Sirion

Actually, would it be worthwhile (probably depending on difficulty on his part) to flag if the underlying is a part of the S&P500? I'm now wondering if we even look at underlings that aren't, just because we tend to require a certain amount of liquidity. The secondary question would be if there was an easy API pull for the current weight within S&P500? As that could easily be a consideration for how strongly VIX would correlate. Just an assumption, but a pretty straightforward one.

Final thought: Saving a user's last inquiry timestamp, allowing for a simple flag of "new earnings confirmations only" feels like it covers most of the use cases while being the simplest.

Is anyone aware if Interactive Brokers has any partners for UIs like TradeHawk with Tradier? I've personally decided to move away from Tradier, but man once you get used to a nice quick options UI it's frustrating to move back. Or maybe IB has tools to let you set up faster trades yourself?

Might save you some processing power/data transfer if it's a filter on the front side of the request, in addition to not breaking people's macros as an additional thought.

Hi, is it possible to add a filter to the return scanner (or a column of data) in order to filter for confirmation date? Specifically, in order to only look at data confirmed in the last 1 day for example, that way you can avoid reviewing the same data every day (or manually filtering based on the earnings date feed). Thanks

Honestly, I'm about ready to cancel my CML trade machine. Just not getting good use out of the individual tools, and they don't post enough private research to make it worth it.

@Hielke Pre-earnings momentum trade. The basic thesis is that leading up to earnings, some stocks tend to have positive momentum regularly. It's a directional, high-risk, relatively short term trade. I'm laying this out because I think many people that were looking at these trades gave up during Q1 of this year, because oh man it was rough. Very small position sizing. Cap the number of trades you run at any given time. If you run this with other bullish strategies (short vol in particular), be even more careful and consider hedging your portfolio. Some stocks that I trade with this will flat out ignore market drops during their pre-earnings phase. Others will not. While many trades are closed at -40% (a widely used rule of thumb is a profit target of +40%, loss limit 40%), I have had 100% loss trades. A stock I did this a couple of weeks ago with had an accounting scandal, dropped 15+% during the day, and rendered the options basically worthless. The only trades were small lots of minimum pricing, I assume traders closing positions for cleanliness/margin reasons. I actually let that one ride through earnings, thinking the commission fees weren't worth trying to close out the dirt, and the company had crushed earnings before - no luck. Anyway, I personally think there's a definite edge here but you have to be careful both in terms of individual position sizing and overall portfolio effect. I'm actually very behind with some of my logging, so I can't even give you my personal profitable trade rate, or the average gain using this strategy. I may even have lost money on it this year, considering how badly Q1 went. This quarter has treated me much better. Some people have looked at an inverse of this trade - negative pre-earnings momentum, in an effort to balance some of the portfolio effects. The return scanner even lists put returns. (also, you can find some non-official straddle opportunities that don't meet SO criteria but may outperform their cost for whatever reason (RV drop might be too high, IV might be out of range, etc.) I don't think there are enough put candidates to justify the effort searching for them, so I haven't really pursued that myself. If someone has been successfully utilizing those trades (as a balance or on their own), I'd be happy to discuss some elements of strategy with them. I think it's not an SO official trade largely due to the directional nature and volatility, though I might be putting words in Kim's mouth.

I almost submitted a post answering this before realizing we're in the public forum. I'll update a post in the "CML Trade Ideas" category and tag you. @Kim Not going to dig back through this whole thread myself, but there may be more trade information in this thread than you'd like. You may want to copy this thread into one of the private forums (while leaving a "public" thread as well for less sensitive feedback / updates). Not really my concern, just a thought

LX was one, zuo another. I noticed a T-# were off, is that all we have to watch out for until you rebuild?

@Djtux Not sure if I've asked this before, but can we please get a minimum number of cycles filter for the return scanner? I'm thinking default should be 4-ish. Kinda annoying scanning through, finding something that might be good, and realizing it has 1 or 2 cycles and is in no way trustworthy as a pattern

@Djtux quick minor bug report: when you have an open window and a "ignore cycle" selected (I was using BKNG), if you plug in a new underlying, it keeps the "ignore cycle" in memory (and will error unless you happen to be looking at an underlying with the same earnings cycle. Don't know enough about web development to know how hard it is to avoid this, but preferably shouldn't have to deselect before going to a new underlying.

