Trading the Gaps

Hello all:

Overnight gaps are my main trading setups. There are two types of setups that I use. The first I call gap-scalping, and it occurs premarket. The second is a straight gap setup, where I enter within the first 20 minutes of trading, and scale at key gap levels, such as half, full, or one and a half--to be sincere, there are not many times I go for gap and a half as it can get pretty hectic.

Anyway, I assume there are a whole lot of people out there who may know a lot about gap behavior, or alternatively, that may want to learn more on the subject. Consider this thread a place to ask questions and answer them as well. I also have a blog where I discuss [what I think is] insight during my admittedly still dilettante trading. Blog link is http://tradingfuturesblog.blogspot.com/.

This spreadsheet also has tracked my trading activity (per one contract) since August 31 (time when I discovered Google Docs). These are trades I have actually taken fully with at least one contract. What it does not account is the various scaling of multi-lot entries, as it is too much accounting and frankly, would be too much information for anyone but me. Hope it's worth a look! http://spreadsheets.google.com/pub?key=to0HdImpeQmE8OTzctxR_Vw&output=html

Let the discussions begin! Happy trading!

Ilir

By the way, just closed my gap setup for the day. Entered short at 1099.75 at 9:36am and exited at 1096.50 at 10:12am. For some reason, my console showed gap at 1091, but based on my notes on yesterday's close, it was about 1095.5, so I chose exit according to my notes.

Short setup just completed. Short at 1103.5 at 9:00am, out 1100.25 eight minutes later. Missed higher short entry at 1105 on the economic release-induced high, but scored a decent entry nonetheless.

ilir.shkurti said:

...The size of the gap, the trading volume in the market, as well as a myriad of fundamental factors lurking in the market's radar will influence how strong price movement will be in the direction of the previous session's close...

Click to expand...

More on http://tradingfuturesblog.blogspot.com/2009/12/closer-look-at-gap.html

Just exited what may be my only transaction today. Got short at 9:24am at ES=1109.50 and exited at 1106.25 at 9:36, my regular target. May regret leaving money on the table, as gap seems poised to fill, but comparatively low volume yields me to believe it may be a choppy day.

I stayed out because I thought it wouldn't fill today with the pre-market hoovering. Just goes to show you don't always get it right.

Technically, gap did not fill, fell three ticks short. Which is why my initial target is 3.25 points (if gap is > 3.25) and second target is gap less two ticks. Premarket was almost dead, and it was a very slow reports day as well. Which means, gap conditions are generally smooth.

I go for a fill < 4pts or close out by end of RTH. If > 4pts, I'm suspect of it filling and don't think the reward is worth the risk on the trade. Each to their own as they say.

I agree on gaps larger than 4 points. However, unless they come as part of a key reversal, such gaps are still good for fading with reduced targets, i.e. less than gap.

Today the size is just right. There are too many economic reports between now and open, so it may 'close' and reverse direction before open. Looking to enter somewhere around 1104.25-.75, but we will see.

Do you half your risk for the half fill?

Well, this sucks! Got in short at 1104.25, resisted position going as much as two and a half points against me, only to mess up on a technicality and cancel a stop. Flattened trade by panic (no stop!!) and got out with three ticks profit. It's days like this that upset me much more than a straight-out stop--lack of discipline.

Long 3.75pts target on open, came out with 2.5pts after it looked like it had peaked at 3.5pts.

Think that will be me done for the day as I can't see anything hitting levels for reversals/breakouts on the channel yet.

You did not fare bad.
I just took a look at the market (had logged off for the day) and noticed my full target and more would have been hit nicely without triggering my stop, which had been at 1101.25. Goes to show everyone never to mess with stops like this amateur (me!) did .

or more importantly have confidence in your target and leave the trade to run.

to be fair I get that wrong a lot too.

