Only on Planet Grim. In the real world, they have negative EV.

Oh man, you still don't get it. Cost and EV are not the same. The cost is the premium of the option. With SPY at 246.91, July 2018 SPY puts are at mid price of ~$12.84, which is ~5% of the SPY. Regardless what the EV is of buying that put is, your cost of 100% downside protection for 1 year is 5%.

Yes. That is the price in the real world. The cost of purchasing and holding a put until expiration on Planet Grim (where the EV of betting on a put is 0.00) would be just over 2%.

EV of a bet isn't the same as how much it costs. It is the expected value at expiration of the bet minus how much you have to lay up to make the bet divided by the bet amount.

At nearly record low EV (the lowest was in 1993), selling a put is still a couple of points above fair value. Just a bit over 50% annualized profit expected given current prices of volatility.

Ah so I missed that they think you cannot detect those stocks, you can only know in hindsight if you actually outperformed.

That is really strange lol, I mean someone has to make this market efficient right? Damn these ivory tower economists.

Grim, if someone buying puts expected to break even on the purchase price over the long term (because they are 0 EV) and given puts protect you against falls in the value of equities, shouldn't everyone holding stocks be protecting their entire portfolio with puts?

If that was the case we should be holding highly leveraged put protected portfolios, assuming the underlying is expected to appreciate more than the cost of leverage

Sent from my Nexus 6P using Tapatalk

Puts aren't free. Like I showed in my last post, 1 year protection in SPY costs 5%.

Sent from my SAMSUNG-SM-G925A using Tapatalk

If the EV is 0, then they are, in fact, absolutely free. If 1 year SPY protection cost 300% and putting on the protection had an EV of 0 (meaning your weighted average expected price to get back out of the position is exactly the same), then they are free.

Upfront price to buy minus price when the position is closed is the cost of protection.

First, 300% doesn't make any sense, so why use it at all even as a hypothetical?

Second, how is it free? The put is protection -- your overall position is still long. You don't make money if SPY goes down. You make a flat -5%, which is the cost or price of the option. If SPY returns X% (positive), you make X% - 5%.

Cost is just the price of the option, so how can it be free? Do you agree if the put cost $0.25 (0.1%), then the put would have +EV? If so, it still costs you 0.1%. You make X - 0.1% on SPY's upside, -0.1% on SPY's downside. Is the cost "free"? No. Is the cost negative? No. Cost is still positive, despite its EV being positive too.

I think you're confusing the terminology. When we say "EV", we mean the EV of just buying a "naked" put. When we say cost, we mean the cost of downside protection of the underlying. In fact, I'm pretty sure your broker add the premium of a married put to the cost basis of the stock, along with other "costs" such as commission.

Correct. It makes absolutely no sense that the EV of putting on a put would be 0. This is exactly what we've been trying to tell you.

Now do the same thing, except buy a call and a put at the current strike price (with no long position), presume that the EV of each is 0, and you will be able to see that options cannot have an EV of 0.

Brian, you are spewing hot garbage, please take some time to think carefully about what you are saying. Cost and EV are totally different things.

Based on your logic you should go buy as many houses as possible today. After all, the ev is at worst 0 (and probably more profitable than that) so the house is free!

I didn't say they were free. If they cost 5%, and the EV is zero, then on average you expect to get the 5% back again, right?

I don't think he's disputing that if a stock is +ev then a put is -ev. His side is that stocks being +ev (i.e. having a risk premium) is not a law of nature, while the other side is that stocks being +ev is a law of economics.

It doesn't seem like people are arguing about the same thing here.

This is beyond silly. Tone down the confidence

Good contribution...

grunching.......... but yes, google "volatility smile", "tail risk", "crash risk"...

problems with black-scholes - even black or scholes wrote a paper on it. i do think many people realize that black-scholes is a great algorithm but that it is based on a bunch of assumptions that often don't hold. people basically use black-scholes and then some heuristics - correct word? - on top of it.

That is what a bet with 0 EV would be anywhere.

It has nothing to do with whether stock markets are +ev. When stock markets are -ev, puts get super expensive just like any insurance.

What/Why? You said:

And I asked, how are they free?


Why?

Yes. Then to address your next (or I guess, original) question of why shouldn't everyone buy protection, it's because while you're fully protected on the downside, it's harder, i.e less likely, to make money on the underlying on the upside.

Because it is -ev to do so.

Not 0 ev, but instead -ev, as many of us have said.

No. It's not (necessarily) because an option is -ev, it's because your cost basis worsened. EV can be 0, and it is still more difficult (lesser probability) of making money on the overall position because your breakeven price on the underlying is higher.

If you look at realized volatility and implied volatility, you will see that puts are usually priced too high to be a 0 EV proposition.**

This isn't some law of physics or anything, but insurance needs to be profitable (+EV, on average) to sell and costly (-EV, on average) to buy if you believe that the market participants are on average at least somewhat rational. It must cost EV to make bets that offload risk onto another market participant, unless you believe that the market is completely silly.* No rational agent would take risk off of your plate for no expectation of profit.

That doesn't mean that every put loses money. It does mean that in aggregate, they must lose money.

*there is a possibility that every single market in the entire history of finance is wrong in their pricing of options.

**this is true of every single market (stocks, commodities, bonds) in the entire world, even those for which there is no expectation that they will go up in price over time.

Cmon, this is hardly an argument. Puts NEED to be -EV? Really? Brian NEEDS to be right? Neither does rationality imply anything here.

And while puts can be used as insurance, they're not necessarily an insurance contract. I use (sell) puts to get long stock. And I'm sure I'm not alone.

There's all sorts of financial instruments out there that one can use to hedge a core portfolio. Are they all -EV too?

It has nothing to do with Brian. It is just a simple fact that, on average, IV > HV and IV > RV across every single asset class.

If this weren't true, then put sellers would be quickly put out of business because of the higher volatility of their returns (highly limited and known maximum upside, large and variable downside).

There is a reason why selling volatility is described as "picking up pennies in front of a steam-roller."

You are selling an insurance contract whether you like it or not. That you don't mind taking ownership of the underlying (if the strike is hit and you are exercised) instead of paying out the cash to get out of the contract isn't an issue. That you are willing to miss out on the upside in order to sell the contract is an issue. It would be ridiculous ("I'd like to limit my upside and take on all the risk") to not just go long if it weren't +ev to sell the put.

Diversification is free.

IV > HV is not necessarily fact. It could also mean the model you used to compute IV in the first place is wrong. Fix/tune your model.

Limited upside is a fact, but simply stating this fact does not imply that it's not baked into the price.


None of this implies that a put can't have 0 EV. Again, limited upside and 100% downside can and is baked into the price.

At least in my personal case, there are other considerations, e.g. the price I'm actually willing to buy the underlying.

Anyway, it's pretty obvious you waving the white flag here. None of anything you're currently saying theoretically/mathematically implies that puts can't have 0 EV, other than your gut feeling.