If you reframe your questions, you’ll get more relevant answers. Opinions won’t change your perspective in any meaningful way, especially if you do not have any distinctions that gives you basic relationships to understand orderflow. You have a fundamental belief that bots can outperform a skilled trader. This belief is limiting any progress you can make in understanding orderflow. Observing changes in orderflow is a distinction based on a tool, one of many. As all tools are made for either general or specific purposes, best to understand the necessity that caused the tool to come into being. As a tool, it also complements other tools in a toolset. As a tool it has it’s own specific benefits and limitations for use. However, unlike most tools which are intuitive to the mechanically inclined, market specific tools require a spectrum of differentiation to be built in the mind. Some approach the differentiation process through trial & error, others more systemically. Others will not do what is required to compare and contrast enough market data to transform ‘noise’ to create ‘signal’. Not all bots are created equal, each genetically coded by the distinctions of the programmer. Consider the dynamic of eagles against drones. https://www.google.com/amp/s/www.wa...drones-france-is-destroying-them-with-eagles/ Be an eagle or be a drone. Either way one has to choose to choose.
why would anyone other than market makers want to publicly announce their pending orders by putting it on the order book
this is a really good post. well, from my experience i can tell you the following: the difference between candlesticks and the DOM is that the DOM Shows you what is beeing traded and how that is beeing traded. this is the Major difference. if you can see Patterns which increase the Odds in your favour than that would make difference in your trading. Regarding to icebergs: from my experience in the US treasuries, EU50, Bund/Bobl and ES i can say that most of the time you will always see icebergs at the extremes (e.g. extremes of overnight session). is there any rhyme or reason for that? well, i dont know as i dont know the market participant who is creating an iceberg order. i dont know his strategy (e.g. is he taking Profit? does he want to fade this Level with his iceberg order? is he spreading? is he an market maker? is he hedging his bet?) the Thing is with These order book Patterns that you dont know the Intention of its creator. but if you say that you saw this type of pattern 50 times and 40 times you could scalp 2 ticks out of it and the other 10 times you would loose 2ticks, than i would suggest that you exploit this pattern and you have an edge in this Kind of Play. it is good to know what Kind of orderflow games are beeing Played at the financial markets but you have to consider what a few guys here already wrote: the hft s are looking for the same type of pattern believe me. they are colocated. they are the first one in the Queue and when they think they will loose than they scratch the trade immediately. the dont pay any commissions. they are programmed by phds of any Kind, physicists, mathematicians, programmers, engineneers and so on. so in this age of algos and hfts the human scalpers edge shrinked, but didnt disappear completely. it is just harder.
I think the price is a fight between sellers and buyers, watching the Order flow will only show you the current status of the fight not the outcome. A bit like watching a football match, anything can happen till the last minute but by then it will be too late to place your bet. It get even worse when you take into account the noise generated by bots
from a scalper's perspective, DOM / price ladder is very useful for watching for real 'paper money' (long time horizon traders) orders. the most simple types used to show themselves as a huge order in the queue. slightly smarter ones used icebergs, time sliced orders etc. if you can spot them as a scalper, you can lean on those orders. it was SO EASY in 2005 that you could literally pickup coins doing intermarket arbitrage on similar products. that was before algos. eventually, some of these human scalpers got smarter & joined the hft side, essentially programming how to conduct intermarket arbitrage / spot 'paper money'. it's not to say they are invincible. if a discretionary trader with a huge pocket knows parameters such as a hft's likely stop loss tolerance, he can accumulate orders & push price all the way to a heavily invested hft's stops. I have personally seen this happen on a few occasions at a co-located trading arcade, during the instrument's cash trading hours. if trading from home now, my opinion is DOM is not as useful, due to the latency & really the influence of hfts now. BUT, the existence of hfts does not mean the end for human traders. as per what CALLumbus said, different groups have different goals. most of what hfts do is to take advantage of their privileges at exchanges by conducting sub-pennying / lean parties / arbitrage. the party of scalping for tight ticks is ruled by hfts. you can still scalp on an intraday timeframe / swing trade on a higher timeframe, where trading is like a game of poker/chess, waiting for enough traders to be caught on the wrong side. hfts do get caught on the wrong side too. on rare occasions, privileged ones with access to important news releases a few micro seconds beforehand can still get screwed if nobody hops on after them. I personally love to watch for these types of moves & screw their asses for a change
These moves you mentioned can be really fast. How do you fade them fast enough if you don't use an algo? Also how do you estimate where to fade them?
Why do you place your orders in the queue? They maybe trying get the best price for their customer. To try and test the conviction of the market for example if i place this 30000 lot sell order at the offer will buyers still lift the offer or will the bids dry up? There are lots of reasons
re: where to estimate how to fade them. if you are referring to very tight scalps, yes you are right, humans can no longer compete with algos. if you are referring to intraday scalps beyond mere ticks, the game is one of if-then strategy. in the past, breakouts may have worked bcos alot of pple were fading ranges. now, breakouts fail most of the time maybe bcos there are alot of pple trying to enter the trend on a breakout. in some markets, the behaviour of range vs breakout participants may be 50-50. it is always an evolving game of poker /chess. no one can tell you when a classical pattern will suceed or fail. it depends on who are trading it & how they are trading it. for eg. a deep pocket participant may engineer a mini failed breakout to shake out early trend participants, reverse it deep enough to trap the range trade crowd, then finally breaking it out. eg. gold on a H4 tf just b4 brexit decision. this is why humans still have an edge over algos on longer timeframes, & why algo strategies are mostly on what I mentioned in my previous post. bcos I am a real emotional basket case, my personal barometer on where to fade, are when I get very emotional & want to jump onboard a move. that's when I know I should standby to fade the stop of the loser in me. generally, traders move higher up the timeframe as we mature & understand ourselves better, developing patience. this is something algos will only catch on IF the thinker behind the program is of such a mindset. we humans can still suceed trading other humans. even longer timeframe low frequency algos are still essentially executing a human's decisions
Rebates, advertising liquidity to attract a buyer/seller, etc. Without saying too much, large traders tend to leave their footprints on the tape. Algos are experts at trading in nanoseconds but they don't move markets. Orderflow is the only tool that can tell you who's doing what in real-time. Often the outcome is obvious before it happens.
The more i think about it the less sens it makes. if i were a big institutional trader i would use iceberg to hide my big orders from BOTs and other traders, they won't show in the market depth but they'll be visible when executed, if i wanted to get more stealthy i would spread those orders across multiple prices, nobody would see a damn thing. Now, as a retail trader you have no way to know where/when those icebergs will hit nor if there will bigger orders to flip them, it happens regularly. pending orders are notoriously fake and have very little predictive value. What you know for sure is that some big guy(s) sold/bought at certain levels, not knowing if they're short or long term traders who could afford to have the price move against them by a margin way larger then yours, there could also be "bigger" fish in the sea that could get in with zero warning and turn the table around. Bottom line is, i don't think orderflow has any predictive value.