Insane Emergency RE strategy

[ERD50]

Regardless what the other side of the trade *expects,* the raw numbers show that the vast majority of options expire worthless, and this benefits the sellers.

Do Option Sellers Have a Trading Edge?

While I agree with the premise that option sellers have the edge, the data within is meaningless. And when I see meaningless data presented as if it is supposed to mean something, I ask - hmmm, whas'up wid'at?

Notice that they never mention the value of the options that expire worthless, minus the losses incurred when they didn't, divided by the entire amount of money at risk over a given time period, compared to the behavior of the underlying over that same time period? Now *that* would give you some indication of how *much* of an advantage the sellers have over B&H. The % that expire worthless means nothing, absolutely nothing.

I'll re-visit the Roulette Wheel. You are the house, and you are "selling options" in effect. You don't care if the bettor steps up and bets either:

Bettor A) On a SINGLE NUMBER - he will only win 1 of 38 times, over 97% of his bets "expire worthless".

or...

Bettor B) On RED - he wins almost half the time (18/38), 52.6% of his bets "expire worthless" - much better performance than bettor A.


Yet, the house will, on average, collect the exact same 5.263% of the money bet from each of them. No matter what bet is placed and what the odds of it "expiring worthless" are. Because simply, the payout matches the chance of the bet. I'm pretty sure "the market" has known this for as long as gamblers have.

Now the only thing that keeps people coming to the table and betting on a single number is that, every once in a while, there is a big payout (but the house still, on average keeps 5.236%). So those people giving you money month after month are going to expect a big payout once in awhile (you take the full loss), or they get bored and don't play (supply/demand says the price they are willing to pay drops). So yes, I think option sellers can skim a bit off because people on the other side are willing to pay to play (either "gamble" by buying a call, or "buy insurance" with a put), but I don't think they will pay anything near 5% a month, on average. If it did, there would be a thread titled - "The easy way to retire early with a nest-egg approaching the size of the GDP", with this approach explained


-ERD50

 

[dixonge]

The % that expire worthless means nothing, absolutely nothing.

I find it interesting that a lot of people are scared off from options because of the oft-cited numbers - "80% of all options expire worthless." This could be because people who buy straight calls/puts lose all their money. Or maybe people buy a lot of puts as insurance for their stock holdings. My spreads expire worthless when my held puts are worthless, and the sold puts are worthless. But this is a GOOD thing from my perspective.

So yeah - we all know what they say about statistics.

Still - I much prefer being on the selling side.

I'll re-visit the Roulette Wheel. You are the house, and you are "selling options" in effect. You don't care if the bettor steps up and bets either:

Bettor A) On a SINGLE NUMBER - he will only win 1 of 38 times, over 97% of his bets "expire worthless".

or...

Bettor B) On RED - he wins almost half the time (18/38), 52.6% of his bets "expire worthless" - much better performance than bettor A.

I was using the roulette analogy just yesterday. I have played at the table for hours. The result? I could play for hours and break even. BORING.

In comparison, option spreads are like laying bets on the table in the section labeled "all numbers except for 1-5" That's an option you'll never see on a roulette table.

So yes, I think option sellers can skim a bit off because people on the other side are willing to pay to play (either "gamble" by buying a call, or "buy insurance" with a put), but I don't think they will pay anything near 5% a month, on average. If it did, there would be a thread titled - "The easy way to retire early with a nest-egg approaching the size of the GDP", with this approach explained

-ERD50

I haven't started that thread........yet....

 

[ERD50]

In comparison, option spreads are like laying bets on the table in the section labeled "all numbers except for 1-5" That's an option you'll never see on a roulette table.

Sure it is. Bet on both RED and BLACK. That is the same as NOT betting on GREEN (0 & 00).

One guess what the payback is .

-ERD50

 

[dixonge]

Sure it is. Bet on both RED and BLACK. That is the same as NOT betting on GREEN (0 & 00).

