Another winning week ... TSLA traitor notwithstanding ...

The NDX trade won again, now 12 for 13 (92.3%) winners. And for some reason volatility was up a bit this week (even though the market remained at record highs), so I received:

• $2.54 (34.05%)
• $2.51 (33.51%)
• $2.00 (25%)

That's over $500 profit on 2 contracts of this trade ...

Now to TSLA:

This is the Gigafactory, where Tesla makes batteries at scale, hoping to bring the price per unit down.

But on Thursday from CNBC, some employees leaked:

Tesla employees say to expect more Model 3 delays, citing inexperienced workers, manual assembly of batteries

I had sold the 340 put and the 352.50 call, and so when I saw this:

... around 11a.m. Friday I went ahead and closed the put spread for a small loss.

But as you can see from the chart, if I'd held on I could have made full profit on this

trade as the price recovered over $340 just before the close.

Profit I would have made: $0.59 / $1.91 = 30.8%!

Another small mistake to get over with during the test phase for this one ... but officially the TSLA trade is now 3 for 3: 100%!

More next week ...

A look at real returns over the period I've been trading NDX

I started testing the NDX trade with 1-lots (the minimum) on Thursday, October 26, 2017. Since then, the main account I've been testing with is up from $5126 to $6903. This is nearly 35%!

... for just those 13 weeks. And I was just testing, trading 1-lots ... if I'd been using the 1/2 Kelly Criterion I'd have lost more the first week but gained double or more for the following 10 weeks ... I calculate the gain would then have been in the range of 50%!

Getting back to the realm of speculation, but standing on the real test results of the past 13 weeks and assuming 91.7% wins on a firm footing, I think:  the one-year Monte Carlo mean result if I assume 15 percent gain on each winning trade and 100% loss on each losing one, risking 33% of the account always:

Finally, if I assume the same as the above run but with an average 20 percent gain: 334%!

And: that's assuming 100% loss on every losing trade, which almost certainly won't be that extreme ...

Wahoo!

This week's results: another win for both trades!

We've reached the expected 91.7% win rate for the NDX trade; it's now 11 for 12.

There was for some reason more volatility this week, so NDX paid $1.75 to $1.80 (i.e. 21.212% to 21.951% on margin). If you (following around 1/2 of the Kelly criterion) risk 30% of your account on the trade, that's over 6% for the week!

The TSLA test trade also worked, returning 57 cents on a $2.50 wide iron condor, so $0.57 / $1.93 = 29.5%! It was looking a little scary end of day Thursday but recovered nicely into the groove on Friday:

We don't have enough data yet to project the Kelly Criterion for this one, but 30% of your account risked would yield 9% for the week!

More next week ...

Updated blog name more accurately reflects what I'm doing

I first thought I could get a trading algorithm to work based on technical analysis:

But I realized after doing some research that most technical analysis trails the market rather than leading it, and there are oodles of other folks trying to do these algorithms ...

I then put my toe into the Tastytrade "trade small, trade often" method .... but I didn't give it too much of a chance before realizing that I could make a much higher return using the Kelly Criterion and the ideas of Ralph Vince.

I evaluate trades by running Monte Carlo simulations on them based on past or projected results.

The two trades I have going right now:

• An NDX 1 standard deviation iron condor on expiring options, put on every Thursday late in the trading day (I've been using around 10 a.m. Pacific) and held until the results are in the next day (based on the a.m. settlement price using a symbol NDS).
• A Tesla 2-day 1 standard deviation iron condor, on options expiring at midnight Friday

I have the NDX trade in "production" using around 1/2 the Kelly Criterion for sizing the positions. I saw a study showing that it wins 91.7% of the time, and it's living up to that reputation.

The TSLA trade is just one I'm testing with the minimal 1-lot, so far. I speculate that it should win at least 80% of the time, but I don't know yet.

Management of these two trades is dead simple:

• The NDX trade is simply "set it and forget it" until you check on the NDS price the next morning. (You can obsess on the /NQ futures price in the interim, but it doesn't help!)
• The TSLA trade is just a little more complex. You have to watch it to make sure the price doesn't hit either short strike ... the NDX trade is cash-settled, but the TSLA trade is not. You get assigned 100 shares, short or long, depending on which option price is breached. So just set an alert at those prices so you get in a close the position before expiration ... you may still make a small profit.

That's it ... otherwise I just sit in cash until the next Thursday. It's very peaceful not being short anymore, in the face of the incessant Trump Tax Cut Rally.

