At the end of the day, options can be as simple or complicated as you want to make them. But every complex spread is comprised of little single options and those little single options are all probability bets. You can combine them many different ways but they are still a total of all the other probability bets you are making. Over time, what determines your success in trading "volatility" is probability and having an edge. Legging out early on a trade is the same as opening a new a trade. Obviously buying back .05 or .10 options always makes sense, but if you are doing spreads for .30 to .50 credits then those nickel and dimes represent substantial costs to the position. Too many times option traders think they can overcome negative edge through adjusting a position a thousand times or they can simply "manage" the risk well. At the end of the day you are simply just making even more trades with negative edge. Over a long enough time horizon and enough trades, the negative edge will become realized through the erosion of your account. Many guys are afraid to trade the underlying and they often end up in the options arena because they are told or they believe that they can be successful without trading direction. Simply not possible. Either you are trading the direction of the underlying or the direction of volatility. Either way, you have to make a bet.
The best solution for a guy that tells he has no feel for direction is to take a very small position. Once you have money on the line, you get a feel for the market really fast. You can replicate the long term returns of an iron condor through the underlying by simply trading a mean reversion strategy with small size. Think about the p&l distribution of both and you will see they mimic each other with about 95% correlation. You trade very very small and you simply add shares on the way down and sell them on the way up. Now this might sound like gamma scalping except you are accumulating deltas as the market goes against you just as in an iron condor. In the end if your volatility assertions are correct, you should make money on both. All you have to do is replicate the deltas of the iron condor. Execution costs are not that great as you will not really be making that many trades and your p&l will be greater most of the time. You might even get lucky and catch a flyer. If you really want to trade volatility, you should be trading 6 to 9 months out. Even selling naked ATM straddles produces very little delta risk as there is no gamma out there with much greater reward. Much easier to remove the delta risk and trade pure volatility. Selling 5 delta spreads is not really trading volatility regardless, you are simply selling the tails that smart money is usually buying. Tails really don't have any useful greeks associated with them, they lie dormant until awakened. And when they awake, it's usually to remove equity from your account, lots of it.
At the end of the day, options are incredibly useful at manipulating your payoff structure. But there has to "be" a payoff. Earning small credits is not a payoff unless you are able to do that thousands of times a day via market making. The idea is to "manipulate" your payoff structure to actually "increase it" not decrease it. It's analogous to a futures trader who trades direction and moves his stop to break even once he is 10 ticks in the money and adds another contract. As the underlying moves higher he is able to geometrically expand his upside while keeping his risk fixed. There is a quote from the book "Ugly Americans". There is a section in there on the eight rules of Carney. Here is number 4: You walk into a room with a grenade, and your best-case scenario is walking back out still holding that grenade. Your worst-case scenario is that the grenade explodes, blowing you into little bloody pieces. The moral of the story: don’t make bets with no upside. Options in general are severely under priced. All of them! But there is more to this statement. Over short periods of time and data sets, option values will appear to be over priced. It's a function of the small data set. This is because of the fat tails. Over long periods of time, option prices resemble nothing even close to being fair. The reason for this is simple. The rare and isolated events that blow options up 100 to 1000 fold simply cannot be factored into the pricing equation. Otherwise there would be no buyers and no liquidity in the market.
The concept of fair value is similar to the concept of God. You are not going to know what it is until it's too late. But all options have realized volatility at the end of any period of time and that we know after that period concludes. Therefore we can look back and see if the implied volatility at that time matched what the actual volatility was. Again, you will find under most circumstances that the realized volatility matches the implied volatility pretty closely. Except on those rare occurrences where it's actually under priced by magnitude of orders. The error is always to the upside, not the downside. In other words, sure in some circumstances the implied volatility was 34 when it should have been 32. But on the upside you see examples where the implied volatility was 25 and it should have been 200. As you can see, if one is going to error, they are better off assuming volatility is priced too low rather than too high because the payoff for volatility being a little too high is minuscule vs volatility being too low.