One guess what the payback is .

-ERD50

not a good analogy - "not 1-5" leaves 33 possible wins (6-38) not 1 (0&00).

BUT - the roulette analogy fails any way since it is random chance on every spin, not a progression down through the numbers.

This *is* giving me some crazy ideas for new games of chance though. LOL

 

[ERD50]

not a good analogy - "not 1-5" leaves 33 possible wins (6-38) not 1 (0&00).

BUT - the roulette analogy fails any way since it is random chance on every spin, not a progression down through the numbers.

This *is* giving me some crazy ideas for new games of chance though. LOL

heh-heh-heh - I've explained that the payouts are adjusted to your chance of winning (at the tables, at the track, or at the market) about a half-dozen ways. If you can't see (or fail to accept) that taking a bet or option that has a greater chance of winning (or of not losing) means that bet or option will have a comparably lower payout, then I doubt that explaining it again will make a difference.

No, I can't seem to offer an analogy for you that matches your criteria 100% perfectly. So please continue to post your trades, and let's see how it goes. Remember, I expect that one can do OK selling options in the long run - I just don't think they can make the kind of rates you are hoping for.

If you don't care to work out the B/E, max/min profit/loss points, then please include the basics:

Ticker/Price/date of the underlying.
Strike Price, exp date, put or call, credit or debit amount for each leg.

Trust me, there is nothing I'd like better than to be proven wrong. If there is a way to consistently make 5% per month, I am all ears.

Back in my working days, people wondered why I spent as much time as I did talking with the "crazy" ones - heck, a lot of time those "crazy" people had some good ideas, but no one would listen. I struck gold occasionally. And the truly crazy ideas were sometimes entertaining, or sometimes triggered a thought that could be made practical. Gotta look outside the box once in a while, but don't forget that there *is* a box.

So keep us informed, this could take a while though. Like I said, I had a pretty good 18 month streak....

-ERD50

 

[dixonge]

heh-heh-heh - I've explained that the payouts are adjusted to your chance of winning (at the tables, at the track, or at the market) about a half-dozen ways. If you can't see (or fail to accept) that taking a bet or option that has a greater chance of winning (or of not losing) means that bet or option will have a comparably lower payout, then I doubt that explaining it again will make a difference.

I do agree that a credit spread (for example) pays less the farther it is OTM. Volatility increases both risk and reward. We're talking basic risk/reward relationship here, right? I think of this as something that is built in to the system, not something being manipulated by the infamous 'them.' But I'm comfortable with the current risk/reward ratio I'm using as it is much more conservative than my previous options strategy.

No, I can't seem to offer an analogy for you that matches your criteria 100% perfectly. So please continue to post your trades, and let's see how it goes. Remember, I expect that one can do OK selling options in the long run - I just don't think they can make the kind of rates you are hoping for.

What kind of rates do you think I *could* make? 4%/mo.? 2%?

You've inspired me to create a spreadsheet to calculate my gains based on the expiration month rather than the month a credit was initially received. It won't track cash flow profit but only realized profit. That might be interesting, and perhaps more accurate.

 

[ERD50]

Volatility increases both risk and reward.

In general yes. Things like small company stocks are generally more volatile, and in general return a slightly higher rate (which the market "requires" to offset the volatility). IIRC, this is called "alpha" (excess return for excess volatility).

But doesn't this work against you? You are saying you expect to "win" most of the time, as your strikes are a distance from the underlying. So, your "alpha" should be small (over the long run)?

What kind of rates do you think I *could* make? 4%/mo.? 2%?

Good question. My gut and my brain says: not enough to get excited about. If there were high profits to be made (relative to the risk) more people would jump on it and supply/demand would erode that to a "non-exciting" level. Options are so closely monitored/traded, I just can't accept that there is anything exceptional to take advantage there - no matter how you slice it with the spread variations.

