Tick data store design: storing all quotes vs NBBO

rohan2008 · May 4, 2015

Hi Everyone,

I have been working on candle stick based strategies and I am now looking at extending them to ticks/quotes. I can envision strategies that depend on NBBO, T&S and aggregated quotes on both sides; a lot of valuable market information can be derived from these three parameters... I was wondering if level2/depth of market quotes can provide similar value? Majority of the (non-NBBO) quotes are fake as I understand (please correct me) and so is there any value in storing all the changes in non-nbbo asks/bids.... this comprises of so much data... why store all the junk when you know its mostly fake.

My question:
1. Are the non-NBBO quotes any useful at all?
2. Does it makes sense to store non-NBBO quotes for future use? Does anyone store all this information?

thanks,

rohan2008 · May 5, 2015

I was referring to storing the non-nbbo data for future backtesting (if it is useful in the first place) in the above question...

i960 · May 5, 2015

Comparing the tape to the book might allow you to determine "how fake" the book is in given scenarios. IMO the majority of the book is not fake the majority of the time. In certain situations spoofing shows up.

volpunter · May 5, 2015

Its the opposite, nbbo is mostly marketing crap. Its a non-tradable feed. All your order size is filled at the actual exchanges. If you trade higher frequencies then NBBO is utterly useless. If not then nbbo should be fine to stream to your strategies. I recommend you first make up your mind what data you really need before thinking of how to store such data.

rohan2008 said:
Hi Everyone,

I have been working on candle stick based strategies and I am now looking at extending them to ticks/quotes. I can envision strategies that depend on NBBO, T&S and aggregated quotes on both sides; a lot of valuable market information can be derived from these three parameters... I was wondering if level2/depth of market quotes can provide similar value? Majority of the (non-NBBO) quotes are fake as I understand (please correct me) and so is there any value in storing all the changes in non-nbbo asks/bids.... this comprises of so much data... why store all the junk when you know its mostly fake.

My question:
1. Are the non-NBBO quotes any useful at all?
2. Does it makes sense to store non-NBBO quotes for future use? Does anyone store all this information?

thanks,

rohan2008 · May 5, 2015

volpunter said:
Its the opposite, nbbo is mostly marketing crap. Its a non-tradable feed.

i960 said:
Comparing the tape to the book might allow you to determine "how fake" the book is in given scenarios. IMO the majority of the book is not fake the majority of the time. In certain situations spoofing shows up.

Looks like I am confused with the term "NBBO"

Today, I need three pieces of data: trades, the lowest ask/highest bid at that time and aggregated sum of the 5 consecutive asks & bids... I get all this in the datafeed from my broker. I can visualize the strategies that can filter out the spoofing and use the information appropriately.

What I am not sure at this point of time is if it is possible to use the remaining asks/bids to derive any useful information even in future. The last thing that I want is to ignore the remaining asks/bids now, capture what I need for a few months and then realize that I need every ask & bid...

Yes, we can detect spoofing by comparing the tap vs order book, but that comparison is only for the tap/trade vs the best ask/bid at that time when a trade happens. Do algo traders also use the depth of the book to develop additional indicators/strategies. For example, if the price of ES is at 2101.00 (best ask @2101.25 and lets say it hasn't touched the ask for a long time) and lets say we see quite a few asks getting added at 2101.75 all of a sudden... ? Is it even possible to differentiate between fake orders and real orders at 2101.75 by looking at the asks/bids when the price hasn't even touched 2101.25? Can we hypothesis anything at 2101.75?

rohan2008 · May 5, 2015

no of best asks/bids + trades is approx 60 entries/sec for ES (@11:12EST)... now, if I include all asks/bids to this, I am getting around 300 entries/sec...