market or limit orders for fast moves after the opening bell?

I see, you're using limit orders. I'm supposing that using an "immediate or cancelled" market order and resubmitting it for the remaining unfilled shares (as long as the price is still within my range) will give the same result.

this morning at the open i experimented by attempting to transmit a whole page of market orders for a range of stocks with very different average volumes and market caps. Nothing at all happened for the first EIGHT minutes after the bell. Then gradually the most liquid ones started going through, often only a hundred shares at a time, and by about 15-20 minutes orders for the least liquid of my selection started being executed. It took a full half hour for me to buy 1000 shares of about 20 different stocks, all with an average daily volume of at least a million, and a market cap of at least 500 million.

So my conclusion is that there is some correlation (since orders for stocks with the largest daily volume were executed first), but that presumably (hopefully!) it is incomparably slower than live trading, for whatever reason.

good luck with your trading
 
I know what you are saying. I paper traded with Investopedia and found that when I hit the button to buy or sell that even though it took about 20 minutes to respond it was at the price that I expected.
 
Quote from brynno:

good luck with your trading
More...

Thanks! We ran our ATS this morning in sim and all trades in the first 20 minutes never filled. Ironically, ALL would've have been profitable! We've decided to go back to 20 shares in the live account for testing. We did that a couple weeks ago and the orders were all executed properly. Each time we tweak the program tho we prefer to use the sim account, but since the best trades often occur off the open, back to the itty bitty shares in the live account.
 
right that's good to hear and i'm almost convinced. however...
don't you reckon as few as 20 shares would be filled instantly under any circumstances? i'd be interested to hear how you get on with larger sizes in the first 20 minutes when, as you say, there is no doubt the best moves do occur.

i mean, perhaps 20 shares might even fill near the open in the simulator! in which case nothing would be proven.

right. i'm going to do that tomorrow then. see how long it takes for 20 shares to fill in the simulation.
 
Quote from brynno:

right that's good to hear and i'm almost convinced. however...
don't you reckon as few as 20 shares would be filled instantly under any circumstances? i'd be interested to hear how you get on with larger sizes in the first 20 minutes when, as you say, there is no doubt the best moves do occur.

i mean, perhaps 20 shares might even fill near the open in the simulator! in which case nothing would be proven.

right. i'm going to do that tomorrow then. see how long it takes for 20 shares to fill in the simulation.
More...

We were only trading 100 shares in sim and no fills on the opening 20 minutes.

Try 20 shares in sim and see if it makes a difference.
 
I'm new to day trading (since last October) and there is much I'm missing but so far haven't noticed unexplained delays in order fills with the NT / IB combo in the opening minutes, either live or realtime/replay paper, mostly limit orders, from 200 to 1000 shares.

In addition I'm trading cross-listed stocks like RIMM & AEM on the Canadian TSX, which is relatively illiquid compared to US markets.

For the record the strategy that tends to work for me more often than not lately in the first 10 minutes is bracketing the early range with 2 OCO (one-cancels-other) buy stop and sell stop orders to capture any early range breakout; i.e., "buy high and sell higher" or "sell low and buy lower".

The orders are positioned using NT's ChartTrader with reference to any support/resistance lines (Gann-like MurreyMath indicator), previous day's OHLC, presence of whole numbers and any other number the stock seems to react to, and what the market makers appear to be up to (Level II displayed as a sidebar on the NT chart). I keep volume and stochastics panels open but rarely glance at them.

I try not to live trade any stock I haven't paper traded (real time or replay) for a week or 2 to learn its idiosyncrasies. This initial exposure to the stock lets me set up initial ATM strategies (number of shares, profit targets & initial stop loss, trailing stop parameters) based on S/R levels but also consistent with usual money management principles (e.g., minimum 2:1 reward / risk, potential draw-down, amount of capital at risk versus account size)

I try not to place market orders ever, since they seem a little undisciplined, the psychology a little shaky, the penalty being missing big moves occasionally. To me missing a big move just means I wasn't prepared, and (bitter) experience has taught me I shouldn't enter any trade if I haven't fully thought it through ahead of time. This habit comes from losing a lot of money over the years discretionary trading :-/

During the trade I'll also keep an eye on movement of the market and the sector as well as any underlying commodity and ETF (e.g., for stocks like AEM), and a finger hovering over the CLOSE button for stocks like RIMM that unfortunately more often than not seem to lead the market.

When it works this (early range breakout OCO) strategy typically yields $50 - $300 / trade and even more once in a blue moon, for 200-300 shares of a stock in the $60 - $70 range.
 
Quote from bnichols:

For the record the strategy that tends to work for me more often than not lately in the first 10 minutes is bracketing the early range with 2 OCO (one-cancels-other) buy stop and sell stop orders to capture any early range breakout; i.e., "buy high and sell higher" or "sell low and buy lower".
More...

Awesome strategy, and I use it a lot. That break out of the opening 5 mins tends give a decent push. My profits have been much like yours on 100-400 shares, with an occasional home run when you catch the start of a strong trend.

Quote from bnichols:

The orders are positioned using NT's ChartTrader with reference to any support/resistance lines (Gann-like MurreyMath indicator), previous day's OHLC, presence of whole numbers and any other number the stock seems to react to, and what the market makers appear to be up to (Level II displayed as a sidebar on the NT chart). I keep volume and stochastics panels open but rarely glance at them.
More...

I'm doing all this manually, because my current charts are not so good. I downloaded NT last week to evaluate for trading bracketed OCO stop orders directly off the chart, specifically to improve my ability to utilize this strategy.

