The Vega Options Way

Paul put on a nice new trade in $AMZN last week. We discussed it a little bit in real time on the Twitter feed, but the trade deserves a more thorough breakdown because it contains a lot of good lessons about price action and trade structure. Paul gave me a set of screenshots and asked me to analyze his $AMZN trades from last week and what adjustments he might be considering this week.

This is a fairly complex trade, and as you’ll see I have a lot to say about it. So we’ll start just by walking through the price action. Then we’ll analyze the trade structure. Here goes….

THE SETUP

Toward the end of August, $AMZN rallied strongly into a large overhead gap and over last year’s high. On Wednesday, September 1st, $AMZN had a gap-and-go higher, right into the upper band of the descending 8-week SMA (and the long-term downtrend line I pointed out on Twitter). Price retraced immediately. The day finished 2u but red, extended from the 8-day SMA, and holding just above the prior day’s/prior month’s high. Paul and I both tweeted that we expected a further pullback.

Paul said that he would be short if Thursday dropped below Wednesday’s low. That level was a powerful bull/bear line not only because it would make Thursday 2d (confirming the pivot at the trendline) but also because it would put price below August’s high.

On Thursday, September 2nd, Paul got his move lower and opened a short position. But if we dig into a lower timeframe, we see that Paul opened his position before price breached Wednesday’s low. So did Paul just get antsy and jump the gun? I don’t think so.

I’d wager that Paul knew very early in the day on Thursday that he’d need to find a new entry for his trade. Why? Opening range! If price had opened near Paul’s key daily level (Wednesday’s low) a trade on a breach of that level would have had a tight stop (day’s high), decent price, and plenty of room to develop a profit intraday.

But the day didn’t start like that. Instead, price opened almost 20 points above Wednesday’s low and set an opening-range high all the way up at 3511–about 36 points above Wednesday’s low. That opening range made waiting for a break of Wednesday’s low problematic for a few reasons:

1) The best stop level (the day’s high) would be 36 points away!
2) Having traveled 36 points just to reach the trade trigger, price might have “less gas in the tank,” increasing the chances that it would stall for the rest of the day.
3) And that 36-point drop definitely would have elevated IV, inflating the cost of long-vega directional positions. After entry, if price were to retrace or even just stall for the rest of the day, that inflated IV would get sucked out and the trade could be underwater due to vega alone.

So buying sooner gave Paul superior location and superior price. But only if he could identify early price action that was consistent with the thesis he’d constructed based on the daily timeframe. So let’s jump into the price action and see what Paul might have seen to allow him to enter this trade.

To do this, I’m going to zoom into a 5-minute chart. Paul definitely wasn’t trading a 5-minute chart. I’m using it here because it’s helpful for reconstructing the “feel” of the price action and seeing exactly where Paul took his trades.

As you can see on the chart, by the time Paul entered his trade at 11:08 a.m., price was showing that buyers were losing their grip. Buyers kept trying to lift prices, but they couldn’t keep it above Wednesday’s HB and every attempt to do so failed faster than the one before. Paul entered when it became clear that buyers were losing not only Wednesday’s HB but Thursday’s developing HB and Thursday’s open as well.

This entry fixed all three of the problems with Paul’s original setup that we discussed above:

1) It gave him a much tighter stop (about 14 points to the day’s high versus 36),
2) It gave the trade downside room to breathe (about 20 points just to get to the Wednesday-low), and
3) By entering during the rotational period of what was then still an inside day, Paul entered before IV popped, giving himself better pricing, controlling his risk, and adding an IV edge to his trade.

[It’s worth noting here that price was also in a bearish posture with respect to the 8-SMA bands on both intraday timeframes we commonly use (the hourly and 30-minute). I’m not showing those here because I don’t know which one Paul was using, and I wanted to demonstrate that this trade set up perfectly just using price’s interaction with key daily levels, regardless of the timeframe of the chart.]

