Did you read my post? I gave a very specific example of the 5k msg/s being evidence of the author's general lack of knowledge. The "article" is the equivalent of: 1) You and I go get a burger. 2) You don't like yours. 3) I take your leftovers back to my chemistry lab (after all, I'm a chemist, so the answer must be chemistry, right?). 4) I proceed to subject your burger to as many tests as I can think of until I find something interesting, defined as "something that looks cool on a graph". 5) I then tell you that you don't like the brand of table salt the restaurant used when combined with the fact that you had Cheerios for breakfast without waiting 3.5 hours for lunch.
Misusing quote is a big deal because quotes are event driven meaning the book needs to change for a new quote to be generated - that said, no one just fires off quotes unless the book changes. In order for the book to change an order must be created, canceled or changed - so valid orders are the only reason why a quote will be sent. The article fails to address the fact that the NYSE halted trading in specific equities which caused the SIAC NBBO (which is pegged to a price that favors the Lead Market Maker) to stall with the halt and caused an arb opportunity between ECNs. Initially when the NYSE halted there was a slight backlog of orders before they started routing outside to other ECNs (as they are legally obligated to do). It was the halt and split second while the NYSE halted and routed away that caused an arb opportunity. The initial order flow that was forced to be routed away and then the arb trigger to buy back up. The article does not discuss any of this at all.
Yes the article discusses in specific detail when stocks were halted and when they entered "slow quote" mode and includes detailed charts of the supporting data: "There are many who believe that the NYSE "Slow Quote" mode or LRP's (Liquidity Replenishment Point) caused the market to drop. However, only 1 stock was in slow quote mode when the final decline began, and only 3 were by the time it was half way done. From the chart below it is clear that the NYSE was not entering slow mode until WELL into the collapse: ...Furthermore, the stocks entering "slow mode" were trailing the market, not leading the market. The stocks that led the market are the same stocks we analyzed in Part 3, The Evidence. PG was such a stock that trailed the market. From the following chart you can see that PG had NO trades on the NYSE when it entered "slow mode" from approx. 14.45.30 to 14.47.15 and that at this time, the market had already collapsed. ...It is also clear that Apple trailed the market and was not the cause of the crash: ...Stub quotes did not cause the crash. You can see from the charts below that while stub quote counts did rise, they did not become prevalent or effect BBO prices until the crash was almost over. One could infer that the presence of stub quotes is a good indication of a lack of liquidity." http://www.nanex.net/20100506/FlashCrashAnalysis_Part5-1.html