Honestly, this forced me to consider the choice of median. Personally, I like this underlying a little more having more closely looked at the underlying data. Given the choice between only average or only median, median is a better rough tool given the theory behind the trade. Nonetheless, extreme positive returns are a pro. I'll have to think on it for a bit.

Fair enough, should have done the calculation myself instead of trying to eyeball it. My bad

@Djtux is the beta return matrix handling errors in the data properly? For example, SHOP https://www.volatilityhq.com/backtester/return_matrix/?symbol=shop&start_date=2016-07-25&show_individual_matrix=No&strategy_type=LongCall&long_delta=0.4&submit=Run+backtest there's a null data point in one of the early entry cycles (17-03), and I think it's throwing off the average calculation for the rest? Looks lower than it should be - take a look at entry t-4, exit t-0. Edit: additional note while I'm actually in the thread so I don't forget again, a filter for minimum number of cycles would be useful (honestly, this could just default to 4+)

Thanks for this every little thing helps Looking to see if maybe people think a minor re-arrangement might be useful visually: Would you like to see columns arranged by time until earnings before delta? Specifically for calls, as they're the only one with multiples by default. (For example, 60DT10, 50DT10, 40DT10... 60DT5,50DT5,40DT5) etc Personally, when I'm scanning the data, I feel like I'm mentally jumping columns to see a trend. Just a thought

@Djtux For the return scanner, if I wrote a quick macro (stuff I do every time I open the excel), would you be willing to implement on the download if you thought it was useful? Primarily, it's freezing the top row (removing the nameplate row, though you could just freeze top two), and then adding some color coding to the different columns. I do very basic coding, if I wrote a macro that'd be permanent I'd add a bit more flare to it.

Derp, I see the problem. Mixed issue of looking at the wrong entry date and not realizing the 2016-12 cycle had one very good day and would have been closed due to profit-taking before it crashed and burned. Hm. Still on the fence about this one - I've seen better pre-earnings momentum trades, though there aren't a lot available right now.

So, https://cmlviz.us12.list-manage.com/track/click?u=8740b0e4e1d86119b842cbdf2&id=a9d07ee62f&e=512930d399 MU pre-earnings long call was published today. However, it seems to disagree with @Djtux's return matrix, which shows some cycles losing. Also, against the thesis of the trade, it seems to only show optimism for a short period - and then holds steady for the last week. I'm leaning against this one personally - I don't see enough to pull me in in this environment, but I am curious about where the difference lies in the CML data and vol hq's?

I've personally started using the pre-earnings momentum trades again and it's gone fairly well, but yeah I'm using CML itself less and less. The only thing I'd use it for back-testing is post-earnings ICs. I need to devote the time again to finding them, but they've been painful so I may just avoid them.

Sounds good! Yeah, I don't think we need updates on backend or typo/formatting fixes I'll make sure to follow.

Just curious, do you have update notes or anything somewhere? I ask mostly because if you roll out a new feature, it'd be nice to check it out and see if it's useful for me in particular - and it'd give people a chance to give some quick feedback on something you just worked on

I almost closed (had a sell order in, just didn't get filled), and my thinking was that it already had an amazing day to day move (over x2). Still, there was positive news among other things to drive the run.. and the bottom fell out of it. In the future, I'll be more likely to take the win (or even close half) on the last day of the trade. If it gets much worse I might just roll the dice on earnings. Not sure.

You may have missed the peak.. but I'm wishing I got out for breakeven now. Down 60%.

ugh.. BA was in the green this morning and I didn't get to close before the bottom dropped out on me. This is brutal. Lesson learned.. the price of the option had more than doubled, I should have just taken the win. I think MSFT flared into the green too, but at least has one more day. Hoping for a recovery.

Yes, that's the one. Price effect might be useful too, if it's just a matter of c&ping some code over