Ed thats an interesting account but I believe that you may be making my argument just a little if you look at the text carefully. Your experience with your buddy was a single instance; you would have to log tens of thousands of those for it to be important and a few of the interesting things we see when traders (be that they hedge funds, CTA's or private) is, they don't recount what doesn't work or when they fail. Part of the issue with the Hedge Fund sector right now is that we can't get a great deal on the attrition rate because many funds just simply stop reporting... We never know if they have gone belly up or what's happned.
Another thing we all do as traders is hang on to whatever affirms what we think, even if we don't have great evidience that it works. There are no doubt many traders here who strongly feel that pschology is extremely important in trading, they may not know why, or they read something and it made sense for them and they have embraced it. That's not a terrible thing, but it can be (if it costs you money in the conclusion ).
This is what we should really be asking ourselves, if we want to demonstrate that plogy (and maybe talking to stuffed animals) has merit (I will tell you what I do in a moment )... How can we trade it and how can we test it.
Lets presume that we all, as a team can agree on a series of plogical activates in the market (it will help if we've got somebody with a PHD in the field), We can probably use some of the high volatility events like NFP and identify price action before and after and then assess the market reply as positive or negative before and after as well, from that information we should be able to extract both long and short term success of the markets reaction to the release... Now to be sure this will become really complied (as it should) because you're going to have additional pscholigical events before and after that also impact the audiences behavior and the significance of the release (you should already be seeing the difficulty with these versions in the market).
I have tons of information and have access to more; enough to derive some thing statistically important, so I'm more than prepared to help in the job.
Letsn't wonder argue, lets version, because in the end the one thing that thing is... Can you trade it. We'll end up with something which identifies reaction or a specific activity is tradable.
Again I think we need some type of pro to help us identify the plogical activates... Someone who knows Conservatism and Representativeness in addition to the dynamics of Belief-based Models. Once we've got a mechanism for accurately (or somewhat accurately) Identifying the trigger, we can then extract the correctness of the reation as time passes.
If you're interested in digging into it let me know...
What inanimate object do I speak too? My keyboard; I maintain about six spairs from the closet for all those conversations. LOL
Alright... less disagreement without evidence, let us grind out some number (in between trades of course). Even in the event that you disagree with the outcome, you'll have facts which you invent a real argument from and can pick.