YT YouTube2Text

YouTube2Text → Transcripts → A Better Way To Power Scale (And Why Most Debates Fail)

A Better Way To Power Scale (And Why Most Debates Fail) — Transcript

by Sokol · 4,115 words · 616 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Stop
0:03it.
0:04HE HAD TO GET HEALED.
0:06GOD, THIS MOTHER is stupid. Power
0:08scaling debates have been going on for
0:09longer than I've been alive, and I've
0:11never seen them answer any questions.
0:13People who go into the arguments will
0:15come out of the same argument having not
0:17changed their opinion at all or even any
0:20understanding of their evidence that's
0:21been had. Some of the biggest examples I
0:23can think of are Minato versus Itachi
0:25from Naruto or even cross-verse battles
0:27like Goku versus Superman. These have
0:29been going on and have been prolific for
0:31forever, and the people don't really
0:32change their mind once they're on a
0:33specific train. And that's a flaw. It's
0:35not really a good thing. We want people
0:36to be able to hold a popular consensus
0:39based off of the evidence that's
0:40available. And people will say this is a
0:41naturally expected conclusion.
0:43Characters in fiction are complicated,
0:45but I don't think so. Power scaling
0:47debates don't fail because the
0:48characters are complicated. Power
0:50scaling debates fail because the
0:51arguments are complicated. And that's
0:53what this video's for. This video is to
0:54teach or reinforce for the more
0:56experienced debaters methods of
0:58argumentation to allow for a more
0:59concise conclusion and disallow for the
1:02crazy chaos that normally goes on in
1:04power scaling debates. He is using Okay,
1:06you do Are you stupid? But before we can
1:09even fix any issues, we need to be able
1:11to diagnose why these problems are
1:13happening. One of the big ones that I
1:14see when I watch these debates is
1:16constant derailment. People will talk
1:18and waste time and energy over
1:20characters or events that don't have any
1:22direct connection to what is being
1:23discussed at that time. Another big one
1:25is people will argue using different
1:27types of scaling like narrative and
1:28quantitative. If you're not online on
1:30the type of scaling, the two debaters
1:32will be talking past each other rather
1:34than with each other. And arguably the
1:36biggest issue I see is that people will
1:37have a conclusion beforehand, find the
1:40evidence to reinforce it, and then argue
1:42with that evidence rather than looking
1:43at all the evidence and then allowing a
1:45conclusion to form naturally. In short,
1:47people aren't arguing about the
1:48evidence, they're arguing about their
1:50predefined conclusions, which is why we
1:52have this problem of power scaling
1:54debates and never answering any
1:55questions. But before we try and fix
1:57these issues, let's take a step back and
1:58answer a very simple question of what is
2:01power scaling. Power scaling is nothing
2:04more than a tool used to measure the
2:06power of characters. It's like a ruler
2:08that is used to measure the length of
2:10objects, which can be used for
2:12comparison, power scaling is to measure
2:14the power of characters, and we can use
2:16that for comparison. Many people will
2:18utilize power scaling to determine
2:20versus battle outcomes like who would
2:22win against who, but it we need to be
2:24very clear that although these two
2:26things are related, they're two
2:27completely separate things. They're not
2:29synonymous. We'll discuss more about
2:31this later, but it's one of the most
2:32important things that we need to keep in
2:34mind going forward. With power scaling
2:36now being simply defined, we need to get
2:38into types of scaling. People will swap
2:40between the two throughout the debate
2:42without much thought into it, but it's
2:44very important because which type of
2:45scaling you're using will determine the
2:47efficacy of your argument. The first
2:49type of scaling and probably the one we
2:51see the most often is narrative scaling
2:53theory. This is where the connective
2:56tissue for the arguments used to power
2:58scale is produced by the story or the
3:00narrative itself. This is powerful
3:02because any coherent story will have its
3:04built-in power system structure for
3:05storytelling purposes, and we can use
3:07that internal power structure to
3:09determine the power of characters in a
3:11narrative fashion. Narrative scaling is
3:13usually the most precise form of scaling
3:16because you're using the actual story
3:18itself in order to deduce chains of
3:20logic. The second type of scaling is
3:22quantitative scaling theory. This is
3:25where there is no connective tissue in
3:27order to power scale the characters
3:29through lines of logic. This is mainly
3:31used for when you have two separate
3:32characters from two separate stories
3:34that have no connective tissue to
3:35connect them together, so we need to
3:37find another method for comparison. This
3:39method relies on the axiom that the laws
3:41of physics are roughly the same
3:43throughout all universes and that you
3:45can utilize those laws of physics to
3:46determine physical quantitative
3:48quantities in which you can use to
3:50compare to other characters to determine
3:53who is more powerful than the other.
