If you click on the 15-minute footprint chart from Market Delta, you'll see (inside each bar) the volume transacted at the market bid x the volume transacted at the offer for trades of 50 or more contracts. This is a slow day, so you wouldn't expect a large volume of large trades, but it does appear that it may still be possible to track large trades in spite of recent changes in the ways in which volume is being reported by CME. I'll be looking into this further later this week.
.
Posts
4 comments:
Of course there will still be some large trades that show up, but drawing conclusions from those may not work well because there could be just as much or more volume transacted on the opposite side with big orders that are split up and thus go unseen.
For example, we could see that there are a couple thousands contracts going off at the offer in lots of 50 or more while on the bid there is very little big lots. This could lead to the conclusion that there is solid buying. But there is a very real possibility that there could be more than that going off at the bid in parceled orders which aren't showing up in the filter. That would give a totally wrong picture.
I think the only hope to get a truly accurate picture from this kind of big-lot analysis is to have some kind of software that reverse engineers the split orders and re-assembles them in their original lot size.
Thanks, Brett, for actually investigating the issue rather than jumping to conclusions or spreading conspiracy theories. This topic has caused many uninformed, scared opinions to spread around the internet.
I've studied volume in the ES for years now, and it is definitely still possible to follow the biggest traders, especially when they are in a hurry.
People are wondering about all of these edge cases, when in reality the general bid vs. ask market trade profile looks very similar on ES whether you split the volume or just analyze every tick. At least, the way I measure it, this has always been the case. The reason, of course, is that the bigger players are such a huge majority of the total volume. So in the worst case you just stop splitting volume out, and you still get a decent picture. In practice methods will be found to get better fidelity readings than that.
The only real consequence is to people without any technology on their side, who can no longer see a bunch of 800-lot bundled trades on their T&S window.
Brett, just looking at this particular MD chart(and subsequent price action), it appears the "large traders" were on the wrong side of the trade.
I would tend to agree w/ Ziad in that there are methods large traders could employ to conceal there activity...and I would think that they would certainly attempt to do so in every possible way. That, or they would submit large orders as red herrings to fool other mkt participants, which is what this chart looks like.
Just speculating...
charleS
I am inclined to agree with Ziad and it would be great to have software to reverse engineer the situation but the mind boggles with that idea. Is that really feasible ?!
I know I can see when large volume comes in on unfiltered cumulative delta (total delta) but a large volume of twitch traders getting ever twitchier is not the same as fidelity volume having the potential to provide direction. It is when filtered and unfiltered diverge that I find interesting and capable of providing an edge. Its also of value to be able to see when larger players are not active.
David.
Post a Comment