It just keeps getting weirder
(twitter.com)
You're viewing a single comment thread. View all comments, or full comment thread.
Comments (220)
sorted by:
The principle of parsimony states that the solution set with the fewest assumptions is the most likely best solution.
The hypothesis is that the shot that resulted was an accident.
The alternative hypothesis would be that the apparent result was not an accident.
Let's look at the first hypothesis. I am looking at this from a statistical mechanics perspective. Let's assume that the shooter was aiming for the center of Trump's head. The actual shot (assuming there was any bullet at all) appears to have had a minimal impact on Trump. I think it's safe to assume that a greater than 2mm deep impact would not have been so minimal. It also hit him in a spot that had maximal dramatic impact. If it had grazed his cheek for example, it wouldn't have caused anywhere near as much blood. It probably would have self-cauterized. That pretty much leaves the left ear (the one that would face the camera) as the only place for minimal impact but maximal drama. But again, it can only graze the ear. Any more and we would see his ear dangling, or blown off, etc. But it also has to get enough of a chunk to cause all the blood and not self-cauterize.
So how many 2mm squares (4 sq mm) are there that would have maximal drama, but minimal health impact? Maybe 20? 50? We'll call it 50 for simplicity. Now, assuming he was aiming for the center of Trump's head but missed, how many squares are there total between the center of his head, and twice the distance to the border of his head where the bullet (if it existed) appears to have hit? That total area has an about equal chance of being hit assuming a miss from center, so all are weighted equally. A bit of quick napkin math gives me about 50,000 possible 2mm boxes he could have hit, each with the same probability. If we were to expand that out to include his body, or the larger air around him there would be a great many more place he could have hit, but they would have a lesser probability, so I will not include them for simplicity.
In order for it to be an accident, we have to assume that he just so happened to hit one of the 50 choices for minimal health, maximal drama impact instead of one of the other 49,950 others. Each of those choices (akin to a microstate) is an individual assumption, because each is individually equally likely.
For the alternative hypothesis we really just need one assumption in this regard. That assumption is that the outcome was what was intended.
Thus, choosing the set with the fewest assumptions demands we choose the alternative hypothesis.
If we then consider Q this becomes more dramatic.
For example, Q stated:
#q/326
#q/35
#q/813
83 different q posts
#q/3387
In order for this to have been an accident, we must believe Q lied or is otherwise completely incompetent. There are thousands of reasons to not believe that, so that requires a new assumption for each piece of contraindicating evidence.
For the alternative hypothesis on the other hand, we only need one assumption. Q was telling the truth.
Here's an example of parsimony.
The shot was deliberate but barely missed it's target resulting in minimal damage.
Your example is over 300 words and says a shot involving the cheek wouldn't involve much blood and would be less dramatic. Which not only increase the complexity of the situation but involve assumptions on your part.
Your example also claims Trump was shot in the ear facing the camera. And of course Trump could move his head and that ear was facing away when it was hit. It was his right ear by the way. It was camera left but his right.
Another way your example reduces parsimony, is you claim that tiny 2 mm area was the target. This would require expert precision in the shot.
I find this assumption to be completely wrong.
I assume the tire the shooter intended to kill Trump and missed. You reduce parsimony again and increase complexity by assuming a deliberate specificity to how he missed.
In short, I think your statistical mechanical approach es leads you astray.
Fair, but not overly relevant. A cheek wound would either have to be extremely shallow (less than the 2mm window I suggested), or would be very devastating. That was really what I meant. I should have been more careful. That isn't really an assumption, but is a conclusion based on just looking it up. (Sorry, that's a bit graphic. NSFW)
Regardless, this assumption is irrelevant in that even if it is not granted (or you don't want to examine the pictures), it doesn't change the argument. It only allows for a few extra boxes to be considered "pro" hypothesis, but not enough to impact the argument in a meaningful way.
I said explicitly, "when he was being led away*, which would be, as you correctly state, his right ear (which I did confuse because of camera left<-> right inversion, my bad). The actual wound (assuming there was one) was not only something the camera couldn't see as it happened (so no possible analysis), it was the side that would be in the focus of the camera when he was led away after the fact. The point is, if the opposite were true on either count, it would have been easier to analyze for the first part, and far less dramatic for the second. Both of those facts of facing support the assumptions of the alternative hypothesis and not the "accident" hypothesis. In neither case however does it have anything to do with the core of the argument. It only changes the number of boxes.