So, if you view this as "safe", then expect something in the range of "safe" returns. The riskier you go, the higher the potential returns, but you need to watch that you don't take a serious hit on a bad streak - which usually means not allocating too much, which means you only realize those high returns on a portion of your portfolio, bringing the average down to earth.

You've inspired me to create a spreadsheet to calculate my gains based on the expiration month rather than the month a credit was initially received. It won't track cash flow profit but only realized profit. That might be interesting, and perhaps more accurate.

I do both. Now, I'm fortunate in that most of my option trading happens in an account that I don't currently add to, or draw from. So I check my ss against actual account values, they are very close (there is normally a small amount not invested - dividend payments or amounts too small to cover another contract).

My ss (not set up for spreads) shows the *potential* profit if they expire worthless, the total amount at risk, and then I update it at expiry to show the actuals. I enter SPY at each close, and compare myself to the market (add in divs). If I were doing spreads, I think I'd just enter each leg separately in adjacent columns or rows - that would be simpler, and will sum up correctly.

If you google BXM and BXY CBOE, you will get a sense of what I'm doing. Here's some older data that I had saved on my computer, I'd need to find something more recent, and make sure divs are accounted for. They show that for BMX selling options just one strike OTM lowers beta significantly and (over their time period) increased returns slightly. With BMY, they go 2% OTM, and increase gains a bit more, still with lower beta than the index (but higher than BXM). So I'm going further out, trying to find a point that maxes gains with little/no affect on beta. But I don't expect much - maybe a few % points above B&H, but that is a lot to a retiree.

-ERD50

 

[dixonge]

But doesn't this work against you? You are saying you expect to "win" most of the time, as your strikes are a distance from the underlying. So, your "alpha" should be small (over the long run)?

I honestly didn't follow you here. Could you clarify, lest I be perceived as dodging a question?

Good question. My gut and my brain says: not enough to get excited about. If there were high profits to be made (relative to the risk) more people would jump on it and supply/demand would erode that to a "non-exciting" level. Options are so closely monitored/traded, I just can't accept that there is anything exceptional to take advantage there - no matter how you slice it with the spread variations.

The longest performance record I have been able to find for this type of strategy goes back to 2005, so it appears to be relatively new on the retails side. I suspect this is partially due to retail commissions being somewhat prohibitive until then. But regarding supply/demand - do you really foresee that many people delving into options, ever? I have yet to find someone at work or in my circle of friends, or Facebook friends, or any one else I personally know who uses options. I have a hard time believing that even a wildly successful options strategy could ever become so popular as to max out one side of the buyer/seller ratio.

So, if you view this as "safe", then expect something in the range of "safe" returns. The riskier you go, the higher the potential returns, but you need to watch that you don't take a serious hit on a bad streak - which usually means not allocating too much, which means you only realize those high returns on a portion of your portfolio, bringing the average down to earth.

ok, how many times do I have to refer you to the title of this thread? LOL

Regarding allocation - I started off with so little money, and such a short time-frame for my goal retirement date, that I was and still am aiming for 85-90% options, 10-15% cash. There is some theoretical point in the future where I will reach a comfortable equilibrium and do more of a traditional portfolio diversification/allocation setup. But when you're desperate and start w/ $5000 there really isn't much to lose! If I do blow out then I'll just finish vesting at work and do a more traditional retirement a few years later than I intended. *shrug* We've got government pensions to fall back on so we have a baseline support, we're just *really* tired of working.

 

[ERD50]

re: alpha -

I honestly didn't follow you here. Could you clarify, lest I be perceived as dodging a question?

I guess it's your linking of "relatively conservative" with "high potential returns" - those just seem at odds with each other.

But regarding supply/demand - do you really foresee that many people delving into options, ever? I have yet to find someone at work or in my circle of friends, or Facebook friends, or any one else I personally know who uses options. I have a hard time believing that even a wildly successful options strategy could ever become so popular as to max out one side of the buyer/seller ratio.

