I watched some of your video and saw that it was not addressing the science... it was engaging in philosohpy.
This excerpt addresses your snowflake argument... but the link also explains the science in a different part of the essay.
On philosophical grounds... your snowflake argument is weak...
"Elaboration of Likelihood Principle
According to the likelihood principle, an event or state of affairs E counts as evidence in favor of an hypothesis H1 over H2 if E is more probable under H1 than H2, with the degree of support proportional to the ratio of probabilities under the two respective hypotheses.[6] The likelihood principle shows why an ink splotch that looks like the face of Abraham Lincoln would support the idea that the splotch was designed, whereas a splotch of random looking ink marks would not. Although the exact details of both kinds of ink marks are highly improbable under the chance hypothesis, only in the former case are the ink marks not highly improbable under the design hypothesis.
One common objection to our likelihood account of why these features count as evidence, which is based on a misunderstanding, is that we are merely arguing from the purported improbability of the existence of one of the above features under naturalism (or the naturalistic single-universe hypothesis). Then the objection goes, very improbable events occur all the time. For example, the exact pattern of any ink splotch is very improbable--never to be repeated in the history of human beings--and yet most of them do not signal design. The premises of the likelihood version of the design argument, however, are not merely that the existence of certain features of the universe are improbable (or surprising) under naturalism or the naturalistic single-universe hypothesis, but that they are also not very improbable (or surprising) under theism. As we saw in above example of the Lincoln-like ink splotch, both of these conditions are necessary for one to claim that these features support theism over naturalism using the likelihood principle.[7]
It is critical to point out that the sort of probability used here is not statistical probability, since this would require that the universe be generated by some physical process that churns out life-permitting universes at some relative frequency, contrary to the assumption of typical forms of naturalism that claim the universe is simply a brute fact. Rather, the probability used here is what philosophers call epistemic probability, which can be thought of as a measure of rational degrees of expectation. For example, when scientists say that the theory of evolution is probably true, they are clearly not talking about statistical probability: they are not referring to some repeatable trial in which the theory turns out true with some relative frequency. Rather, they are saying something to the effect that given the total body of available evidence, a rational person should expect that the theory of evolution is true.
Put in terms of epistemic probability, the likelihood principle can be reworded in terms of degrees of expectation instead of probability, in which case it becomes what I call the expectation principle. According to the expectation principle, if an event or state of affairs E is more to be expected under one hypothesis H1 than another H2, it counts as evidence in favor of H1 over H2--that is, in favor of the hypothesis under which it has the highest expectation. The strength of the evidence is proportional to the relative degree to which it is more to be expected under H1 than H2. Rewording the likelihood principle in this way is particularly helpful for those trained in the sciences, who are not familiar with epistemic probability and therefore tend to confuse it with other kinds of probability, even when they are aware of the distinction. Given this rewording, the central premises of our argument become that the various features of the universe mentioned above are very surprising (unexpected) under naturalism (or the naturalistic single-universe hypothesis), but not under theism, and thus they provide evidence for theism over naturalism.
One might question this use of epistemic probability on the grounds that it is merely subjective. One response to this objection is that epistemic probability is used, and needed, for many widely accepted inferences in everyday life and science. For example, as the above quotation by Edward Dodson illustrates, the support for the thesis of common ancestry (evolution) is based the claim that a variety of features of the world--such as the structure of the tree of life--would not be improbable if evolution is true, but would be very improbable under the other viable nonevolutionary hypotheses, such as special creation. This improbability is not statistical improbability, nor can it be justified by an appeal to statistical improbability, since we have no statistics regarding the relative frequency of life on a planet having these features under either the evolutionary hypothesis or some nonevolutionary hypothesis. Neither do we have any model from which to derive those statistics. Thus, if it were a statistical probability, it would be completely unjustified. Rather, it should be understood as a form of epistemic probability--e.g., as claiming that various features of the world would be very unexpected under the various contender nonevolutionary hypotheses, but not under the evolutionary hypothesis. Further, since we have no statistical models on which to base our judgments of epistemic probability (especially for the nonevolutionary hypotheses), I contend that these judgments of epistemic probability are not rigorously justified. Rather, after (hopefully) doing the best job of looking at the evidence, scientists and laypersons make judgments of what kind of world we should expect under each hypothesis, and then they simply trust these judgments. This sort of trust in our judgments of epistemic probability--that is, what we should rationally expect under various hypotheses--is a pervasive and indispensable feature of our intellectual life.
This same kind of reasoning is what is going on in the likelihood rendition of our argument: we look at the various features of the universe mentioned above, and judge that they are very surprising under naturalism, but not under theism. Then, as in the case of evolution, after a careful analysis of the evidence, we trust our judgments of epistemic probability in deciding the strength of the evidence. What if someone does not share these judgments of epistemic probability? One can either appeal to how widely shared these judgments are by those who are relevantly informed, or one can attempt to provide a deeper justification of them. In this regard, it should be noted that the judgment that features of the universe such as beauty and discoverability are surprising under naturalism is widely shared by intelligent, informed individuals, as some of the scientists and philosophers cited above illustrate. I believe, however, that a more rigorous, deeper justification can be offered.[8] For example, in the case of the fine-tuning for life, I base the claim that an (intelligent) life-permitting universe is very surprising under the naturalistic single-universe hypothesis on the relative smallness of the range of the parameters of physics that allow for embodied, intelligent life, along with a revised version of the probabilistic principle of indifference. My point here, however, is that although such justification is nice to have, even if it were not offered, that should not undermine the claim that the above features provide evidence, via the likelihood principle, for theism over naturalism, just as it does not in the analogous scientific cases...."
http://www.infidels.org/library/modern/robin_collins/design.html
This excerpt addresses your snowflake argument... but the link also explains the science in a different part of the essay.