Not true. I do not assume there was a shot. I only assume that the outcome was not accidental. It could have been a faked wound. In addition, it is not that hard of a shot to make (for an expert marksman). I say that not as an assumption, but based on the knowledge that I could make that shot. I couldn't do it all the time (I'd give myself one in ten at best), but I have made plenty of shots like that hunting or at the range, and I am NOT an "expert" by any measure (I know people who are much better than me). Regardless, I know it can be done intentionally, so no "assumption" required.
Regardless, there is no parsimony reduction here. That claim of a 2mm area was simply a guesstimate to set up the statistical argument. ANY assumption of target size works for the argument. It only changes the number of microstates. The argument itself is the same. If you want to say that assuming there is a target is an "assumption," well, sure, but it's the same for both hypotheses, so we'll call it a wash. Personally, I think it's more appropriately labeled a statement of known starting conditions than an "assumption" in the sense of parsimony.
It is not me who increased complexity, it is the number of possible end states that increases the complexity. That is the foundation of statistical mechanics and/or probability and/or information theory. The analysis is statistical. The number of end states is relevant to the analysis. You refusing to consider the number of states does not change that this type of analysis exists, is common, is a part of decision trees (the only place where Occam's Razor is relevant), and is considered both relevant and reasonable by many analysts.
From my perspective, you are reducing complexity in a contrived way by ignoring the number of possible end-states for each hypothesis.
I can't make you allow for my analysis, I can't make you consider the possible end states as relevant in the claim of the "accident" hypothesis, but I CAN say that you have not properly addressed the argument as presented, and have instead chosen to ignore it rather than addressing it.
Dude, I saw FARGO. Steve Buscemi's wound was gnarly enough for me!! He got shot in the face!!! He's taking the car.
And it is an assumption. The hidden assumption is that there would be any difference if the wound grazed his cheek.
I suggest that if the same events occurred and the wound barely grazed Trump's cheek, there would be absolutely nothing different about this board today. We would all still be just as concerned and upset.
To me cheek versus ear is a distinction without a difference.
Whoa.
This doesn't seem very Occam's Razor. Now you have to account for gunshot sounds timed perfectly with multiple different injuries spaced out a bit, and synced to video and the picture that captures the bullet in flight.
This introduces complexity. Your main issue is you are assuming intentionality in the location of the wound. You are assuming meaning in the location of the wound.
Which is why I think this is the wrong approach.
There's only a few end states that actually matter.
1 bullet completely misses 2 bullet is accurate. Trump is killed 3a Bullet it is not accurate, Trump severely injured. 3b. Bullet is not accurate, Trump with minor injury
That's how you look at something parsimoniously.
And you're right, I didn't look at the argument fully. Because it's not an Occam's razor analysis.
Occam's Razor is strictly a question of theory complexity. The problem here is, we are defining complexity differently. I am defining it as it is defined in formal analysis, in the information theoretic way. I am thinking about how to write computer programs to solve the problem of which is most complicated (similar to the travelling salesman problem). You seem to be defining it in the "how do I present this argument to my coffee shop buddy" sort of way.
If we define the basic terms differently, of course we can't have a meaningful conversation about it.
I couldn't disagree more. You are loosing too much information in your reduction. A loss of information will automatically make for a simpler model, but will also be counter to one of the requirements for the application of Occam's Razor, which is that ALL of the evidence must be considered. This elimination of information (or relevant evidence if you prefer) is, in my experience, very common in people's attempt to apply Occam's Razor, which is why it is so often misused/abused/misunderstood.
Occam's Razor doesn't in any way reject a theory before it has been analyzed. It can't be "not very Occam's Razor" until after all of the entire theories have been put into their "sets of assumptions" and then weighed.
You reject it because you think that:
Is something difficult. On the contrary, this type of stuff happens all the time in crisis actor situations. It is daily practice for agents provocateur.
In addition, all of the information you know about you received via video. Such things are faked all the time. Ever seen the movie Wag the Dog? Ya, that's not bullshit, that's the daily news. There are specific programs in the intelligence agencies whose job it is to do exactly these things. These things are trivial to make happen for these agencies.
That doesn't mean that's what happened, or even that I think it has, but your assumption that such things are extremely difficult (or somehow increase complexity) is a "hidden assumption" that I completely disagree with. On the contrary, I've seen too much evidence that make me think such things are trivially easy. There are too many examples of exactly those types of things happening. There are explicit statements of such programs existing by the FBI, CIA, corporations, etc.. They don't call them crisis actors because they suck at acting so bad it's a "crisis."