3 ms·
I would not draw same conclusion. The website visit figures are estimates by SimilarWeb. I never heard of them, but would take their estimates with a grain of
by LittlePeter 7y ago
I would not draw same conclusion.
The website visit figures are estimates by SimilarWeb. I never heard of them, but would take their estimates with a grain of salt until proven otherwise.
Next, flagging volume fake just because average volume per estimated visit is high? That's a leap I would not make. Some sites cater to market makers who have larger volumes and trade over API.
What about dark pool trades where one exchange does publish the figures while the other does not?
Also, what is fake? Reporting volume without the actual volume of reported trades adding up to the reported value? Or are some trades fake? If the trades are fake, are they wash trades? Is the exchange the counter-party in these trades?
I am unable to trust this analysis.
- gruez 7y agoI agree the evidence and reasoning is pretty shaky, but the conclusion isn't exactly surprising. See report from a few months ago with more rigor: https://news.ycombinator.com/item?id=19465606 https://news.ycombinator.com/item?id=19465606