Hi
is there any way to know the overall percentage of people buying a pair and selling it (LIVE) ?
Yes - half are buying and half are selling. But if you want order flow, its not available to retail traders like us.
Oanda shows it for their traders. But this is only a part of the market. Also available to non-Oanda people.
If its level 2 order book that you are after, its not much helpful in the FX market, normally. Most big orders come from institutions who are moving money around, and since these orders can impact the price movement adversely, they break large orders and route it in small fractions across multiple brokers. Each broker holds its own book, which for some ECN brokers is accessible on their FIX or other API based platforms. However, this book only represents a fraction of the global dynamics (though under efficient markets all brokers converge to the same price due to several parties looking for arbitrage opportunities). So even if you have the access to a local order book, it would not help in predicting anything.
However, if you are either an institutional trader or a hedge fund with substantial financial and technical resources, then this data is useful for arbitrage strategies (in which case I presume that you would not be asking this question). Some time back I came across a data feed which collects level 2 data from multiple brokers and provides it for a price. I do not remember the name, but google should be your friend.
And now the more generic part: if you are trying to read market emotions, I prefer to rely on news feeds, volume analysis and common sense for that.
It is not possible to get accurate volume data for the whole forex market, because it is decentralized.
That is one advantage stock traders have over forex traders.
The volume figures offered by various brokers are always only those of the broker’s own orders or at best those of the broker’s LP.
Cheers,
P.
It’s easy, When the price is going up the majority are buying, when it’s dropping then the majority are selling.
Just an observation, if half are selling and half are buying the price will not move.
Well it also helps to confirm your thoughts on the direction of the market. If you see the opinion of a 1000 traders, that helps to soothen your uncertaintiy about your own opinion. So, it can be helpful, but it should not lead you.
Obviously, yes.
But that’s not the point.
The idea behind VSA is gauging the strength of a movement: how fast price will move and how far it might go.
There is an interesting thread dealing with this topic here on Babypips.
Cheers,
P.
Always half are selling and half are buying, the movement of price it is given by the flow of orders as CodeMeister said
I have to go with codemeister and yunny on this.
Half buying and half selling could be all liquidity takers buying and all liquidity providers selling, all liquidity takers selling and all liquidity providers buying or more likely some mix of the two.
As to the order flow end of things when the liquidity takers are predominantly on one side of the market (initiating buys), it tends to push the market in that direction (up) as long as there is concerted buying. When the liquidity takers run out of gas to keep buying, the price will naturally start dropping because the liquidity providers will tend to be net short and will adjust their quotes to favor their net position in the absence of concerted buying.
For more information regarding the roles of market participants including liquidity providers and takers, consult a market microstructure text, such as Trading and Exchanges by Harris.
By the way, I recently read a paper that showed evidence that trading volume was not (primarily) causative in moving price! I realize that this inconvenient study will tend to make the VSA / market profile proponents do back flips! 
Could you provide a link (if you saw it on the net, that is, and not in paper form)?
Would be interesting to see how valid the evidence is.
It seems logical that a 100k buyers/sellers will move price faster and further than 10k buyers/sellers would or could.
That’s the whole idea behind supply and demand, after all.
There’s more to volatility than volume
By the way after reviewing the conclusion I need to walk back my previous assertion a bit:
We do not dispute that volume and
transaction frequency affect prices, but for these data other effects are more important.
Even Einstein could not understand this:15:
Yeah, all they say is that volume isn’t the only factor influencing volatility … which isn’t exactly news.
In addition, their study refers to the stock market, not forex.
lol, it’s just maths.
The only important parts are the introduction and the conclusion … the rest is long-winded evidence.