Missing Catalyst - December 14, 2011

[Billy]

Billy,

I was thinking to the very interesting example you showed us a few days ago, using 3 different (with 2 tighter) trailing stops for your triple leveraged position.

That could be a way to limit the loss, or even to have a little profit even if the official robot trade incurs into a loss.

The reverse of the coin is that long term cumulative performance still risks being deteriorated if compared to a "full" position.

Roberto,
That’s exactly what I described in my post below as a discretionary “choppiness” risk management strategy. My mistake was to switch to an aggressive discretionary “trend-following” strategy mostly based on too small a statistical sample from stage analysis. But I clearly mentioned to all at the time that it was not robot-related.
Billy

[TraderD]

Excellent idea! Now, how do you detect choppiness with confidence?
We have tested selling on targets and it consistently deteriorates long term cumulative performance.
Billy

This question breaks down for me to asking whether choppiness can be identified at trade pre-entry or at position post-entry. I believe that the decision to not enter on weak signals represents the first case and the accompanied belief that the expected reward isn't worth the risk (in relative terms, when compared with entering on strong signals). So now what about the second case where you already have a position and an apriori-known trade mgmt plan (which takes the form of EOD-based position exit rules, recalculated stops, etc.)

While the position is held, the market keeps on injecting new information that can be analyzed, not merely particular to the trade itself. The question I'd ask is, what would warrant altering the original trade plan and playing defense (profit taking) instead of offense (riding a trend). Is it the signal becoming neutral? Is it the position failing to "move" after X days? There's clearly a myriad of possibilities. I gather that you have tested some of them, but maybe a more dynamic market evaluation method could find that golden path of switching between offense and defense while the position is on.

$.02,

Trader D

[Billy]

This question breaks down for me to asking whether choppiness can be identified at trade pre-entry or at position post-entry. I believe that the decision to not enter on weak signals represents the first case and the accompanied belief that the expected reward isn't worth the risk (in relative terms, when compared with entering on strong signals). So now what about the second case where you already have a position and an apriori-known trade mgmt plan (which takes the form of EOD-based position exit rules, recalculated stops, etc.)

While the position is held, the market keeps on injecting new information that can be analyzed, not merely particular to the trade itself. The question I'd ask is, what would warrant altering the original trade plan and playing defense (profit taking) instead of offense (riding a trend). Is it the signal becoming neutral? Is it the position failing to "move" after X days? There's clearly a myriad of possibilities. I gather that you have tested some of them, but maybe a more dynamic market evaluation method could find that golden path of switching between offense and defense while the position is on.

$.02,

Trader D

Most interesting clues, and we will seriously try to backtest and improve if possible.
The main problem with choppiness is that it is very similar to randomness. No matter how much you backtest it, it hardly can provide meaningful edges for the future.
Billy