Chops and theekret, I have heard the term you used many times but I looked it up on the Urban Dictionary anyway, quite a good site!
Anyway, back to substance. I am not a big fan of market depth, very open to manipulation, but I would like to hear more about gavank's approach.
2. The relationship between market depth and subsequent small cap. share price movement is a complex one, at times the price will move to size, at other times not. Knowing when the behaviours will be different is the challenge.
Theekret:
The two points you raise are valid, again I would say knowing when and when not to use the market depth is the challenge. Also, can you expand on your time and speed comment?
Hang about , I may be on my 5th Stella , but you answered this question .
But without sprouting algorithmic strategies till I puke , lets just say the hunter can become the hunted .
If one could backtest executions effectively , it would have to deal directly with liquidity ( good subject at present , ...... cheers ..... ) , then real time data , which is 9/10ths the domain of third parties or more , which in itself can be a mine field when the transparency and accuracy needed to support the theory can be conflicting to any data messages structured by the software one is using , be it their own or another parties .
If you can merge all the suggested , lets call it a theoretical montage , where you could actually read each layer of the composites , the missing link would then be true sentiment , which in itself presents a wider timeframe than that available in market depth . Sentiment can be seen as realtime movement but the changes can't or at least not yet , to date they can only be observed once committed to execution and that can change twice in a blink primarily because of off market trades . Now if you could see market depth in off market , match it with liquidity projections , maaaaaaate , then we would have definition , with true definition who needs a crystal ball ( rhetorical question ) . It would be a transitive construction within a realtime market , which could take into account systematic differences and be bl**dy marvellous if it could be done . Even algos can't predict change in sentiment only price reflects that , then the equation can then start again as long as a trend can be summised . That then starts another barney in the market depth theory , because trend identification again is still post market .
Anything relying on pre is in for a post shock , so that cuts out any avenue on that theory straight away , as there is no measurement for pre acceleration phenomena for shares , methinks it can be concluded that market depth can only show a portion of the mass , there is no dimension to the viewer that could possibly take into account velocity .
Then we have another arguement again in velocity .
Bollocks was easier .
PS .. I like your stop theory .