Why does the market move towards the heavier side of the order book?
Someone said to me recently:
I have observed that this tends to be the case as well, though I don't pretend to know how to use the DOM effectively to trade. Do you agree? And if so, why is it the case the the side with more orders tends to draw price to it? While spoofing occurs all the time, I'm assuming a relatively spoof-free environment for the sake of this discussion, although if spoofing is simply part of the equation of why this works, then please I'd love to hear your thoughts on it too.
The following user says Thank You to josh for this post:
Well, a lot of spoofing goes on and generally that involves stacking the order book on one side to fool people into trading in the other direction. At the same time they will be putting in an icebergs on the opposite side to scoop up the trades of the fooled traders. Then the size will flip to the other side, those fooled traders will be trapped and price will move through that large size as they exit.
Then again, there are times when the order book will just naturally be stacked higher on one side. At the high of the day it is normal for the offer to be much larger than the bid and vice-versa. This does not however mean that price is going to break higher.
Sometimes it's real, sometimes it isn't. Sometimes there's a big fish gaming other players and sometimes a bigger fish comes along and eats him.
Basically, it's a game. A game that requires particularly large 'spheres'.
So - you are on the right track, there's just a lot of games being played. There's also some attempts to make algo's jump. You will occasionally see (on the ES) the order book flash up 6000 or so on a level and then it'll disappear again. That's not there to make you & me react, it's there to fool the algos.
As well as the order book, the time and sales is needed too. The DOM is a statement of intent, the Time & Sales represent actual trades (against that intent). So - combine the two and you have a good toolset to assess the gameplay.
The following 8 users say Thank You to DionysusToast for this post:
But the demand you talk about (16 on the bid) is passive demand. Demand shows itself, IMO anyway, as active buyers, that is, market orders buying the offer. At the very least, in your example, is supply not stronger, as sellers aggressively unload (i.e., supply) to passive buyers?
There was a lengthy discussion on this a few years back on ET started by a guy named Viper Speed Trader. He developed a strategy based on the cumulative volume of the first 5 levels of the orderbook on both sides (he referred to it as ACV). When he saw that one side of the orderbook was twice as large as the other, he would scalp toward the direction of the size. You might want to check out his thread for more details/examples (search for "VSTscalper" on ET).
Someone else created a indicator for him to plot the ratio of the cumulative bids vs. cumulative asks. You can see in this example that when price is moving down, the indicator (3rd panel) does in fact show that the cumulative bid volume (1st 5 levels) is twice as big as the cumulative ask volume (e.g., the reds bars are hitting "-2"). http://www.whitmarkdevelopment.com/projects/images/VST%20Indicators.jpg
After reading the ET thread a few years ago, I tried out the indicator for a few weeks, but ended up ditching it.
Not sure if this info helps your question, but just passing it along since I thought it was an interesting topic when I first came across it.
Last edited by Haverchuck; May 18th, 2011 at 03:35 PM.
The following user says Thank You to Haverchuck for this post:
According to their explanation, price often trades towards size displayed on the smaller FX networks they supply data for, in response to movements on the larger networks such as Bloomberg, EBS, and Reuters. It would make sense for FX futures, which are quite small in volume compared to the large cash networks, to be used the same way.