Question on Dan Harvey's Updated Boxcar

[Keanan]

Last Wednesday (1/20) during a round table with Tom, Dan Harvey discussed updates to his boxcar strategy, and I have a few questions about making adjustments.

• You mention that if the market goes down and triggers your orders to buy back a portion of the put credit spreads, depending on conditions, you may decide to rolldown those spreads you closed out. When/how do you decide that rolling the spreads down is the best way to adjust?
• When rolling the put credit spreads down, do you again place them about 18-20 delta out from the current price?

I see how it’s crucial to enter these trades on a down day. Do you wait until you are near the end-of-day to place this trade, or could you enter earlier if the market is down a significant enough amount?

Finally, is there a fixed dollar or percent risk you have going into the trade after which point you would close out the trade?

Thank you Dan and Tom for putting on this round table, I greatly appreciate all the help!

 

[DGH]

Hi Keanan. You pose some excellent questions. Unfortunately, the answers are more subjective than objective. If an order to buy back a credit spread within the Boxcar structure is triggered I generally wait until I see a bit of stabilization of the underlying before rolling down to a strike at the 16 to 18 delta level. Sometimes I will add another debit spread which I can close if the market rallies. This serves to decrease deltas, gamma, and vega. When launching a new position I wait until mid-morning on a down day; then I study the price action in order to make my best guess of a good entry time. I often let an order sit while awaiting a good fill. I fill the credit spreads first, then add the debit spreads. Exit points for profit or bail-outs for a loss are very subjective and depend on the trader's tolerance for risk and potential loss. I can't give you a fixed dollar amount, but I can say that it's best not to lose more than 6% to 8% when the position is finally closed in order to maintain positive expectancy since most trades make about 4% to 6% with a high win rate.

 

[Keanan]

Thanks for explaining that, Dan! I'm eager to start backtesting and practicing this trade

 

[DGH]

Keanan, you might wish to try setups with highly liquid ETFs for the 1 3 2 fly. If you launch on a down day you should have small risk with decent premiums. I have this set up on Power Options (poweropt.com) to find them easily. Best of luck to you.

 

[status1]

Wow it's been almost a year since the Boxcar started and continues to evolve
I am not sure if I should start a new post or continue with this one
I like this latest new and improved version of the boxcar that Dan presented in the latest TG1 presentation on Jan4 2022 and the spreadsheet that was shown
I have a question for Dan about the spreadsheet or anyone who could answer

I was trying to create a similar spreadsheet based on the one Dan presented and I got a lot of it done but something did not make sense to me
In the last column on the credit spread side there is a column called "Current net pcs p/l" and the values in the cells I am not sure what they supposed to represent or at least it doesn't make sense to me Could that be an error in the formula for that cell ?

In the cell 2 cells to the left of that in the column "PCS P/L Open Credit" there was a value at one point in the video showing 2022 on the first line and at that point in the column "Current net pcs p/l" column it was showing 20.22 which looks to me like that is 100 times less but I am not sure what that represents
I was wondering if perhaps that value was supposed to be divided by the lot number which was 6 for that trade so the value should have been 3.37 instead of 20.22 So the 3.37 would represent the current value of that leg of the spread which would make more sense to me

 

[DGH]

Good observation regarding the spreadsheet. The cells in the columns mentioned above do not contain errors (to the best of my knowledge) , but could be represented differently (see below). I create my rudimentary sheets "on the fly", and often do not "make them pretty" once they have become functional because they are for my use only and are not for sale or public dissemination. Of course, viewers are certainly free to construct their own sheets or similar versions. I try to give whatever information might be necessary to create a similar sheet for the viewer during a presentation Referencing the cells and columns described above, the cells under the heading "PCS P/L Open Credit" are used to calculate the current real time P/L of the PCS, based on the differences between the initial premiums and the current Mark values. The cells under the heading "Current Net PCS P/L" are used to calculate the Net PCS P/L value by summing the PCS P/L Open position plus any gains or losses from adjustments of the initial credit spreads. So, actually, the formula could be modified to simply add the PCS P/L Open plus the value of any previous gains or losses from PCS adjustments (calculated by hand calculator or by exporting the Trade History into another sheet for these calculations). In my personal array, I have a separate tab which I use to Copy/Paste trades from a Trade History export and thereby calculate the P/L of adjustments. In order to accurately reflect the real time P/L of the entire position the values of any P/L from adjustments must be added to the Open P/L of the position. A similar process is used for the Debit Spread side. I have checked the accuracy of the position P/L value against TOS values (by importing previous trades in TOS) and by exporting trades to TraderSync.com. Hope this helps.

 

[status1]

Thanks for the reply
I just wanted to follow up to make sure I understand

the cells under the heading "PCS P/L Open Credit" are used to calculate the current real time P/L of the PCS, based on the differences between the initial premiums and the current Mark values

I assume that includes the lot size ?
So based on that the Initial credit was 4.22 and the mark was 0.88 so the difference is 4.22-0.88= 3.37 and multiplying that with 600 for the 6 lot it comes to 2022 which matches the value in that column so I would consider that correct unless it's calculated some other way

The cells under the heading "Current Net PCS P/L" are used to calculate the Net PCS P/L value by summing the PCS P/L Open position plus any gains or losses from adjustments of the initial credit spreads.