On philosophical grounds... your snowflake argument is weak...
"Elaboration of Likelihood Principle
According to the likelihood principle, an event or state of affairs E counts as evidence in favor of an hypothesis H1 over H2 if E is more probable under H1 than H2, with the degree of support proportional to the ratio of probabilities under the two respective hypotheses.[6] The likelihood principle shows why an ink splotch that looks like the face of Abraham Lincoln would support the idea that the splotch was designed, whereas a splotch of random looking ink marks would not. Although the exact details of both kinds of ink marks are highly improbable under the chance hypothesis, only in the former case are the ink marks not highly improbable under the design hypothesis.
One common objection to our likelihood account of why these features count as evidence, which is based on a misunderstanding, is that we are merely arguing from the purported improbability of the existence of one of the above features under naturalism (or the naturalistic single-universe hypothesis). Then the objection goes, very improbable events occur all the time. For example, the exact pattern of any ink splotch is very improbable--never to be repeated in the history of human beings--and yet most of them do not signal design. The premises of the likelihood version of the design argument, however, are not merely that the existence of certain features of the universe are improbable (or surprising) under naturalism or the naturalistic single-universe hypothesis, but that they are also not very improbable (or surprising) under theism. As we saw in above example of the Lincoln-like ink splotch, both of these conditions are necessary for one to claim that these features support theism over naturalism using the likelihood principle.[7]
It is critical to point out that the sort of probability used here is not statistical probability, since this would require that the universe be generated by some physical process that churns out life-permitting universes at some relative frequency, contrary to the assumption of typical forms of naturalism that claim the universe is simply a brute fact. Rather, the probability used here is what philosophers call epistemic probability, which can be thought of as a measure of rational degrees of expectation. For example, when scientists say that the theory of evolution is probably true, they are clearly not talking about statistical probability: they are not referring to some repeatable trial in which the theory turns out true with some relative frequency. Rather, they are saying something to the effect that given the total body of available evidence, a rational person should expect that the theory of evolution is true.
Put in terms of epistemic probability, the likelihood principle can be reworded in terms of degrees of expectation instead of probability, in which case it becomes what I call the expectation principle. According to the expectation principle, if an event or state of affairs E is more to be expected under one hypothesis H1 than another H2, it counts as evidence in favor of H1 over H2--that is, in favor of the hypothesis under which it has the highest expectation. The strength of the evidence is proportional to the relative degree to which it is more to be expected under H1 than H2. Rewording the likelihood principle in this way is particularly helpful for those trained in the sciences, who are not familiar with epistemic probability and therefore tend to confuse it with other kinds of probability, even when they are aware of the distinction. Given this rewording, the central premises of our argument become that the various features of the universe mentioned above are very surprising (unexpected) under naturalism (or the naturalistic single-universe hypothesis), but not under theism, and thus they provide evidence for theism over naturalism.
One might question this use of epistemic probability on the grounds that it is merely subjective. One response to this objection is that epistemic probability is used, and needed, for many widely accepted inferences in everyday life and science. For example, as the above quotation by Edward Dodson illustrates, the support for the thesis of common ancestry (evolution) is based the claim that a variety of features of the world--such as the structure of the tree of life--would not be improbable if evolution is true, but would be very improbable under the other viable nonevolutionary hypotheses, such as special creation. This improbability is not statistical improbability, nor can it be justified by an appeal to statistical improbability, since we have no statistics regarding the relative frequency of life on a planet having these features under either the evolutionary hypothesis or some nonevolutionary hypothesis. Neither do we have any model from which to derive those statistics. Thus, if it were a statistical probability, it would be completely unjustified. Rather, it should be understood as a form of epistemic probability--e.g., as claiming that various features of the world would be very unexpected under the various contender nonevolutionary hypotheses, but not under the evolutionary hypothesis. Further, since we have no statistical models on which to base our judgments of epistemic probability (especially for the nonevolutionary hypotheses), I contend that these judgments of epistemic probability are not rigorously justified. Rather, after (hopefully) doing the best job of looking at the evidence, scientists and laypersons make judgments of what kind of world we should expect under each hypothesis, and then they simply trust these judgments. This sort of trust in our judgments of epistemic probability--that is, what we should rationally expect under various hypotheses--is a pervasive and indispensable feature of our intellectual life.
This same kind of reasoning is what is going on in the likelihood rendition of our argument: we look at the various features of the universe mentioned above, and judge that they are very surprising under naturalism, but not under theism. Then, as in the case of evolution, after a careful analysis of the evidence, we trust our judgments of epistemic probability in deciding the strength of the evidence. What if someone does not share these judgments of epistemic probability? One can either appeal to how widely shared these judgments are by those who are relevantly informed, or one can attempt to provide a deeper justification of them. In this regard, it should be noted that the judgment that features of the universe such as beauty and discoverability are surprising under naturalism is widely shared by intelligent, informed individuals, as some of the scientists and philosophers cited above illustrate. I believe, however, that a more rigorous, deeper justification can be offered.[8] For example, in the case of the fine-tuning for life, I base the claim that an (intelligent) life-permitting universe is very surprising under the naturalistic single-universe hypothesis on the relative smallness of the range of the parameters of physics that allow for embodied, intelligent life, along with a revised version of the probabilistic principle of indifference. My point here, however, is that although such justification is nice to have, even if it were not offered, that should not undermine the claim that the above features provide evidence, via the likelihood principle, for theism over naturalism, just as it does not in the analogous scientific cases...."
http://www.infidels.org/library/modern/robin_collins/design.html