Running A Negative Control on ID Math

I love these conversations and but I recoil when I read this (above.) This discussion has come up many times over in terms of lottery selections and lottery winners. I don’t believe that what you are saying here, T, is exactly what you mean or what you should mean.

It is not that according to ID theory that a random sequence generator could not have produced that rare of a sequence. Every sequence that a random sequence generator produces is that rare.

The challenge is to create a sequence that is significant in its own way. To generate a lottery ticket is a unique event, but not a rare event. To generate a winning lottery ticket is a unique and rare event.

I could misunderstand, but that you can create this is one thing:

That you can create it again at the push of a button is something altogether different. (If I’m wrong, then let me know.)

You are perfectly right. And to see why it is something altogether different, you can have a look at the reference I give at 14.

IOW, ID commits the lottery fallacy.

1 Like

The problem is that significance is determined after the protein sequence emerges. This is called drawing the bulls eye around the bullet hole.

It isn’t that rare at all. People win the lottery all of the time. Before a drawing, the ticket that ultimately wins is no different than any other ticket, and it is only given significance after the drawing.

To push this analogy a bit further, we could look at the probability of 5 specific people winning the lottery. Let’s say that the odds of winning the lottery are 1 in 200 million, there are 200 million tickets sold for each drawing, and there is 1 winner for each drawing. The winners are John, Susy, Tom, Leslie, and Dawn. The odds of those 5 people winning was 1 in 200 million to the 5th power, which is a pretty big number.

This is where the ID smoke and mirrors comes in. ID supporters would claim that the odds of those people winning is too astronomical, and therefore there had to be a designer who helped things along. They would also claim that these are the only possible winners since these are the winners we see. @stcordova would call it a violation of large numbers. However, with the conditions we set out, 5 winners are inevitable. Whatever 5 winners we end up with will be exeptionally improable, but still inevitable.


This is what happens whenever this conversation occurs… and maybe it needs to move to another thread.

I understand all that… When we talk about apologetics, and you see an argument that is not good, you point it out and say, “Well, at least we have a chance to discuss which kinds of arguments are valid and which are not.” You aren’t taking a position regarding the Christian apologetic, just having a discussion.

When you take a random number generator and generate a sequence and say that it is rare, no problem. But that rare sequence, in all probability, has no significance. It is just garbage. So, while the ID position that winning the lottery may not be as rare as one may think, the explanatory power of your example is equally terrible. It is just another bad example or metaphor for what happens. So, don’t perpetuate the problem by using it. Explain it away differently. This is a bad way of defending evolution.

This is my point. If you believe that this is so, you can’t use random number or letter generator examples to defend evolution. If it is bad there, it is bad here.

@Michael_Callen I respectfully disagree. Perhaps it needs to be better explained but the analogy is spot on.


No, because those who accept evolution are, in effect, saying that someone can win the lottery without it being fixed. Whereas the creationist is saying that it is impossible for anyone to win the lottery unless it has been designed for a particular person to win it.

This is a better analogy to the evolutionary process: You have millions of random number generators running at the same time. Every so often, and it is probably not so rare, two of them will match to, say, 25%. These two now no longer generate numbers entirely by random. Rather each of the numbers that match will only change random, say once every 1000 draws. The numbers that do not match continue to be generated entirely at random, but when two of them match between the two generators they now follow the rule of the ones that already match.

There is now a race, of sorts, in which some of the numbers that match will be no longer match after enough rounds, while others that did not match will now match. And depending on how things turn out, the numbers might fall completely out of sync, or become increasingly more alike until they are almost identical.

That turned out being much more complicated than I set out for it to be. I hope it is understandable.


The analogy is not spot on if it doesn’t make sense… and if it is also analogous to the ID analogy that fails. I’m happy to learn, so please explain.

1 Like

Yes, thank you, this is a much better analogy that better explains the process. I’m not sure if it is correct, or not, but it is understandable. It is also, and I’m sure you’d agree Faizal, was much different than what was being explained above.

I’m not trying to explain away ID, nor am I trying to punch holes in evolution. I just want everyone to use analogies and examples that make sense.

1 Like

The point is that @T_aquaticus is applying the same math used in ID arguments to scenarios that are obviously not designed, at least not how ID means it. They show astronomically I likely things happening, according to ID math, which demonstrates that this approach to detecting design fails by detecting design were it doesn’t exist.

Scientists call this type of reasoning “running a negative control”. The ID logic and math produces a positive where it should produce a negative. This example can be generalized to many many more examples. For this reason, most scientists conclude that we can’t trust positives from the ID reasoning, as it’s easily confused.


Thank you, that is a very good explanation. So, here’s my problem, and maybe you can articulate it to the greater group, because I know that this is an issue for other laypersons as well. T says:

Is this a valid result? Is this a functional (viable?) protein and, if so, was it returned on the first try? My assumption was that this is not. That this is garbage… But maybe i’m wrong. Maybe this is actually something and it was returned on the first try using the random character generator.

1 Like

Part of the point is that “function” is a hopelessly ambiguous term. This sequence certainly is “functional” because it serves a purpose in @T_aquaticus’s argument. So yes, it is functional. It is not garbage. It is a valid amino acid sequence that does not, for example, include any Z or X letters or punctuation. It certainly is not garbage.


Awesome, thanks very much Joshua. So here’s one main misunderstanding that I had, and also a clue as to why this is so challenging for many of us on the outside.

It seems like T was saying, see, I can randomly generate something and it is easy, when Gil was saying random generation of anything significant is too difficult to entertain (statistically speaking.)

Is this what was being said?

1 Like

6 posts were split to a new topic: The Foundation of the ID Argument

Is it okay if we put aside the problem with ID or evolution for now? I would really like to focus on what the analogies mean and whether or not they are good (or the best) ones to make the case for one or the other.

Then we can move on to the arguments themselves.

1 Like

@T_aquaticus I can already see that I owe you an apology, as I misunderstood what you were saying.

1 Like

I’m really hoping to explore the examples and analogies here, so that we can all understand what is being conveyed by them and whether or not they are good ones.

1 Like

You don’t owe him an apology. You are objecting in good faith, and taking the time to understand. No one should apologize for good faith resistance.

That is about right. However, @Giltil is speaking non-precisely, without mathematical rigor. There is an intuitive argument he is trying to make, but the problem is that statistics is not intuitive. @T_aquaticus is proposing that negative control to demonstrate a failure case for the logic his is using, it weree to be taken aligned with how ID argues the case.


In this case, it isn’t an analogy so much as a negative control.

1 Like

Okay, here’s what was said:

I went to a webtool that produces random protein sequences and got this result for a 100 residue protein:


The chances of any specific 100 aa protein is 1x20^100. Therefore, the chances of getting that protein was 1 in 1.6x10^260. According to ID arguments, there is no way that a random sequence generator could have produced that rare of a sequence, yet it did. In fact, I can keep doing this with ease.

In the argument from the ID side, it is said that a protein cannot form on its own due to the statistical improbability argument, which says, essentially, that there are so many possible combinations that stumbling upon what has obviously formed is inconceivable.

In contrast, the example above was put forth. I objected (incorrectly it seems) that what the random string generator made was not significant, but it seems that I was wrong.

What I am really hoping that someone can explain is how or why the example above is a good analogy for evolution. If, it is so. Or is it merely a counter to the improbability from the ID side?

1 Like