THE HEDGE

Paul hedged his original risk at 3:44 that same day. As you can see in the previous chart, momentum had slowed, and price was running out of time for another drive lower. Rather than wait until the last minute when spreads can get wonky and little adverse moves can disproportionately skew options prices, Paul locked in his hedge with plenty of time.

THE FOLLOW-THROUGH

On Friday, price gapped lower and dropped off the open, but then rallied to break the opening 30-minute range to the upside almost immediately. After the morning push, price seemed to stall in the bottom of Thursday’s range, but pushed higher in the afternoon. Paul made his final adjustment 10 minutes before the bell, after price found acceptance above Thursday’s close and Friday’s developing HB. His timing was perfect, as price spiked almost 20 points higher into the close!

THE POSITION

Okay, now let’s talk about the specifics of Paul’s options position.

Paul opened the trade on Thursday with five units of the Sep24 3450/3375 put vertical. Three things jump out at me here: First, I believe this is about half Paul’s typical size in $AMZN, which probably reflects the fact that this trade was taken in anticipation of the breakdown below Wednesday’s low. Second, the vertical is quite wide, which increases cost and initial risk but, facilitates adjustments. Third, even though he was anticipating only a short-term pullback, he bought over three weeks’ worth of time. Better to have too much than too little.

The Thursday adjustment is where things get really interesting. Paul rolled the long 3450 puts down to 3425, narrowing his debit spread to 50 points. Simultaneously, he sold the 3325/3300 put vertical. This adjustment created a broken-wing condor, collecting $14.50 in credit and reducing original risk from $21.00 to $6.50.

[Edit 9/7: This adjustment is worth studying/learning for two reasons. First, it can yield substantial credit much earlier in a trade than simply bolting credit positions onto the far edge of the position. This is especially true if your spread is far out of the money, the debit spread is quite wide, and/or time to expiration is short. Second, and equally important, this adjustment took substantial risk off the table without choking the trade or altering the directional bias of the position.]

Then on Friday, as $AMZN showed reluctance to follow through lower, Paul rolled the 3300 puts down to 3275, widening the credit spread to 50 point, creating a symmetrical condor and collecting a further $2.65 credit.

As you can see from the risk graph below, the position still has negative delta and will befit substantially if price continues lower. Theta is positive, and the reward-risk is almost 12:1.

HEADING INTO NEXT WEEK

So what to do with this position now? With Friday 2d and the weekly bands trending down, it isn’t time to exit the short. That said, the daily chart looks like a bull flag, which means this pullback might just be the rest before the next leg higher.

For me, the bull-bear line is last month’s high of 3472.58. Below that and price remains neutral to bearish. Above that and the daily chart will begin to look bullish, and the weekly chart less bearish. If price remains below that level, Tuesday’s candle would be a 1 at best Price below Friday’s HB of 3459 is bearish, below Friday’s low of 3436 strongly so. 3483 would make Tuesday’s candle 2u and signal the potential move higher.

Paul is considering three adjustments. Each involves buying one or more Oct15 call verticals. The first adjustment buys one 3500/3600 call spread. I’m not a big fan of this adjustment. It adds significant delta to the trade, but you get paid for it only if $AMZN moves several hundred points higher.

The second adjustment buys a wider Oct15 3500/3700 call vertical. This adjustment flips delta positive but only slightly. It won’t hurt if price slides slowly slower. And the new position’s upside breakeven point is substantially lower than in option 1, meaning you’ll build profit quicker if price rallies hard to the upside. Note, however, that theta is now negative and vega is strongly positive. This adjustment will require the stock to move substantially. This would be my preferred adjustment if price broke above last month’s high.

And finally, Paul proposed a third adjustment that simply buys two of the Oct15 3500/3700 call verticals. Now, delta is solidly positive, theta is more negative, and vega is even larger. This is a strongly directional play to the upside. If price rallied, I would execute adjustment #2 over last month’s high and adjustment #3 (by simply adding a second unit of the call vertical) if price continued over Friday’s high.

I hope this post will be as helpful for you to think through as it was for me to write. Let me know your thoughts!