Said by who? Not by chemists anyway. In fact, we know that a mixture of NH3 and CH4 under electrical discharge or UV illumination is converted to amino acids with aproximtely 4% efficency. This means the the probability of any given NH3 and CH4 molecule ending up as an amino acid is roughly 1 in 25. (of course, this is due to the chemistry of the molecules rather than simple randomness, the formose reaction is well understood.)The probability of a given page of any text being generated randomly is roughly 1 in 10^1000, assuming 200 words per page, and average word size of 5 letters (not counting punctuation or whitespace).
Can you see the difference?
Now let's have a look at your output. Yep it's a series of random letters, this is a random letter generator. Nice Java programming, but not relevant to amino acids.
And here I was hoping for a reaction path output a la MOLPAC. It's possible to calculate reaction probabilities explicitly, using quantum mechanics, although in practise approximations are used. The formose reaction for producing amino acids is short enough that you could apply MOLPAC type analyses to them.
However, You have apparently confused the probability of generating amino acids from simple molecules and the probability of generating proteins from amino acids. Even then you are way off mark. Generating the 32 amino acid "self-replicating" Ghadiri ligase (Lee et al 1996) purely randomly has a probability of 1 in 4.2x10^41. The probability of generating a given sequence of the 101 amino acid peptide cytochrome C is 2.5 x 10^131 (however, there are about 10^65 sequcences that represent cytochrome C (Yockey, 1977)).
In a global ocean of 10^20 litres, with an amino acid concentration of 1x 10^-6 molar, there are something like 1x10^50 potential starting chains (Chyba and Sagan 1991). Assuming that it takes a week to generate a sequence (Ferris et al 1996), Then the Ghadiri ligase could be generated in one week, and any cytochrome C sequence could be generated in a bit over a million years (along with about half of all possible 101 peptitde sequences, a large proportion of which will be functional protiens of some sort).
Thus, at the very least, Kaufmann/Eisner type hyper-cycles could be established very rapidly on early Earth (Lee et al., 1997).
Courtesy of Ian Musgrave, Ph.D, Prince Henry's Institute of Medical Research
Lee DH, Granja JR, Martinez JA, Severin K, and Ghadri MR. (1996 Aug 8). A self-replicating peptide. Nature, 382, 525-8.
Lee DH, Severin K, Yokobayashi Y, and Ghadiri MR. (1997 Dec 11). Emergence of symbiosis in peptide self-replication through a hypercyclic network. Nature, 390, 591-4.
Yockey HP. On the information content of cytochrome c. J Theor Biol, 1977 Aug 7; 67(3): 345-376.