one product. My question is: Which IV number is most important to you? For me I pay attention mostly to the IV of the product and also look at the current IV percentile and also IV rank. I get my IV numbers from thinkorswim and the IV rank from tastyworks. I'm just curious how you guys interpret these several IV numbers. 2. What affects the price of this option the most? Well lets look at the greeks. TLT $133 JULY (19) = $1.12 Vega = .12 Theta = -.03 Gamma = .13 Delta = .48 All else being constant, the delta affects the price of this option the most. 3. Why might this be the implied volatility? Not sure what you mean by this? $TLT has been boomin since November and its IV is around 10% which is pretty low in its percentile (35%). But whats interesting is, since TLT is a bond product its vol seems to react differently than lets say SPY, IWM, AAPL etc... observe and see when TLT makes a good run up its IV tends to also follow and expand. I've heard traders say other products like commodities also show signs of high iv on up moves. I read that in the roaring 90's in the late 1996,97,98,99 when the nasdaq was crashing up that vol was also very high and stayed high. This is very interesting since it shows us the different aspects of the nature of volatility. 4. Can i price the option better? You stumped me. Please help lol
A lot of traders believe that vol is mean reverting. Let me play devils advocate... Even if vol is mean reverting, how can this benefit your trading strategy? They say predicting vol is "easier" than price... really? Explain why please. Just because vol is mean reverting doesnt mean vol with revert to its mean lol Lets say IV is in its 90th percentile and you decide this is a great opportunity to sell an ATM straddle. Well just because vol is high doesn't mean its going to revert down. Vol can stay high for long periods of time. Its interesting that traders think just because vol is high that it must go down now.. High vol begets higher vol, low vol begets lower vol.
In this book I'm reading the author says...... "The goal of the iron butterfly is for the underlying not to move as the decay from the short ATM straddle outpaces the decay from the protective wings. Volatility: Much like an iron condor, for an iron butterfly implied volatility does not need to be high, but only HIGHER than the average true range of the underlying. Thus, the iron butterfly can be traded in just about any type of vol condition. It only matters that the expectations are that IV is HIGHER than ATR." Reading this, I wanted to ask you guys this question. When the author says "average true range - ATR" is this synonymous with historical/realized vol? If not, I'd like to hear your thoughts on this thinking process.. I understand traders sell vol when IV>HV, but what about IV>ATR?
When you trade, you are always trading the distribution of out comes not what is exactly going to happen. So yes, if vol is in the 90th pctile and the asset is mean reverting, what is more likely? Vol going up or vol coming down? Volatility is "some what" bound. We call this a stationary process. It is much easier to predict something that is stationary than something that is not.
Great post, So we know the underlying has a distribution. Each instrument's distribution is dependent on its nature. Walmart's price distribution is vastly different than Tesla's distro. Knowing this, a trader can map out the statistics of the underlying she wants to trade and use optionality to extract alpha from it. The alpha comes from measuring the distribution and implementing a strategy that correctly assessed it. When you trade a short strangle you are trading a "normal" distro, when you trade a jade lizard you are trading a "lognormal" distro. Thus, it is important for the trader to equip his arsenal with a statistical mind to properly measure and evaluate the distribution of returns and prices from the underlying. Also, your comment about vol being in the 90th percentile, yes it would be a higher probability of success if the trader sold that vol rather than buy it, but 90th percentile gives us a 10% margin of error in a way, most of the time (in this market enviroment at least) the VIX lives under 20 and vol is always relatively low, so percentiles are rarely in the 90th besides earnings. But I agree, one should look at many metrics to order enter a trade. Percentile is just ONE of many tools.[/QUOTE]
Can you expound on this please? Bound meaning "range-bound"? For ex, IV can be 5% or even 150%, but looking at the IV over a certain period of time the IV tends to stay in a range. This is why IV rank is a great tool for seeing if premium is rich or cheap. Is 150% IV rich? You could only know that by measuring that products IV from the past to determine that. Also, is price-action non-stationary? A stock can go from $5 to $500 which is basically "unbound" but the vol of that underlying will only fluctuate in a much smaller range.
Imo, past IV can be an input, but it’s not the only way or even the primary way I’d use to determine if current IV is rich. IV is only rich if we’re going to realize less than it for the duration of the trade. Having a short term forecast of stat vol is a necessary tool. Also, there’s a decent amount of places where vol trends pretty well. I’d rather sell “cheaper” IV than the rich 90th percentile vol if we’re passing through the former as IV descends following an extended down shift in stat vol, vs. a blind reversion trade where vol is blowing up and finally happened to hit 90th percentile of its historic values over your lookback window.
Thank you for this post, I agree with a lot of it. I totally agree that you need to use stat vol as a metric when trading IV. A trader I know will usually sell premium only when IV > HV and ALSO doesn't enter until he sees stat vol starting to decline first BEFORE IV. This is very important. Volatility trading IMO has much more "merit" then many other strategies. The boundedness of IV and the price distribution gives the retail opportunities to use optionality to ride the distro. A iron fly, strangle, anything delta neutral is a way of mapping out the distro with distro strats. I'm reading an older journal from @atticus and it seems he day-trades fly's in fast movers like AAPL. He will enter and exit usually within a day or two. I'm not sure how he determines his entry/exit but I'm very interested in this method.
Once price paths through your midpoint on unhedged short vol structures the r:r doesn’t really warrant further holding unless you have really strong conviction price will oscillate there in order for you to collect further decay. Likewise if theta flips it doesn’t make sense to burn through time and money hoping price will come back.. once your initial target is already shown to be wrong. If strong conviction still exists you could always restructure. The concepts don’t change much and can be applied to different resolutions. Timeframe decisions depend upon what you can best model for. As frequency increases commissions erode more % of profits and the infrastructure disadvantage becomes more significant. Longer holding periods arguably contain more unknowns and risk. Like everything it’s a trade-off.