If you are wondering as to why does it happen, well there are tomes of academic papers on that, but it boils down to two factors largely, one is cost/barriers to trading in a given product, and relative volumes.
So to sum this point while type (a) is a fundamental reason for index moves, type (b) can also happen and may provide a minor edge to the participants.
Hope you all are still with me.
So far we looked at
1) What is an Index? 2) Why does it move/What causes its moves?
Now lets get to the topic of this blog post.
Market Profile and Order-flow Charts. I will not spend time in explaining the basic concept of Market Profile and Order-flow, that I did in the previous posts, do refer to them for the basics. But here I am going to explain Order-flow more than Market Profile. Both are unrelated but for some reason a lot of sellers and vendors offer them together. Let’s move on.
Let’s understand the process flow of an order, i.e. an order you put to buy or sell one lot.
Buy/Sell Order Placed by you ->
Order goes to Broker’s OMS (Order Management System) ->
Then goes to Exchange’s OMS ->
Finally reaches the CLOB (Central Limit Order Book) ->
Order now gets queued based on Price and Time priority (Depending on Market or Limit Order type) >
Finally once it matches another Buy or Sell order it gets executed.
If you notice I have not used the word order-flow anywhere yet in this sequence of events. The reason being, only after an order is executed we get to see the Order-flow Information i.e. Bid/Ask Volume traded at a given Price.
To reiterate Order-book (LOB - Limit Order Book) comes first and Order-flow later.
^LOB is the information that you see in the Market Depth window of your trading platform
So in a way Order-flow is stale info. Its all done and over by the time you see it (*Assuming what you see is what it is).
If someone claims that there is a “Predictive Edge” in Order-flow they are essentially claiming that “if X volume at bid or ask happens at a given Price” it means the price will go further up or down.
For a second ignore predictive edge, even a statistical edge will do? Show me one Order-flow based back-test and I would be happy to update my views here.
If this is not enough, you also need to understand how Level 1 Data Feeds work.
None of the feeds in India give you tick by tick data, we don’t have the infrastructure as retail traders to receive it, what data-feed providers like TrueData, GDFL and E-signal give is a Per Second Aggregate of Ticks. So what you see in an order-flow is an *aggregate information for a second or as some call it “Snapshot Data”. Which can never be accurate, to put it differently its not meant to be, there will always be “missing” info. in it.
And lastly always ask yourself, if Order-flow info had such an edge, why wouldn’t these indicator sellers keep it to themselves and print money.
Personally, I have used and tested both Order-flow and Order-Book information to the extent it’s possible with retail level latency and infrastructure and have not found any edge there.
So to sum it all up, Order-flow, if at all has an edge, it would be in a market with type (b) information flow, which we are not. And secondly, in markets with type (b) information flow, you might-as-well use Order-book Info. why would you want to look at stale order-flow info.?
This is all I had to share, hope it helps you, saves a bit of your time (by helping you avoid rabbit holes) and more importantly your money.
1) Does Index Futures Dominate Index Spot? Evidence from Taiwan Market - Ching-Chung Lin, Shen-Yuan Chen, Dar-Yeh Hwang and Chien-Fu Lin
2) Domestic and international information linkages between NSE Nifty spot and futures markets:an empirical study for India - Sanjay Sehgal & Mala Dutt