Sluggo's Remorse Ratio

Discussions about personal psychology for the individual trader.
Post Reply
alp
Roundtable Knight
Roundtable Knight
Posts: 309
Joined: Mon Aug 27, 2007 8:09 pm

Sluggo's Remorse Ratio

Post by alp » Sat Apr 25, 2009 10:24 am

I have found that sluggo's remorse ratio, practical money management or decision making rule of thumb, very useful both for discretionary trading and money management.

By balacing the seemingly conflicting greed and fear feelings the decision making process gets smoother and clearer. At least it looks like so for myself.

When using the remorse ratio I basically divert away the attention from being right or wrong to thinking about how much remorse I might feel if I don't implement the desired action and I end up being right about the outcome and also if I do implement the desired action and I end up being wrong about the outcome.

I find it very interesting, insightful and practical, contributing to a clear mindset.

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Sat Apr 25, 2009 11:25 am

I suspect these articles may have been alp's starting point: (ref 1), (ref 2) .

Neither of them mentions a "ratio" specifically. Rather, they perform what quants call a "minimax optimization": find a policy which minimizes the worst case remorse. (i.e. a policy which minimizes the maximum remorse among all cases. Minimize the max --> "minimax")

The basic idea is "When in doubt, do half; that way you can only be half-wrong." I didn't invent the concept; I learned it from video tapes of a lecture series by an Original Turtle, who said that Richard Dennis taught this concept to the Turtle students. (RD majored in Psychology at university.)
Last edited by sluggo on Tue Oct 26, 2010 9:55 am, edited 1 time in total.

alp
Roundtable Knight
Roundtable Knight
Posts: 309
Joined: Mon Aug 27, 2007 8:09 pm

Post by alp » Sun Apr 26, 2009 10:32 am

sluggo wrote:Neither of them mentions a "ratio" specifically. Rather, they perform what quants call a "minimax optimization": find a policy which minimizes the worst case remorse. (i.e. a policy which minimizes the maximum remorse among all cases. Minimize the max --> "minimax")

The basic idea is "When in doubt, do half; that way you can only be half-wrong." I didn't invent the concept; I learned it from video tapes of a lecture series by an Original Turtle, who said that Richard Dennis taught this concept to the Turtle students. (RD majored in Psychology at university.)
Yes, thanks. I find that by considering the potential worst case remorse of both decisions I divert attention away from being right or wrong and focus instead on risk management. I find it even helpful for bet sizing on a discretionary basis. The "when in doubt, do half..." is equivalent to decreasing the desired bet size, by half. I mentioned ratio because it seems that the policy leads us to find the best compromise between greed and fear.

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Tue May 05, 2009 2:59 pm

A book by M. Gunn, copyright 2009, (link), contains Richard Dennis's "when in doubt, do half" advice, stated this way (on p.23):
I have been working in the currency markets for a long time and that has encouraged some of my friends and family to ask me whether they should buy their foreign currency for their upcomng holidays (for some reason they think I will be able to "forecast" what the currency will do!). My stock answer is that I, like everyone else in the world, do not have a clue of the exact path a currency will take between now and your holidays so you should buy half the currency now and half just before your holiday. In that way they are perfectly hedged. If the currency goes up, they will feel relieved that they bought half, but if the currency goes down, they will feel relieved that they are still going to buy half at the cheaper rate. Common sense, but as someone said, common sense is not so common.
Gunn speaks of feeling "relieved," the emotional inverse of feeling remorseful. He advises a policy of being half-relieved (and therefore, half-remorseful). Same as Richard Dennis.

alp
Roundtable Knight
Roundtable Knight
Posts: 309
Joined: Mon Aug 27, 2007 8:09 pm

Post by alp » Sat May 09, 2009 10:19 pm

Thanks. I guess I have taken it to another level. I think of it as a practical way of reconciling two seemingly opposite feelings (such as greed and fear) so as to achieve a clear mindset ( or "common sense").