I take it you're pleased with it? Any issues of note?

Quote from bnichols:

I try not to live trade any stock I haven't paper traded (real time or replay) for a week or 2 to learn its idiosyncrasies. This initial exposure to the stock lets me set up initial ATM strategies (number of shares, profit targets & initial stop loss, trailing stop parameters) based on S/R levels but also consistent with usual money management principles (e.g., minimum 2:1 reward / risk, potential draw-down, amount of capital at risk versus account size)
More...

I did this as well and now only trade 4 stocks. It makes life simple to wake up each day and know short term and long term S/R levels without having to look them up, and also to know proper position size for each setup, and where survivable stops can be placed. Less is more, IMHO.

Quote from bnichols:

I try not to place market orders ever, since they seem a little undisciplined, the psychology a little shaky, the penalty being missing big moves occasionally. To me missing a big move just means I wasn't prepared, and (bitter) experience has taught me I shouldn't enter any trade if I haven't fully thought it through ahead of time. This habit comes from losing a lot of money over the years discretionary trading :-/
More...

The more I thought about the question this thread poses, the more I realize that market orders in the opening minutes are a bad idea. As I said earlier, I actually use marketable limit orders when I want in early. I've used market orders a few times in the past, and on at least 2 occasions ended up with serious slippage on the trade.

I call fear of missing a big move "miss the boat disease". The main symptom is chasing entries and the sad effect of this disease is shrinkage of the tradingus accountus.
 
My opinion of NT may be slightly biased since I eventually bought the lifetime licence (it has to work). Still learning, and there seem to be a few quirks, but support has been great (and patient...found that out when I ported it to 64 bit Windows 7) and I'm pleased overall.

I chose NT over others, including Trade Station, because I wanted lots of features as well as the ability to chart trade and to program indicators and strategies, but mainly as a Canadian citizen wanting to trade Canadian stocks on Canadian markets the platform/broker combos are pretty limited; the IB / NT combo seemed to be the only option, and I like the $1 / 100 share commission since I prefer to trade fewer shares of pricier stocks.

One issue with NT may be that the Dynamic SuperDom (similar to IB's BookTrader) doesn't show any Level II info except for the inside bid/ask for stocks, although the data does print for futures and is available to NT. In fact the 3rd party indicator I'm using (jtRealStats) with ChartTrader prints Level II as a sidebar and appears to keep up with Level II in IB's BookTrader. Not sure if this will be addressed in Version 7, which is in beta (current version is 6.5)

Also, I still haven't figured out how to default the ATM strategy properly. This means when I switch from real-time paper to live trading, say, the selected strategy changes for some reason I have yet to fathom and I've got to double check that it's the one I want to trade. This has nailed me a few times :-/

Finally, absolute volumes seem to be wonky, but haven't attempted to figure out what's going on with that.

To illustrate the setup I like to use, I'll try to attach a screen capture. The content is a 3x realtime replay just now of Friday's data for AEM with my usual OCO early range breakout ATM strategy for AEM (didn't actually trade AEM Friday). The pic shows the situation after the trade completes: short sale accepted cancelling the long buy, the first target hit followed by the remaining trailing stop a little later. I just let the strategy run here but will often fiddle with the targets if it looks like I can squeeze a few more cents out of it, and if I'm feeling brave I might go after the retrace.

Note: price is CDN dollars and times are AST (1 hour later than EST)
 

Attachments

Just an addendum to the demo described above to show placement of a the initial OCO orders straddling the early range.

The orders have to be placed not only above & below the range limits but also with respect to the minefield of Thursday's high & close, Gann-type S/R levels drawn in by the MurreyMath indicator (e.g., "Top of Trading Range" and "Major Support/Resistance Line") based on statistics from previous days, as well as the whole number 60.00 and perhaps to a lesser extent 61.50 and 60.50.

The main issue in this example IMO is that the relatively wide (50 cent) early range makes choice of initial stop tricky if we want to maintain a decent risk/reward ratio, in particular given that AEM could easily take out either order and then reverse if we jumped the gun and the range is still setting up.

In terms of execution using NT, orders are placed by right-clicking on the chart at the desired price and selecting from the pop-up menu as shown in the pic. Order selection is followed by an optional order confirmation popup (confirmation can be turned off). Before it's filled an order can be modified by left-clicking on the price and moving it up or down, left-clicking on the quantity to change it, or canceled by clicking on the red X on the right hand side price axis.

Another way to play the opening range as it develops might be to place a single limit order with a very close stop in anticipation of where price is going to be and play it manually, trying to buy near the bottom of the range or sell near the top, placing one or more targets considerably outside the range so we're set up properly if the price decides to break out in the "right" direction (otherwise the stop saves us). This approach may require that you're somewhat more nimble and are quick to pick up on how the pattern is evolving, however. In this case for example it turns out the lower low and lower high in the first 10 minutes were predictive of a breakout through the bottom (see pic in previous post), at which point we would not have wanted to be long. In particular I'd avoid options like "auto-reverse on target hit" in the first 10-30 minutes since that may be building too much expectation into the trade, based solely on my experience so far that it (trying to be clever :-/) seems to be a recipe for disaster more often than not.

Finally, while I tend to trade on the 1-minute chart in the first 30 to 60 minutes I keep an eye on the 5- and even the 10- minute as much for the longer term S/R levels as price pattern.
 

Attachments

You must log in or register to reply here.
Back
Top