Imagine that you've just checked the option chain on a security. You're studying volume and open interest. You're ensuring that the security's options are liquid enough for your needs. You plan on trading a butterfly centered at that particular option's strike. You notice that 300 have already traded that morning and open interest is 27,775. Open interest refers to the number of options or futures contracts that remain open or undelivered that day.
You decide that the option is liquid enough for your needs. You enter a butterfly that includes selling 10 of those options you had been checking and buying 5 at each of the wings. The volume on the option that forms the body of the butterfly ticks up by 10 contracts, but you notice that the open interest doesn't change.
Imagine that another trader out there is watching volume in that particular option, trying to decide what other traders are doing. That trader who watched the volume tick up by 10 contracts wouldn't have been able to tell by comparing the volume and open interest alone whether you had bought-to-open or sold-to-close those 10 contracts. While volume is updated constantly, open interest is calculated only once a day, at the end of the day.
If that other trader had been watching the "last price" column, and volume had been changing slowly enough that she could catch the price at which the 10-contract transaction had occurred, she might have been able to guess whether those ten contracts had been sold or bought by comparing the price at which they traded to the bid and ask prices. A last price close to the bid probably meant that that those 10 contracts were being sold while a price close to the ask probably meant that they were being bought.
But that's a clunky method that isn't always reliable. Astute traders who are trading complex trades such as butterflies in highly liquid vehicles can sometimes place their trades at or close to the mid price, so that a trade a penny or two off the mid price might not be clearly a bullish or bearish transaction. Even if the trader were able to conclude that particular option had been sold, unless that trader had been closely watching other options and noticed the uptick in volume in them and rightly concluded they were part of a butterfly trade, that trader wouldn't have noticed that the trade wasn't bullish or bearish at all, but part of a neutral butterfly strategy.
If volume is ticking up rapidly, yet the price of the option is going down, we can be fairly certain that more options are being sold than bought, and vice versa, but we don't always see such obvious signs. Perhaps some institutions are using a temporary rise in price to gradually unload some calls since they know many retail traders will be buying as prices climb.
Some companies and institutions have developed proprietary methods of determining whether an intraday volume change can be attributed to buying or selling of a certain option. In the live portion of the Option Investor site, the Market Monitor, commentator Todd Shriber often repeats reports on such option activity. If you watch CNBC or catch the CBOE podcasts, you've probably seen one of the Najarian brothers commenting on whether options were being bought or sold.
However, while most brokerages provide information on unusual volume activity, retail traders find it more difficult to obtain information on whether such volume can be attributed to buying or selling. Companies and institutions guard their methodologies. As far as I know, it's more difficult for the retail trader to determine whether option activity can be attributed to buying or selling, and I don't personally know of a free site that provides such information on daily option activity. Perhaps if some of our readers do know of such sites, they'll let me know and I can convey that information to readers.
Only by watching how the open interest changes overnight can option traders ascertain whether most of the previous day's volume can be attributed to opening new positions or closing old ones, but even then, opening and closing positions can't always be attributed to pure bullish or bearish action. For example, retail traders can be bearish during a sharp decline, buying puts, while institutions can at the same time be managing risk by selling previously held long put positions, locking in partial profits. Maybe institutional traders sell more puts that day than retail traders buy, so that open interest declines, but is that clearly bullish action, a sign that markets are due to turn around immediately? Not necessarily. We've said on this site before that institutions can move early as they manage their risk and lock in profits.
However, that doesn't mean that we should ignore open interest numbers, even intraday. If volume is ticking above open interest, we know that some sort of unusual activity is occurring. We can be alerted to check out what's going on before we enter our planned trade.