3:54Because of this axiom, quantitative
3:56scaling theory is generally considered
3:58to be less accurate or less precise than
4:00the narrative scaling theory. This idea
4:02of less precision is important because
4:03now we can set up a hierarchy. Narrative
4:05scaling is of higher precision than
4:07quantitative scaling and therefore takes
4:09precedence if it's available. We only
4:11utilize quantitative scaling theory when
4:14narrative scaling fails. I've seen this
4:16in power scaling debates all the time
4:17where someone will try to argue that one
4:18character is more powerful than the
4:19other, but the narrative just does not
4:21support that argument, so they will
4:23utilize quantitative scaling with its
4:25implicit less precision in order to
4:27bolster that argument which the
4:28narrative does not deem to be true.
4:30Being able to move mountain-level
4:31expansion at lightning speeds does show
4:33he can perform country-level supports. I
4:35already posted the math and the calc for
4:37it.
4:38>> Tailed Beast Bombs should be
4:40multi-continental, and I have a category
4:42Now, in normal debates, people will just
4:44accept these quantitative reasonings,
4:46and that's not really a good thing. With
4:48this hierarchy, we can now call this out
4:50as bad practice. If a narrative
4:52connection exists and the narrative
4:54connection is saying one thing, why
4:56would I listen to your quantitative
4:58scaling, which has less precision
5:00overall compared to narrative scaling?
5:02We can formalize this in a very simple
5:04rule. If narrative precedence exist,
5:07then narrative scaling should be
5:09utilized. Quantitative scaling is only
5:12utilized when there is no narrative
5:14connection between the characters. For
5:16this particular video, we will talk
5:17about narrative scaling theory and we'll
5:19leave quantitative scaling for another
5:21video. So, how do people normally do
5:23narrative scaling? Well, it's very
5:24simple. We determine the power of
5:26characters in relation to each other
5:28based off of what is shown in the story,
5:29whether it be through feats, statements,
5:32databooks, you name it. But one of the
5:34biggest issues I see within narrative
5:36scaling is people will be arguing over
5:38which interpretation is more valid. Is
5:41it feats are the greatest? Is it
5:44databook statements are the greatest? Is
5:46it statements are the lowest? This
5:47entire concept is subject to serious
5:50scrutiny and have whole debates just
5:52over this specific topic. But I think a
5:54very satisfying answer to this question
5:56is that it is not that one is more
5:58important than the other, but rather
6:00that all of them have an equal amount of
6:02importance. A bad way to argue this is
6:04saying that feats are the top, then it's
6:06databooks, then it's statements, and so
6:08on and so forth. But scaling using this
6:10type of rigid outlining for the level of
6:13importance leads to serious logical
6:15conclusions that no one would
6:16intuitively agree upon. If you're
6:18someone who believes that databooks are
6:19the most reliable, then you're led to
6:21believe that Temari from Naruto could
6:23destroy a universe with a single blow of
6:25her fan, something that no one really
6:27believes. If you're someone who believes
6:28that feats are the greatest, then Vegeta
6:30from the Saiyan Saga didn't destroy a
6:32planet, so he must not destroy a planet,
6:35which no one largely agrees with either.
6:38Using these rigid lines of arguments
6:40lead to conclusions that no one
6:41intuitively believes. In order to
6:43account for this as well as my statement
6:44that all of the pieces of evidence have
6:46the about the same amount of priority,
6:48we need to define something which we
6:49will call narrative consistency or or
6:52narrative cohesion. Narrative
6:53consistency is the intuition of the
6:55power structure presented by the media.
6:57This can be defined as a claim being
6:59strong if it's consistent with the power
7:01structure of the story and a claim being
7:03weak if it is inconsistent with the
7:05power of the story. Therefore, we can
7:06look at the two examples that we had
7:08previously like Temari being able to
7:10destroy a universe. If we look at Temari
7:12throughout the series, we see that the
7:14power structure of the series has
7:15developed has shown that Temari is not
7:18that strong and is not universal.