RedRock
Roundtable Knight
Roundtable Knight
Posts: 941
Joined: Fri Jan 30, 2004 3:54 pm
Location: Chicago

Post by RedRock » Sun May 10, 2009 10:20 am

alp wrote:Thanks. I guess I have taken it to another level. I think of it as a practical way of reconciling two seemingly opposite feelings (such as greed and fear) so as to achieve a clear mindset ( or "common sense").
But then, "common" sense is not necessarily the most profitable. Perhaps this could be tested.

alp
Roundtable Knight
Roundtable Knight
Posts: 309
Joined: Mon Aug 27, 2007 8:09 pm

Post by alp » Sun May 10, 2009 10:58 pm

RedRock wrote:But then, "common" sense is not necessarily the most profitable. Perhaps this could be tested.
Sure, thanks. I have been doing some discretionary plays. :)

On the other hand, not a topic for this thread, but as far as mechanical trading is concerned, I also hold that such a 100% pure mechanical system doesn't exist. There is a lot of decision making in the process, apart from the emotional strength to follow it.

RedRock
Roundtable Knight
Roundtable Knight
Posts: 941
Joined: Fri Jan 30, 2004 3:54 pm
Location: Chicago

Post by RedRock » Mon May 11, 2009 12:09 am

hmm. I'm reminded of the argument presented in the book Inside the Mind of the Turtles ... master risk. Where the decision should be judged not on the outcome of the event. But on the merits of what information was available at that point in time when the decision was made.. How the best decision given the known facts, may turn out bad. Yet, it was the proper decision.

What I am curious about, is aside from the psychological need to do something, the agony of doing nothing. Is doing half, more likely to yield a positive result than doing nothing, or everything. If it is, then fine. If not, and the net result is the same as consistently doing nothing/everything. What are we gaining by doing half. Is this a crutch to compensate for a human weakness.

ahh, thats why I only dwell in the rule based world. such ponderances can usually be reduced to probability. More difficult to evaluate when the mind is the initiating decision filter.

But then again, as the two handed accountant once said... :?

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Mon May 11, 2009 7:43 am

RR, I agree. The "when in doubt, do half" suggestion is explicitly a pre treatment for soon-to-be-wounded feelings. A salve for human emotions. Richard Dennis supposedly said although it is bad to occasionally deviate from the mechanical rules, it is 100X worse to become "emotionally destabilized". When destabilized, traders often abandon the mechanical rules entirely, with disastrous results. (In poker, the expression "Going On-Tilt" describes the same mental state and its always-disastrous results.) The "Do half" suggestion temporarily gives you permission to cheat on the rules a little bit, in situations that are incredibly painful and potentially destabilizing, to preserve your sanity. You cheat just enough that you remain stable, so you can go back to following the rules.

Think about it: "when in doubt, ... take this prescription ... " Theoretically, when is a mechanical system trader ever in doubt?
Last edited by sluggo on Tue Oct 26, 2010 9:56 am, edited 1 time in total.

alp
Roundtable Knight
Roundtable Knight
Posts: 309
Joined: Mon Aug 27, 2007 8:09 pm

Post by alp » Sun May 17, 2009 3:38 pm

The mechanical trader executes her system faithfully, flawlessly and, more important, emotionlessly. There is no emotion involved in the quantitative, programmed algorithm.

So, unless a T-888 terminator-like trader designs and runs the system, eventually some emotions might try to express themselves and render the system's pilot emotionally destabilized. This could be out of a losing streak, a winning streak, out of having nothing to do, etc. (The emotions could have already been part of the design process as well, when the human-trader decides on the rules, the backtesting procedure and chooses the optimal values and portfolio).

How can machine (system) and human become one? Or otherwise, how can the trader preserve her sanity?
Attachments
Mechanical Trader.jpg
Mechanical Trader.jpg (58.19 KiB) Viewed 10741 times

Post Reply