Sure point one you have already made for me:
Thinking heads is more likely after a series of tails is a text book example of a Gambler Fallacy.
Sure they are not coins but we can think of them in the same way becuase we calculate an estimated future expectation. In a coin-toss we KNOW it’s 50/50 per toss but with a Darwin we have to estimate and I submit that whether or not it is a recent Darwin high or low is similar to whether or not the coins have been giving heads or tails recently. They both can produce an illusion. They are both prone to Gambler Fallacy.
Second point I’d like to make is I want to clarify the context of this discussion to be about Darwins with a more or less positive expectation that are not “manufacturing” a smooth equity curve. Darwins with enough statistical significance and that we correctly expect to be profitable. For instance I want to leave out Darwins that are too young that have a recent run up. These lucky young Darwins may be more likely than not to be losers (and here is a potential high probability of a DD immediately after a run-up) so we must leave them out.
The third point I’d like to make is lets leave out the argument about parked money is not making money. Lets assume we have a large diversified portfolio already and have found two more qualifying uncorrelated Darwins with the exact same “stats” but Darwin A has had very recent DD and Darwin B has had a very recent run up and we can pick either or both to add to our already large diversified portfolio.
The fourth point is the Darwin price. The price is not a value oriented price, it does not go up or down based on the way other traders are buying and selling it. It is a marker of sorts for calculations such as P/L. Buying a Darwin is not like a traditional “trade” as we are not trading with anyone, there is no supply or demand. (EDIT: well there is but it’s not connected or effecting price, it is capacity) Buying a Darwin is just giving a trader your money for him to trade and selling the Darwin is terminating that agreement. Darwin price at any given time prior to entering that agreement does not matter at all in this context EDIT: This does not rule out market cycle timing though!
The fifth point is about market cycles. The most arguable point:
But then how do we know how long is the cycle? Where will the current cycle end and anew one begin? Will the trader herself eventually compensate for such cycles? I think the trader herself is in a much better position to identify and adapt to (or not trade during some or develop a filter for) market cycles as a trader knows her strategies best and she can analyse the underlying market structures of a single strategy (and adapt or incorporate a filter) instead of just analyzing the equity curve of her entire strategy portfolio (within a single Darwin). Is it possible an investor can identify and exploit a weakness in a traders ability in relation to market cycles just by looking at the equity curve of the entire Darwin? I guess it’s not impossible but it seems extremely unlikely and each exploitable Darwin would have different degrees of exploit-ability and different ever changing market cycle timing. Adapting to such cycles is one of the largest challenges a trader must face, why should an investor attempt to do it isolated from the strategies and underlying markets? I’m suspicious that any testing “proving” otherwise is flawed by lack of data or is outside the context of “safe” profitable Darwins. EDIT: I can see now how some Darwins may be somewhat time-able if they are very one dimensional but…
In the context of profitable Darwins I would not expect lucky short term gains to be necessarily replicated right after a lucky streak BUT I would not expect a DD necessarily either. Instead I would expect… the calculated estimated expectation based on ALL the data!
In the context of equally profitable Darwins, any such Darwins on a recent high or low shouldn’t matter. We do expect smaller DD and risk if we are properly diversified in to more Darwins but we can ignore whether or not they are on a recent high or a low.
This seems absolutely correct to me. On a side note I would generaly try and enter when the trader is flat as in has no current open positions. That would be the only timing I can think of that is logical.