Why is this taking so long?

I tried 3 times to close a Natural Gas trade (yes it populated the trade window & I clicked Place Order). There was NO indication that the system recognized a close was placed or that the trade completed. This shouldnt take 5 minutes or more to close one Natural Gas trade. I am not trading Butter!!!



STO 1 NGZ6 7.790 10/11/06 10:20 0 7.450 (11:28t) $0 $8,000 $3,399 n/a No calc $3,399 Close Reverse

For heavens sake, it took 7 minutes to close!!!

Yeah. I’m investigating. The issue apparently was that NYMEX wasn’t reporting a valid bid/ask spread for 7 minutes. I’m going to try to make the system handle these cases a bit more gracefully - specifically, to ignore the bad “ask” if we only need a bid (for a sell order, for example) or to ignore a bad “bid” if we only need an ask (for a buy order, for example).



Of course, if you are trying to sell, and NYMEX does doesn’t report a bid, then we’re still SOL. But in this specific case, the proposed fix would have eliminated the problem. Working on it now.



MK

Of course like always, it cost me a few hundred the entire wait time. Why don’t these glitches ever seem to work out in our favor? “Gee, I made $800 due to a broker problem” :slight_smile:

> "Why don’t these glitches ever seem to work out in our favor? “Gee, I made $800 due to a broker problem” :slight_smile:



Actually glitches in C2 have made a few vendors $$$$$$$ on

vacuous trades.

One just tends to remember those times where it works against you. If one has only had a few of these trade-affecting issues, the results can be lopsided either way. Collective2 results are not any different from real results in this regard from my experience.



As someone who deals with all manner of human/technological issues that affect trades on a daily basis, I can assure you that over time they even out.



I tend to agree that many glitches tend to cancel out against each other in the long run. E.g. this week I had an autotrade delay of an hour or so. I lost $130 in one account with one system, but had $120 more profit in the other account with another system.



But logically the degree of compensation depends on the degree that the system has optimal timing. E.g. if the system has always the optimal time to exit a position, then every delay will lead to a suboptimal result. Also, it is said that many systems have a narrow margin for error and then I would think that the small part that isn’t cancelled out could eat a relatively large part of the profit. Moreover, both the positive and the negative glitches can decrease the Sharpe ratio if they are large (but I think this happens only if you over-leverage).



Even so, losing many $$$ unnecessarily spoils a few minutes of my day even if I know that it can go the other way next week.