If you don't want to take anybody's word for it, you can calculate a fairly reasonable estimate yourself. Let Numerator = Max Open Interest on any one day Let Denominator = Sum of Volume of all trading days Then (Numerator / Denominator) is a fairly decent approximation of the percent of contracts that are delivered. This estimate will be an "upper bound", i.e., it will be equal to, or greater than, the correct value. (It's an upper bound because it assumes that the number of contracts delivered against, is equal to the max open interest throughout the lifetime of the contract). I just did this little calculation myself for the March 2008 contract of Crude Oil and got this result: approx 3.88% OR LESS of the traded contracts of March 2008, were delivered against. See attached Excel spreadsheet.
Thanks for your reply, but please explain the logic behind this calculation. If there is a lot of daytraders, volume will be quite high. If there is a lot of position traders, volume will be low. This has nothing to do, IMO, with how many contracts get closed.