Again you fail to comprehend. I have stated and the chart shows the BSM delta vs. the True delta on a daily basis. And yes the price of the option WAS effected by change in something other than the price of the underlying. Of course I would agree that something else could affect the price of the option during the sample period. That's the whole point. It makes perfect sense, the comparison with the BSM model Delta and the true delta showing wide discrepancies was the purpose of the analysis. And since you state that "something other" & "something else" affects the price of the option in relationship to the underlying then you would have to agree that utilizing the Greeks based on a "Theoretical Model" for trading decisions is hazardous at best. That has been my assertion from the start.
It seems your definition of delta is different than that of BSM. Your delta is not the delta as defined by the price change of an option due to the change in the underlying, but the day over day change in price of the option coinciding with an 'x %' move in the underlying. BSM provides the greeks and enables one to trade and discuss options in a standardized and comparable format with other market participants - seems pretty useful to me.
Your "true delta" is really an amalgamation of all the B-S greeks, so comparing B-S delta to your "true delta" is like comparing apples to oranges.
Thank you, MTE. Ironically, "apples and oranges" is where I started too, but I then foolishly tried to provide a more detailed explanation. Sure, but that's not a sensible assertion, for reasons stated by myself and others repeatedly. Now if you managed to plot the difference between the theoretical price of the option as explained by all the greeks, rather than just delta, and the actual price of the option, we could have a potentially somewhat more meaningful conversation.
I will no longer make any post to this thread until at least one proponent of BSM puts forth their own analysis with real world data to back up their position. I have requested same. They are either unable or unwilling.
Who will be the brave contender(s) to take up the challenge and enter the steel cage greek death match chamber? Let us hope all the greek fighters (both sides) enter so we can lock the cage, throw away the key and put an end to the et bloody, endless, meaningless greek wars. (actually we all like to watch a good fight.)
Clearly both sides can have winnrs : http://www.elitetrader.com/vb/showthread.php?s=&threadid=168873&highlight=greek* Practically it depends on individual's trading timeframe, style (directional, swing, etc.), strategy, etc.
It is not a matter of being a proponent of BSM but understanding its scope, applications and limitations. Which came first - The option price or the model? The models were designed to measure, label and help to explain the complexities of option pricing. In its theoretical snapshot world BSM is always correct and all real world data will support this. There are, of course, issues with the model - starting with the subjective nature of some of the inputs. So real world data assumes that we all use the same values. But this is not a flaw of the model. I was tempted to use MTE's line in the GLD thread. Here, I'll paraphrase - don't confuse knowing the greeks for thinking.
This is not about the proponents or the opponents of BSM. The BSM delta is an instantaneous measure of the option's sensitivity to changes in the price of the underlying. What you are measuring and incorrectly calling "true delta" is in fact delta-gamma-vega-theta-rho. If you want to make real life comparisons then you have two alternatives: 1. Take into account changes in implied volatility, time to expiration, size of the move (to account for changes in delta), and changes in interest rates, and then compare it to your "true delta". or 2. Take out the effects of above changes from your "true delta" and then compare it to the BSM delta.