Boxcar video on Oct 4th and TOS error on analyze tab calculation

[Michael B]

Dan, thank you for the best and most comprehensive one to date.

However I don’t have any question or comment about the trade itself or the adjustments. It’s about this error we keep having on thinkorswim and how it impacts you.

For months now the analyze tab seems to have a problem calculating theta and P&L for next day. For any position you look at on the analyze tab, when you advance one day from the current date, the new P&L will be in error. Sometimes it can even show a loss despite having some reasonable positive Theta the day before. So it’s not the sum between the P&L and Theta from previous day. But if you advance one more day, it will then get in sync and P&L will increase each day by the amount of Theta (assuming positive Theta here). Based on the option chain data, Theta for current date seems to be correct, but the way it impacts the following day P&L looks flawed. Then when looking at following days all gets back to normal.

Here is a reduced boxcar shown today. Please disregard the setup, this is just to have a look at Theta and P&L. The prices moved a bit while I was capturing the screen shots, but not to the extent to justify the error I’m talking about.

Here it is today:


And now I advance one day and P&L is not as expected:


And one more day. Now P&L looks as expected based on previous day Theta.


And another day. P&L looks fine.


Or you can look at it using day steps. One day increment here. Clearly the first step is way wrong.


There were few posts in the forum about this issue and some of us contacted TD, but there is no fix in place so far.

For your initial setup and potential adjustments you mentioned you look at theta and theta to delta ratio and you may also use the day steps. My question is how you deal with this issue we have with thinkorswim.

Thank you,
Michael B.

 

[DGH]

Good observation, Michael. I have noticed the same error. My general approach is to view the hypothetical P/L curve in day steps of 3 day increments, so I am not particularly troubled by this error. Also, I set up 2 or 3 price levels at selected intervals and observe the greeks and T+0 curve in order to make decisions for placing GTC orders or real time orders. I have great confidence in the performance of this strategy when properly managed, so I am focused on harvesting profits and minimizing losses on either side of the trade rather than worrying too much about tomorrow's theta. As you saw in the TG1 session, I constantly analyze both the CS and DS side of the trade during the decision making process. I don't like to let the DS "go negative theta" or lose too much money. Similarly, I add more DS or DS/CS combos as necessary to maintain the integrity of the trade as well as peeling off CS when it becomes absolutely necessary. I begin to unwind on Tuesdays with Friday expirations and let a small portion of the original trade run for even more profit. I suggest that you set up a Boxcar trade in Nathan Meyers' Monte Carlo simulator (javalinux.net). I have done so, and am very pleased with the expected return and high probabilities when compared with condors, butterflies, or (heaven forbid!!!) a time spread. For example, with my typical setup used in the Option Risk Calculator, the expected return (for a 100,000 trade simulation using 2 SD to expel outliers) for an 8 lot CS/3 lot DS is $1000 or more. Compare this with the negative expected returns for a typical unadjusted condor or butterfly.

 

[Michael B]

Thank you for the reply and additional explanations.
I know the option risk calculator, but there's no need to simulate. I know the boxcar is a great trade. I tried it many times (the previous variants) and it worked most of the time. For me the problem is I still go to work every day (on the other hand that's a good thing) and only have the 1st hour after the market opening to do any adjustments. Sometimes I do them in anticipation of a market move, other times I do nothing and often I'm wrong, but never had a big loss with it.