I'm tracking dark pool trades and noticed that today after hours an AAPL trade was reported at a price below any price available today, while the price actually matched another dark pool trade executed yesterday after hours. I often see such trades reported late, usually assuming they happened earlier in the day, but more than 24 hours looks excessive and possibly obstructive. The tick timestamp also doesn't show correct date/time (some trades can come in late but carry an earlier time-stamp). This also confuses various charts, including Google. Is this even legal? (BTW, there are thousands of such trades every day across various instruments, with smaller and hard-to-detect discrepancies, confusing every back testing system and possibly some individual traders)
A few of these types of threads floating around recently. Ignore the trades. They were trades executed probably in a block at the previous days close and are indicative of nothing. They are supposed to be reported within a half hour of the next days open but I often see them get peppered in on the tape throughout the day instead. I have screenshots of this type of stuff over the last few years but was never able to find a use for it. I don't see a dark pool designation on the 'tape' that you pictured, so I'll assume that you know what you're talking about there but yes, most electronic reporting is pretty seamless but it's not perfect.
I do ignore them, though not sure that I can detect all of the skewed trades with invalid timestamps. I've spent couple years cleaning and fixing various 'data bugs', but there are many smaller and not so obvious ones that can trip variety of trading systems, not just my own. Some people also manually trade based on price charts, without Level 2. So I just thought there may be some legal boundaries in terms of reporting any trades late, especially with wrong timestamps. At least I'm hoping SEC will be dealing with this type of stuff because otherwise everyone may be be trading on random price data...
If your only intention is to filter these trades out, that should be pretty simple. Ive had to do that in the past and it was an easy fix... but i cant remember how. I'll repost here if i can think of it... what software are you using? The likes of you and i are not going to compel the sec to do anything and anyway, who cares.
I find it interesting that some choose to ignore them as noise and some choose to focus on them as information to benefit their trading.
I use some block and dark pool trades when analyzing data, but it’s hard to make a case that old/stale/corrupt/manipulated data is nearly as good as fresh/correct data. Otherwise everyone could manipulate data and you’d imagine you’re given something useful rather than being manipulated.
FWIW, I'm thinking of the likes of Stefanie Kammerman. Also I believe Linda Raschke makes use of block print analysis. Idea being, even a day late, the block prints give an idea of how so called "smart money" is positioning themselves.
Understand, and I see similar value. But holding and reporting trades very late is a form of manipulation. And usefulness obviously has to decay with time. What if someone holds trade info for an hour, someone else for a day, or for a week, or for a month. They all cannot be equally useful. And if at some point you’ll see a mix of random trades held up for random lengths of time, your own trading would become random. Btw, if Stefanie was any good then she’d be running a hedge fund, not scamming people out of thousands of dollars. https://tradingschools.club/reviews/stock-whisperer-review-stock-trading-room/