Wouldn't surprise me. Earnings and a lot of other data products are released outside of normal trading hours.
Sorry to go in somewhat of a tangent, but a question has been asked a few times here and never really answered about how much of the theta of an option decays throughout the 24 hours of the day. This would seem to indicate it should decay more during the close to open period then it does during the trading day. Or at least a similar amount. Is that what we see?
Actually the study doesn't say that at all. Everyone here is getting frustrated with you because you're not grasping what the data actually says. How about we do this, I will find a single week in the past 25 years (in your referenced data set) where the market gains during that week all happened during market hours and none, nada, zip occurred during after hours. Since that data would prove your statement incorrect, you'll donate $100 to a charity of my choice. Or just take a class in statistics and stop making nonsensical statements, I'd settle for that. How about it?
Thanks, I do this for a living. It's pretty simple, for the millionth time, net gains "after hours only" were nearly 600% and "during hours only" were -4%, at the end of the 25 year period. I don't care about 1 week, this thread is regardling the phenomena that occurred over the 25 year test period. I don't have an opinion about the data. It just is what it is. The hopes are more data sets and studies would be posted here about this phenomena, and there have been a few good ones so far. I hope more come in. "Everyone" is not getting frustrated. Some people are providing data sets of their own and adding value, some are using this thread as a soap box, some don't know what they are talking about but act like they do. It's that first group that makes these threads worth while even with all of the extra rabble inbetween.
https://bruceknuteson.github.io/spy-day-and-night/ To add to some of the other research already posted here. This is essentially showing the same phenomenon of all net % gains occurring overnight, but in multiple markets including Canada, Germany France and Japan. The most jaw dropping is the NASDAQ results which show that net % gain after hours has been over 3,000% and during market hours has been -65%. Essentially showing the bulk of daily losses happen during market hours not overnight, which has allowed the overnight-only gains to grow exponentially faster. Overall it seems compelling and worthy of more research.
Just to be clear, you stated " all net market gains have occurred after hours, on none, zero, zip have occured during market hours, over the past 25 years." What was the point of the "none, zero, zip" part of that sentence? It is good to see that you're now saying that "during a specific data mined period, if you summed all the returns that happened overnight vs all the returns that happened during the market session....you would find the results cited". If you really do this for a living, perhaps you'd consider being more precise in your wording since the wording I quoted actually says something far different than the study and to be frank makes it sound like you're one of the some who "don't know what they are talking about but act like they do." No one is claiming "an opinion on the data", the opinion is on how the data was chosen. We actually had a fun assignment back in stats where we competed to find the most absurd data set that explained the last 10 years of stock returns and as I recall something related to if the AFC or NFC won the Super Bowl had the most robust predictive power. The data was dead accurate, that didn't make it at all useful for future predictions. It makes sense that this data is showing simply that larger moves occur between the close and the open because that's when most data is released, therefore if the market is trending up it will tend to amplify that if you buy at the close and sell on the open and vice versa. I could equally well tell you to buy and hold a basket of stocks with a beta greater than 1 and you'd get similar return results and profile, no? As evidenced by the Sharpe ratios you see.
To emphasize it, since that is the phenomenon. I stated from the beginning that this was a 25 year study and the study was comparing gains during "market-hours only" and "after-hours only". I couldn't break it down any more simple without breaking out flash cards. Now there have been some white papers that contain studies that lead to various hypothesis on why this may be happening. You on the other hand "simply" know. What must it be like to have that level of market consciousness to look at a data set and immediately know all cause and effect. I bet you are posting from your mansion in Barbados aren't you. You are clearly trying to save face, but I am not interested in word gymnastics. Post data or a study that relates to the topic or move along.