Another winning week: "production" NDX and testing the TSLA trade

The NDX trade is now 10 for 11 ...

and TSLA's test trade is 1 for 1!

Volatility is still (of course) very low, so I didn't get a huge amount of credit on either trade. For NDX I got credit between $1.25 and $1.29 (14.3% to 14.8%) and for the TSLA trade I got 32 cents (on a 2.50 wide Iron Condor) so that's 14.7% ...

The Kelly Criterion calculator gives the optimal risk amount for the NDX range, assuming the previously published 91.7% win rate, as about 1/3 of the account ... which is a bit more than I risked for these. I was at about .8 Kelly for a couple of the accounts. But anytime you win, it's OK!.... I'll have to watch this while volatility is so low. Also it's impossible to fine-tune in small accounts with the number of contracts risked going up in $800+ chunks.

But the NDX 10 for 11 is 90.9% ... and one more win brings it to the mythical 91.7%!

More next week ...

A speculative look at possible TSLA returns

The 2-day TSLA trade I mentioned in an earlier post should work around 80% of the time.
I assumed this win rate, earning an average of 30% on wins (and 100% loss when losing) ...
I ran 20000 Monte Carlo simulated 3-year trials ...

Starting with $20000, at the end of 3 years you average having $150381.55 ...

Not bad at all! Now we'll see if it works, starting this week ...

A note on TSLA trading through expiration and a Dumb Thing I did

Tesla makes interesting vehicles:

But its options have also been very volatile and tradable over the past several months. The Tastytrade Guys say that TSLA is one of the most profitable trading vehicles (no pun intended) they had in 2017.

I've normally been trading "cash-settled" indices including SPX and NDX. When options on these finish "in the money" the difference in price is just deducted from the seller and given to the buyer and it's all over.

TSLA and other stocks are different in this respect. If you have a short call or put that finishes in the money, your option gets assigned:

• short call gets assigned -100 shares of stock at the strike price of the option (per contract)
• short put gets assigned 100 shares of stock at the strike price of the option (per contract)

I had a tiny little TSLA iron condor on a few weeks back and watched as it went a few cents into the money at expiration, thinking that I would lose $14 of my projected $52 profit, something like that.

But nooooo .... I received a short position of those 100 shares of TSLA ... just before they announced they sold some self-driving trucks. And I wasn't paying attention so didn't jump up to get out of this position at the opening on Monday a.m. So my $35 win became a $1000+ loss by the time I closed it. (I could have held on for a while and gotten this back, but I was out of buying power and it's bad policy anyway ... I kept thinking the S&P futures had to come down and they still haven't!)

Dumb, dumb, dumb. Fortunately only a 1 lot .... and I will Never Do This Again ...

NDX trade wins AGAIN ... but profit was shaved a bit

That's 9 in a row and 9 for 10 overall on the NDX weekly trade!

I did start "production" this week, risking 25% or so of the accounts where I was doing this. Volatility is still quite low so I didn't get as much premium as I have been getting normally. In 3 different trades I got:

• $1.42  (16.55%)
• $1.35  (15.61%)
• $1.20  (13.64%)

But a funny thing happened on the way to getting these full results:

I had sold the 6220 call, and the NDS report for Friday (where the NDX settles and that's how the trade's results come out) was 6220.84. This means that $84 of the potential profit didn't happen, so yields were:

• $1.42 - $0.84 = $0.58  (6.16%)
• $1.35 - $0.84 = $0.51 (5.38%)
• $1.20 - $0.84 = $0.36 (3.73%)

I'll take it! But ... getting the NDX trade filled is weird and difficult. The prices bounce around so and the markets are 'wide' (i.e. the bid/ask spread is probably 50 cents wide on this spread) ..

So I'm going to start testing an alternative (or an adjunct; I can do both): Tesla! Symbol is TSLA and its chart looks like this for the last week:

Putting on a 1 standard deviation smallest iron condor on Wednesday the 3rd would have returned $0.66 ... but on a $2.50 wide iron condor, so return for that would be ... 35.89%! And it would have worked this week ....

So I'm going to start running this ... probably early Thursday morning for 2 days next week, for 10 or 20 weeks just to see how often it works. It should be much easier to fill than NDX, and I can use the 20 weeks of data as a rough estimate (I'll certainly assume that it wins less often than the results show,  and that it returns a bit less) ... and then I can use the Kelly Criterion again if and when I decide to take it into "production." ...

Watch this space next week ....