Are you saying that the values in this column contains adjustments that are not on this spreadsheet ?
It just seems a little odd that the adjustment is equal to the PCS P/L Open Credit divided by 100 and the long side is also adjusted by the same amount
2022/100=20.22
-1467/100=-14.67

If there was any adjustment for the PCS would that be in the AG column or is that on a separate sheet ?
Since I did not see anything in column AG I assumed there was no adjustment so based on that I don't understand the calculation behind those values
Assuming those values are correct the result in the subtotal would be 5.55 Maybe the resolution on the video is not high enough but I don't see a decimal point in that value It looks like 555 which matches the PCS P/L Open Credit subtotal which includes the lot size

The Current Net PCS P/L values still don't make sense to me

 

[DGH]

I did not include the adjustment columns. They were "off the screen". You may have more success by building your own sheet from scratch using the principles I outlined rather than trying to drill down into my calculations which may not reflect your approach. If you chose to make your own sheet, the only piece of information required in order to calculate the real time position P/L is: Position P/L = (PCS P/L Open + arithmetic value of total PCS adjustments) + (PDS P/L Open + arithmetic value of total PDS adjustments). In my sheet the formula for my SPX Net Position P/L is: =AG37+AN37 where AG37 is the arithmetic net sum of PCS Open P/L + arithmetic P/L of PCS adjustments (if any) and AN37 is the arithmetic net sum of PDS Open P/L + arithmetic P/L of PDS adjustments (if any). Again, the adjustment P/L of both the PCS and PDS must be hand calculated or calculated by TOS export of appropriate trades into a spreadsheet. While I don't want to disseminate my entire sheet, I have attached my methodology for calculating adjustment P/L from trades exported from TOS into the sheet. It is important to exclude trades in which you are not interested...in this case, exclusion of the PCS, to make the calculation accurate. The value of $600 shown is my net PDS profit after closing a profitable PDS when the market was down. This value was added to the PDS P/L Open calculation in order to generate the Net PDS P/L. I subsequently rolled the PDS down, choosing strikes in the manner I described in the TG1 session. All positions, including RUT, were profitable at Friday's close. All SPX profits in both accounts so far were generated from the PDS profits. I expect the PCS to become profitable next week since the risk metrics are very favorable.

 

[status1]

I understand that you are calculating in the adjustments it just does not end up in column P which is the Current Net PCS P/L
The debit side is different and makes more sense maybe because there are no adjustments but it's just the difference between the Initial debit/credit and the mark Doing the same thing on the PCS the numbers do not work out the same in column P
It looks like all the adjustments in column P for SPX and RUT are adjusted the same regardless of the size or value It's just the value in column N divided by 100 I find it hard to believe that all the adjustments work out to be exactly the same

If you don't want to share the formula in column P8 I understand and you don't have to
I will just use the same formula for the credit side that is similar on the debit side which is the difference between initial credit/debit and the mark than add/subtract any adjustments to those values

 

[Kobus]

Hi

I really liked the trade as per the TG1 presentation video on Jan4 2022. My question related to this trade is what do you do when the market goes up? as you only purchase 1 debit spread. Do you roll the trade up? Or what is the best way to manage the trade in this scenario?

 

[DGH]

Hi Kobus. My Boxcar trade is monitored nearly entirely on my enhanced spreadsheet rather than the risk graph, which does not provide sufficient detail with respect to the interplay between the credit spreads (PCS) and the debit spread (PDS). My spreadsheet highlights the debit spread component in red when the PDS has lost 18% of its cost. This parameter can be easily changed to 20% or 25%, for example, if desired. When the PDS side of the trade is "hit", that means that the PCS side of the trade is making money. Since my net return depends on the balance between profits from the PCS and losses from the PDS (a "mathematical trade"), I need to adjust the PDS side of the trade when the market rallies sufficiently in order to prevent further PDS losses. In that situation, I simply close the entire PDS and monitor the market action. If the rally persists during the trading session, then obviously the PCS is continuing to make money unencumbered (hedged) by the PDS. However, I always add another PDS with suitable strikes prior to session close in order to protect the position against a strong overnight downturn. When I have received sufficient profits I close part of the trade to bank some profits and continue to monitor the trade, hoping to capture more theta. I do not roll the PCS side of the trade up. Downside adjustments are entirely different and are too detailed to be summarized here. Also, this strategy is heavily nuanced because I am only in the trade for about 5 or 6 days before launching a new one. Therefore, my protocol changes a bit with each passing day prior to expiration Friday. Hope this helps.

 

[Kobus]

Thank you @DGH much appreciate the feedback and makes sense what you are saying. I will backtest the trade a bit and "play" with the settings. Enjoy the rest of the weekend!