My point exactly. Every transaction has a buyer and a seller. Using buy volume vs sell volume based on bid and ask transactions by itself to make some kind of judgement call on buying/selling interest is prone to error. The data is basically saying if the trade went off at the bid it was a sell. If it went off at the ask it was a buy. That's not really accurate because you're making an assumption of comparing motivated sellers to patient buyers or vice versa and there isn't enough data to make that determination one way or the other. In some cases you may be buying or selling to/from an MM's inventory as opposed to another shareholder. That again doesn't tell the whole story of what might be going on.
The only way to really draw any significance from that data is to combine it with price movement.
If there was 7:1 sells to buys and the share price tumbled, then that data is consistent with motivated selling and you could say that there is selling pressure. Conversely, if there is 7:1 buys to sells and the price rises dramatically, that indicates motivated buying.
And to continue with that, if there is a 7:1 ratio of sells to buys, but the share price doesn't really move, well that divergence tends to indicate that there is strong bid support to absorb the selling...like today.
This also refutes the notion of end of day tape-painting. If there is 7:1 sells to buys on over 2 million shares traded and the share price finishes up, that's not tape-painting - that's patient buying until all selling is exhausted.
News 
Market Data 
Discover 
AI