Irrelevant. I touched on this before - option pricing is somewhat independent of supply/demand of that option, because there is that underlying stock component to it. It is a derivative, not a product unto itself. Plus, I'm pretty sure there are banks of computers comparing the current bid/ask against the Black-Scholes models, scooping up "bargains" which can only be taken advantage of at the wholesale level. They simply won't get "out of whack" for a retail buyer/seller because of this.

An easy example: Look at any option chain on a highly traded index. You will see that some options are rarely traded (there is more action at certain round numbers, or maybe certain % points away from the underlying. But, the bid/ask will be in-line with the rest of the sequence up/down the chain. The fact that there is less "demand" for it has no bearing on the price (it might increase the bid-ask spread a bit). But it isn't like the low demand of an OTM call will cause the price to drop lower than a more "popular" strike price above it.

ok, how many times do I have to refer you to the title of this thread? LOL

But when you're desperate and start w/ $5000 there really isn't much to lose!

- Hey, there are always lottery tickets!

-ERD50

 

[dixonge]

I guess it's your linking of "relatively conservative" with "high potential returns" - those just seem at odds with each other.

To clarify.....my previous strategy was to buy calls in AAPL - a group of AAPL investors regularly spoke in terms of 'baggers' - It was not unusual to refer to a call having become a 3-bagger (having increased to a value 3 times the initial investment). Making those types of returns in a 6-18 month timeframe began to be seen as the ultimate goal.

Compared to that, my current strategy is *relatively* conservative and expects *relatively* high returns.

- Hey, there are always lottery tickets!

-ERD50

I'm sure I can find a strategy for those on the interwebs somewhere.....

 

[FIRE'd@51]

What kind of rates do you think I *could* make? 4%/mo.? 2%?

Actually, I believe you can calculate the expected return on an SPX credit spread held until expiration in a straightforward way. Here's how I would do it:

Assume the VIX at the time you put the trade on is an unbiased estimator of the the future volatility of SPX. Using this (adjusted for the time until expiration) as the standard deviation of a normal distribution of returns (continuously-compounded), calculate the probabilities (A) that the spread expires worthless (you pocket the entire net premium), (B) both options expire in-the-money (you face your maximum loss), and (C) one option finishes in-the-money (the index ends up between the two strikes).

If the strikes are K1 (upper) and K2 (lower), and the net premium received is P

Expected $ Payoff = A x P + B x (P - (K1 - K2)) + C x (P - 0.5 x (K1 + K2))

Monthly Return = (Expected $ Payoff) / (Margin set aside) / (number of months until expiration)

 

[dixonge]

If the strikes are K1 (upper) and K2 (lower), and the net premium received is P

Expected $ Payoff = A x P + B x (P - (K1 - K2)) + C x (P - 0.5 x (K1 + K2))

Monthly Return = (Expected $ Payoff) / (Margin set aside) / (number of months until expiration)

K1=upper strike
K2=lower strike
P =premium
A = ?
B = ?
C = ?

 

[FIRE'd@51]

K1=upper strike
K2=lower strike
P =premium
A = ?
B = ?
C = ?

You can look them up in a table of cumulative normal probalities.

B = area in left tail (to left of K2)

A = area to right of K1

C = area between K2 and K1

Or did you mean that you want me to do the calculation?

 

[ERD50]

You can look them up in a table of cumulative normal probalities.

B = area in left tail (to left of K2)

A = area to right of K1

C = area between K2 and K1

Or did you mean that you want me to do the calculation?