Reason I moved stop was because I was seeing resistance against my targeted direction, and wanted to tighten stop. I have over a year worth of data comparing my profit factor on transactions with and without this active management due to perceived resistances and supports. While my original parameters would have been more profitable in the aggregate, I have often reduced or eliminated would-be full stops.

Did not trade today, as I just came back from some chores, but interesting market action. Would have expected CPI to have much more of an impact. Market is factoring in some Fed speak ahead of the 2:15pm speech. Gap may not happen today after all, and its too late for any gap setups in the day, anyway.

ilir, have you done any specific research for gaps on fed days??

I have not, jm99. My data is still shallow to have any statistical significance, about a year's worth. Other gap experts do not see any significant relationship either, for what it is worth. Gap guys I know avoid days such as today altogether specifically because they disrupt the validity of the gap.

I as well find it very hard to come up with an effective way to incorporate any fundamentals (economic reports) into the gap play. The closest I have come--and it has worked for me a couple of times--is to put a limit very far from the normal price range. If the market influence of the release is such that price will penetrate this unlikely level, it is likely it does so on a candle's tail. As soon as it does so, it is very likely to retract. I have been filled like this a handful of times. The flip coin is that, short of such an influential release, you may end up sitting out an otherwise smooth regular setup.

Otherwise, releases may be used to piggyback on ongoing gap behavior--if, and only if, there are no other major factors disrupting the gap. 8:30am is such a moment, where releases often cause a gap soar of 3 to five points with no follow-through to speak of, which means that even if gap does not close completely, you have a 3 to 5 point fading opportunity. And honestly, that is all I would want out of a gap setup anyway.

Long answer to a short question, sorry.

Just recently started looking at gap trading as an adddition to my other trades.
Just a starting point and may wind up in the trash but first look and back testing looks to be possibly profitable.
using 5 min bars on ES
If opening gap = 4 or less (even if very small gap), right at open bell enter stop limit order 2 ticks towards gap close (1/2 point from opening price towards yesterdays close) . Stop is 3 points, target 2 points (yes sounds rediculous but...)

If gap is greater than 4 points then enter stop limit at low (gap up) or high (gap down) of 3rd bar close.
Hence if gap doesnt try and fill( runaway) then no fill. Also enter stop limit at high of 3rd bar close to play the runaway using same 2 point target 3 point stop(one direction should fill).

Yes of course there are many variables to consider such as multiple contracts, stop and targets and trail a portion to cover those fairly frequent huge runs on both the gap close and runaways. But I am looking to set a base strategy first and then enhance.

Some rough data from backtesting 9/24 through 11/13 - 55 trades (2 no trade days) 40 2 point winners, 15 3 point losers) net 35 points. Sounds a little scalpish (less than a point per trade average) but with some fine tuning may be able to enhance.
Caveat - backtesting can be misleading, on some of the close calls I had to go to 1 min or even 89 tick bars to see the order high and low was hit. Could have made some mistakes, when questionable I counted as a loser.

PS
Ilir can you increase the font on your google cloud sheet, as I can hardly read it on my commodore 64

razor99 said:

Ilir can you increase the font on your google cloud sheet, as I can hardly read it on my commodore 64

Click to expand...

Done, razor99.

Your approach sounds a whole lot more sophisticated than mine. Interesting results on back tests. Your two points are good as a first target, but would be very conservative as an exit for entire position, at least based on my live experience. Gaps have a surprising tendency to make overt moves towards closure in the first hour of trading. I admit my stop of 3 points plays too close in many cases, which is why most my entries are premarket--you may get a much better entry than right at open.

Today's was an interesting gap. I sat this session out, and likely would not have participated, or at least hung around that long to wait. But, a 3-point stop would have worked great (hindsight, right?) with a decent entry.

Exited first trade at 1099 seven minutes after open. Followed long entry at 1095.75 at 8:40am. Entry was a retested low of premarket price action. Took the trade after downward pressure from initial jobless claims ran out of steam.