Epic

Well, it was a horrible week last week. Was not at my computer for a couple days with many trades still open, and Murphy's Law took over. On the one hand, leaving trades unattended was pretty amateur. On the other hand, I discovered a spot where the strategy needed some adjustment to handle certain events. Also, there was a missing portion of my algorithm that I didn't realize was missing until this event. Took a couple days, but it is included now.

Always have to remember that just because something hasn't happened before, doesn't mean it won't happen ever.

I wanted to post a snapshot, but for some reason the numbers coming from IB are inaccurate. Probably due to the way that they are calculating the value of a hedge. Consequently, the snapshot makes things look better than they actually were last week. I don't feel that it is an accurate representation, so I will just give an update.

The two days that I was gone there was a drawdown of 3% each day. I did not close the losing positions because after they ran against me, the all presented setups in which I would open a new position if I didn't already have one there. In the end, I've recouped about 1/2 of the losses so far this week. We'll see where I get to by week end.

YTD return = 46%
 
Covered 2/3 of my positions today. Almost had 80% covered but then es jumped off support and missed me by a couple ticks.
 
I should also point out that the main reason for the severity of the losses last week was due to extreme divergence between the core positions and the supplementing hedges. This divergence was on the order of three sigmas and at max drawdown I was only utilizing 1/3 of available margin.

The three sigma divergence creates a very high probability stat arb. In this case, price action on the core positions takes a back seat to mean reversion of the hedges. The outcome is that I will likely see this type of event once yearly, and as long as I've got dry powder to avoid margin call, mean reversion takes over and max drawdown remains statistically theoretical.

Backtesting suggests that I would be able to average into the position throughout the divergence period and come out with a nice profit. To this point I have opted to avoid averaging in during such an event and simply adjusting targets instead. This ensures that I will not profit from the event, but it also creates a smooth risk profile over the long term and guarantees against blowup.
 
2r2unoi.jpg


This week went pretty much as expected. I'd hoped to be all the way back to previous highs, and it looked like I would get there this morning, but then things slowed down for me and I didn't get a whole lot of trades in during the afternoon.

Last week should be the largest peak/trough drawdown that I see this year. We'll see how next week goes.

YTD return = 64%
 
This range bound stuff is pretty boring... No money to be made this week if this keeps up.

Most opportunities currently accompany a bear move... so that is where the focus is.
 
Some end of day analysis indicates that conditions are right for increased size here. Intend to double the size of trades tomorrow, barring a large overnight gap. The positions might then move against me temporarily causing a larger than normal drawdown (roughly 5-6% worst case), but the resulting gains far exceed the risk which is well defined.

This is an example of conditions in which my strategy and the associated high sharpe ratio allow (and in fact somewhat require) increased leverage.

I should note that this is not a contradiction to an earlier post suggesting that I don't increase size based on higher probability. I'm not increasing size based on higher probability of profit, but on lower risk conditions in the underlying. All sizing is based completely on accepted levels of risk and target profit remains constant on a per contract basis.
 
This week was incredibly slow but toward the end of the week I was able to start increasing size on the positions. The upcoming week should be a little bit more entertaining.

Also I checked and the snapshot from IB for the week that was messed up is working now. Not sure if anyone is actually following this thread, but if you are and you want to see it, then I'll include it upon request. It's not gonna change the overall numbers though.

YTD Return = 67%

2lu8mrp.jpg
 
Well, a couple of indirect hedges are completely not working out right now. That is certainly the risk with indirect hedges, as they don't necessarily maintain their statistical correlation all the time.
 
Going quite well thanks.

Returns since the last update were as follows.

May = +5.35%
June = +8.20%
July = +5.16%
Aug = +9.82%

YTD +92.2%

I realize that usually when someone abandons a journal it is because of poor performance. That was not the case here. Recent business developments have prevented me from participating much on public forums. I apologize for getting this thread started and then not following through. I'm going to be trying to stay under the radar as I've realized that I don't really like the attention. If anyone is interested in my progress, feel free to PM me.
 
You must log in or register to reply here.
Back
Top