That makes sense to me. And, if we are dealing with SPY options (I am), the VIX should back right into that:

VIX - Wikipedia, the free encyclopedia

The VIX is quoted in terms of percentage points and translates, roughly, to the expected movement in the S&P 500 index over the next 30-day period, on an annualized basis. For example, if the VIX is at 15, this represents an expected annualized change of 15% over the next 30 days; thus one can infer that the index option markets expect the S&P 500 to move up or down \frac{15%}{{\sqrt{12}\ months}} = 4.33% over the next 30-day period. That is, index options are priced with the assumption of a 68% likelihood (one standard deviation) that the magnitude of the S&P 500's 30-day return will be less than 4.33% (up or down).

the formula didn't copy so well - for VIX=15, it is:

(15%/sqrt(12 months)) = 4.33% over the next 30-day period

So that is one std dev, then plot the everything relative to that. Good exercise, but I should get some stuff done around the house


-ERD50

 

[dixonge]

You can look them up in a table of cumulative normal probalities.

If statistics knowledge is required for investing then I'm in trouble. Perhaps I should just stick with CD laddering...

 

[FIRE'd@51]

If statistics knowledge is required for investing then I'm in trouble. Perhaps I should just stick with CD laddering...

OK, you've twisted my arm, so here goes:

For the 2-month put spread you described in post #44, I calculate that A=88%, B=10%, and C=2%, assuming a VIX of 30% and an SPX of 900 at the time you initiated the trade (which appear to be about where the market was in mid-June)

Expected $ Payoff = 0.88 x 342.30 + 0.10 x (-3000 + 342.30) + 0.02 x ( -1500 + 342.30 ) = 12.30 for 2-months

Expected monthly return on margin set aside = 6.15 / 2657 = 0.23% (about 2.8% annual)

Note that I have neglected any fees associated with settling any options that get assigned. IIRC, the SPX's are European-style and cash-settled, but still may have assignment fees.

While this strategy has a high probability (88%) of achieving the maximum payoff of 6.4% per month, the times it goes against you are very costly. If the maximum loss occurs, it will take nearly 16 months to make it back. That is what really drives down the expected return.

 

[dixonge]

While this strategy has a high probability (88%) of achieving the maximum payoff of 6.4% per month, the times it goes against you are very costly. If the maximum loss occurs, it will take nearly 16 months to make it back. That is what really drives down the expected return.

Based on the numbers you were using, with 3 contracts my max loss would be the margin set aside of $2657. How are you calculating the 16 month figure?

 

[FIRE'd@51]

Based on the numbers you were using, with 3 contracts my max loss would be the margin set aside of $2657. How are you calculating the 16 month figure?

Assuming you hold the position until expiration (2 months)

Maximum loss = 2657

Maximum gain = 342.30

Max loss / Max gain = 2657 / 342.30 = 7.8

7.8 x 2 months = 15.6 months (nearly 16 months)

 

[ERD50]

OK, you've twisted my arm, so here goes:
....
Expected monthly return on margin set aside = 6.15 / 2657 = 0.23% (about 2.8% annual)

FIRE'd@51 - thanks for actually doing the exercise! I got my basement cleaned up a bit, and had a nice dinner while you slaved away over data tables

Well, this is very encouraging to me. A 2.8% annual is in the ballpark of what I was hoping for, assuming other option sales would net about the same. My gut tells me they would (within reason), and my gut seems to be guiding me pretty well lately (maybe I should reward it with another beer? ).

Like my gut said in response to:

dixonge:What kind of rates do you think I *could* make? 4%/mo.? 2%?

ERD50: ... I don't expect much - maybe a few % points above B&H (note: I should have specifically said 'annual' here), but that is a lot to a retiree.


dixonge, here's another way to look at some of this: Since there is an 88% probability of winning, and a 12% chance of losing, that gets you to 12/100 losing periods, which is 3/25, or (rounding) 1/8. So, you can expect to lose 3 times in 25 periods, or roughly 1 in 8. Which again, is roughly the same as the payout ratio (max_credit /max_loss). And it is also not surprising that you can go many, many months in the black.

And that is where the hurt can come into play - if you happen to hit a bad streak of two losing periods close together, you can find yourself taking 30 months just to crawl back to break-even. And if I interpreted an earlier post correctly, you are 85% invested in this strategy; if that means that 85% of the account is in set-aside (at risk) that doesn't leave you much of anything to work your way back with.

