If you purchased a lottery ticket once a day for the next 50 years....maybe not. If you purchased a lottery ticket once a day for the next 50,000 years.....it would be a virtual certainty. And that would be a blink of an eye in the time frames we are dealing with.
Let me repeat - a new gene will do absolutely nothing unless it is expressed.
Often they are expressed. If the upstream promoter elements are duplicated as well then what is the problem? Remember the example of the bacteria in xylose I gave you before? Those duplicated genes were expressed just fine and confered a definite survival advantage. In time enough changes will take place on the duplicated genes so that those enzymes will have a higher affinity for the xylose metabolites than for glucose.
Here is the key - those changes which will optimize the gene for xylose metabolism do NOT have to happen all at once. You (and many others) have a hard time with this concept. If just ONE point mutant now allows the gene product to metabolize xylose more efficently, it will quickly predominate in the population. Now ANOTHER mutation makes the site have an even higher affinity. Keep in mind, during this time, there are MANY MORE mutations which have a NEGATIVE effect on the xylose binding site will cause the bacteria carrying them to die - we will never see the "mistakes" . Over many generations eventually you end up with a gene which was apparently "designed" to metabolize xylose.
In other words, what I am saying is that even a single gene is an irreducibly complex entity which needs the help of many other functions, systems, in the organism.
Why are you having such a hard time conceptualizing that even a change as simple as a duplication could not be tolerated by the system? You end up with a slightly different animal perhaps. You just cant predict what will happen when you start deleting or adding genes to the genome! I remember several instances where a gene which was shown in tissue culture to be important in many diverse processes in a variety of cell types. Many predicted its loss would have a drastic effect on development and function of the adult mouse. Yet Lo and behold, they knockout the gene and the mice are perfectly viable. The genome has a lot of built in redundancy. Conversely, extra genes can and do sometimes find a role in the system. The genome isnt a delicate house of cards or even a "blueprint" in a strict sense.
The above is absolutely wrong. It is the reason why so many gamblers lose their shirts. If the odds against winning the lottery are say 1 in 100,000,000 and you played it 99,999,999 times wihtout winning, the chances of winning on your next bet are still 1 in 100,000,000. Random events do not have a memory.