2009: Markets Move More Overnight than Ever Before

I am going to post some data from TraderFeed blogspot, run by the very well respected Brett Steenbarger. Readers will know we've been noting some very strange action - after the initial surge in March - we began noticing a lot of the upticks were driven by "the persistent buyer" in premarket. Many days futures were negative overnight (I'd check at midnight) then when I woke in the morning I'd notice they'd be flattish or slightly green. Then they'd run up from 7 AM to 9 AM. And not just on days with economic news. This was happening notoriously throughout April and May... over and over. Who is so insistent on wanting to buy the market via futures... based on no news... and willing to place such sloppy orders that they move the market, rather than placing orders slowly so as to get better prices? Only 1 type of "person" - someone who does NOT want better (lower) prices for themselves - instead, that someone wants to pay higher prices (with sloppy future order and/or a mountain of them at once) to get the program traders all chasing in, and leading the dog by the tail.

I don't have access to hour by hour historical futures in the S&P 500, otherwise I would of taken the time to do this myself, since I had a running joke with someone that for a few months there you could buy futures at midnight and sell them at 9:35 AM each day and the winning percentage had to be 80-85%+. This has changed a bit the past 2-3 weeks and it's starting to wear off, but for a good 10-12 weeks of this rally this was the rule. I would love to quantify how much of our epic, history making run was achieved (1) during non market hours (9:30 AM to 4:00 PM) OR (2) a combination of non market hours + 3:30 PM to 4:00 PM. I'd guess almost half our gains were overnight ... which says a lot. (but I don't have the data - that is my thumb in the air guess)

This data from Brett is not so specific but it does look at how much of our movement is done during normal market hours versus after hours in 2007 v 2008 v 2009. It supports my anecdotal view that "the persistent buyer" has been doing a great job of marking up this market during off hours when thin trading allows "him" to move the market in much easier fashion (and with less of our tax money) than during the day. Ahem.

Those who do not believe in grassy knolls might just say that we react much more strongly to economic news in 2009 than we did in 2007 or 2008. If that's the case then why did we surge so many nights between 7 AM and 9 AM on no news at all?

****************** Brett's data below

How much movement can we expect in the U.S. stock market (S&P 500 Index; SPY) during the day session, and how much movement can we expect overnight?

It turns out that the ratio of day movement to overnight movement has been shifting over time. I haven't seen this reported elsewhere, but it's been striking to those of us who follow markets both during U.S. day sessions and overnight.

During 2007, the median absolute change from open to close was .61%; the median absolute change from the prior day's close to the current day's open was .37%. Day movement exceeded overnight movement by about 63%.

2008 was a more volatile year, so we saw more absolute movement during both time segments. The median absolute change from open to close was 1.06%; the median absolute change from the prior day's close to the current day's open was .58%. Day movement exceeded overnight movement by about 83%.

2009, however, has given us a different look. The median absolute change from open to close has been .93%; the median absolute change from the previous day's close to the current day's open has been .86%. Day movement has only exceeded overnight movement by about 9%.

Stated otherwise, we've seen less movement during the day session from 2008 to 2009, but more movement overnight.