I think you said you are spread across different indexes, and time frames, etc - that can smooth it, but it won't change the end game.

So while 2.8% annual might disillusion you, I'm thrilled with the prospect of getting even a fairly reliable 1~3% annual boost to the portfolio, which I think is in line with what the BMX and BMY are doing, though they seem to be targeted a bit more towards trading off reduced volatility over maximizing total return. I'm trying to weight it a hair more towards total return.

I've got to go find recent data on BMY that accounts for divs...

-Good conversation - thanks to dixonge for bringing up the questions, and to FIRE'd@51 for the analysis.

-ERD50

 

[dixonge]

dixonge, here's another way to look at some of this: Since there is an 88% probability of winning, and a 12% chance of losing, that gets you to 12/100 losing periods, which is 3/25, or (rounding) 1/8. So, you can expect to lose 3 times in 25 periods, or roughly 1 in 8. Which again, is roughly the same as the payout ratio (max_credit /max_loss). And it is also not surprising that you can go many, many months in the black.

Wouldn't this point to 12/100 losing *trades* as opposed to periods? (assuming all the trades were 88% probability)

 

[ERD50]

Wouldn't this point to 12/100 losing *trades* as opposed to periods? (assuming all the trades were 88% probability)

Sure - I was just thinking in terms of one trade per period for analysis purposes. But if all the trades were 88% probability, then 12/100 trades would lose, on average. But of course, it will be randomized, will some fairly longer than average stretches of wins and losses.

I read somewhere, that some stats teacher would have half the class write down an imaginary random string of "H" and "T", and the other half actually flip a coin and record H/T. He supposedly could identify with high accuracy the imaginary lists - they avoided long strings of Hs or Ts. They happen more often than people think.


-ERD50

 

[dixonge]

Sure - I was just thinking in terms of one trade per period for analysis purposes. But if all the trades were 88% probability, then 12/100 trades would lose, on average. But of course, it will be randomized, will some fairly longer than average stretches of wins and losses.

aha! That makes sense. So if I did, say, 10 such trades per month, some months would have no losers, some would have several, but 12/100 over a long period of time. After a quick count it looks like I'm overdue! So far I've placed 37 of the recommended trades, 2-4 contracts each, none have gone ITM or required a buy-back.

I also broke the gains down to expiration month. So far averaging over 6% on realized/expired spreads, 7.5% if you count July and August non-expired spreads.

I read somewhere, that some stats teacher would have half the class write down an imaginary random string of "H" and "T", and the other half actually flip a coin and record H/T. He supposedly could identify with high accuracy the imaginary lists - they avoided long strings of Hs or Ts. They happen more often than people think.

-ERD50

Reminds me of the roulette table gain. I was always surprised how many times black or red could be hit in a row. If only it was more predictable....

But then with the right money management strategy you can give yourself some edge, but there are entire books on the topic...

 

[haha]

Reminds me of the roulette table gain. I was always surprised how many times black or red could be hit in a row. If only it was more predictable....

But then with the right money management strategy you can give yourself some edge, ...

I do hope you will explain that one!

Turning an unfavorable game into a favorable one by "money management" sells books well, but it seems to me to be akin to creating a perpetual motion machine. Not just unlikely, but truly impossible.

Ha

 

[ERD50]

But then with the right money management strategy you can give yourself some edge,...

heh-heh-heh- this time it will be *me* reminding *you* of the word "insane" in the thread title

But do keep us informed of your trades - it should be interesting, and we are all bound to learn something.

-ERD50

edit - whoooops! haha beat me to it!

 

[dixonge]

I do hope you will explain that one!

instead of 'give yourself some edge' I should have said 'reduce the house's edge somewhat'

MIT students have worked magic with systems, counting and blackjack, but then they got to have fun meetings w/ casino security in dark basements...