Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
You can't get bid/ask, so you have little confidence as to whether your trade would have happened in reality. That's maybe not too bad if your timescale is a couple of hours and you use market entry, but not if you look at tick data and your trade lasts 30 mins. Also I seem to remember that if your primary series is 5min (which you might want on your chart) and you add a second series of one tick to get the granularity, you then have to go unmanaged otherwise you won't be able to enter on that secondary series. If you switch 1tick to primary, you've lost what you're looking at - I may be wrong on that, but if not, it's a lot of hassle for something which should be simple)
I don't see how a trade duration has anything to do with if you have bid/ask or not.
What you probably mean is that you need it for scalping. I know that scalping is a bad idea, so I never code those kind of strategies. The only problem that I had in not having bid/ask was when I tried in Forex. In futures usualy the spread is minimal.
The problem of having bid/ask data is in not seeing your trades on a chart exactly as they executed. (On chart you show only last prise).
Actually the problem is more with market orders then with limit orders. With limit order if you choose default fill type. Then the order is filled only if prise is penetrated.
Of course you can have 1 tick secondary series and use managed orders. I do it all the time.
I would like to see an improvement it Chart Trader that would allow for quicker order entry. Currently one has to right click while the cursor is over a price and choose an order option. I would like to select the order type, be it stop limit or limit, whatever order types will be available, market if touched would be a nice addition.. after the order type is selected it would be armed and ready to be entered on a price level in the chart with a single click. Kind of like loading a shotgun. This would allow the trader to stay closer to the chart/price action and able to fire an order in moment. The right click/selction process at the time of order entry is time consuming and distracting, a bulky process for high frequency discretionary traders.
Well, if you're trading much longer the lack of a bid/ask is less problematic because firstly the trade has a longer time to enter and still be similar to what would happen in reality (with a short term trade, if it doesn't enter then the whole trade may be gone), and secondly the tick or two difference is a much smaller percentage of the overall expected movement.
What I mean is exactly what I said. I look at ticks and usually stay in the trade for around 30 mins. Doesn't matter whether you call it scalping or not. So, ignoring the sweeping generalisation, you're saying that because you never code my sort of strategy then my problem is somehow invalid?
It's not the size of the spread I'm worried about at all, so that assumption was wrong. It's actually whether or not the trade gets executed. To put it simply: too often in backtest a trade is either taken or not, where the opposite happens in replay because replay has bid/ask, and bid/ask is a more accurate fill criteria than price traded.
It's my understanding that managed entries and exits execute only on the primary series, except where you have multi instrument and you can specify the bars index for entry. If that's the case, and I'm happy to be corrected, then your order does not execute on the tick series and you're back to the original granularity problem, which was the original point about it being useless for testing those sorts of strategies.
Regarding backtests and optimizations and the result set generated, please cache all the tabs at the time the cycle is run.
Right now, if I select one run (say line 1 in the result table), it has to re-run the entire backtest cycle. For some of the stuff I am doing now, this can take several minutes (ABC hives). It is incredibly annoying.
Would be really great if the results for each tab were simply cached somewhere to prevent the entire cycle from having to be re-run each time you move from one result to another.