85% of options that are held to expiry may expire worthless, but that's not the same as 85% of all options that exist during the life of the contract.
Here's a simple example. Suppose, during the life of an option 100,000 contracts are open, and subsequently 60,000 are offset (i.e. closed out prior to expiry). That leaves 40,000. Out of those 40,000, 35,000 contracts may expire worthless, which means that 87.5% (35,000/40,000) of options that were held to expiry have expired worthless. However, those same 35,000 are only 35% (35,000/100,000) of the total number of contracts that have existed throughout the life of the option.
In other words, it is completely misleading to state that 85% or whatever of options expire worthless so you should write options rather than buy them. Taken at face value writing is neither better nor worse than buying. Options are priced at zero expectancy, which translates into negative expectancy for retail traders because of bid-ask spread and commissions.