7:20Therefore, that statement is akin more
7:21to hyperbole rather than a direct fact.
7:24As for Vegeta, we have seen feat-wise of
7:27characters weaker than him like Piccolo
7:29being able to easily destroy a moon as
7:31well as characters stronger than him
7:33like first form Frieza being casually
7:36able to destroy planets. This leaves
7:38Vegeta in that little middle sweet spot
7:40where he is most probably able to
7:42destroy a planet, thereby giving his
7:44statement a lot of credence. With all
7:46this stuff now in mind, the question
7:47comes, how do we actually scale? The
7:50answer is we draw. No, I'm serious.
7:53These lines of argumentation followed by
7:55chains of logic or analyses, which means
7:58you could literally draw out your
7:59framework and follow it in a logical
8:01line. If you're not able to do this,
8:04then it probably indicates that you have
8:05a bad argument. Some of the best power
8:08scalers in the world already do this
8:09intuitively. They don't do it explicitly
8:12by literally drawing it out, but in
8:13their mind, they're able to construct
8:15these visualizations in order to form
8:17their arguments. This is usually what
8:19separates some of the best power scalers
8:21from some of the worst ones.
8:23Fortunately, not everyone needs to be
8:24these intuitive geniuses. We can
8:26literally take something as simple as a
8:27pen and paper and draw these out
8:29ourselves explicitly. But how do we draw
8:31these lines of chains? Well, I think the
8:33best way to do it is something using a
8:35power graph where the higher you go up
8:37on the graph, the more powerful a
8:39character you are. This visually allows
8:41us to see how the characters line up in
8:42comparison to one another with the one
8:44higher on the graph being more powerful
8:46than the one lower on the graph. Now,
8:47there are three different ways we can
8:49connect characters on the power graph.
8:51The first method is linear scaling. This
8:53represents a direct increase in power
8:55from someone who is stronger to someone
8:57who is weaker. For example, using a
8:59Naruto example, if I wanted to power
9:01scale who is more powerful, Hokage
9:03Minato or Part 1 adult Kakashi, we can
9:06first set up the graph with Hokage
9:08Minato. Right under Hokage Minato, we
9:10have young Hiruzen, and under him, we
9:12have old Hiruzen, and finally under old
9:15Hiruzen, we have Part 1 adult Kakashi.
9:17Using this power graph, we can visually
9:19see that adult Hokage Minato is more
9:22powerful than Part 1 adult Kakashi,
9:24which is an intuitive answer. Next is
9:26parallel scaling in which characters are
9:28roughly approximate to each other in
9:30power based off of either narrative
9:32cohesion or what's explicitly shown in
9:34media. For example, if we wanted to plot
9:36where end of Part 1 Naruto and end of
9:38Part 1 Sasuke are on the power graph, we
9:40can put them right next to each other.
9:42We can do this because in their final
9:44fight, they were presented as being
9:45equal-footed rivals. And the last method
9:47to connect characters on the power graph
9:49is through indirect scaling, arguably
9:51the most utilized one of all the three.
9:54This is where two characters are
9:55connected to each other via a
9:56combination of linear or parallel
9:58scaling logical steps. This type of
10:00scaling is usually used to compare
10:01characters who have never had any
10:03interaction before within the narrative.
10:04For example, if I wanted to analyze two
10:06characters who have never interacted
10:08before such as part one Itachi or part
10:10one Temari, I could set up the power
10:12graph like this. With part one Itachi
10:14being at the top, part one Sasuke being
10:16below him, with Naruto being parallel,
10:18and Gaara being parallel, with Temari
10:21being under Gaara. This conclusion shows
10:23us that part one Itachi is more powerful
10:26than part one Temari, which is a logical
10:29conclusion that most people would agree
10:30with. That's a really easy example. We
10:32didn't need to set up a power graph to
10:34answer that question. What if you wanted
10:35to do an example that is much more
10:37contentious? Let us say I wanted to
10:38discern who is more powerful, Hokage
10:41Minato or part two alive Itachi. We can
10:43start our first graph with Hokage Minato
10:45being stronger than younger Obito when
10:47he fought him believing it was Madara.
10:49Minato also saw Naruto fighting Pain
10:51within part two of Naruto. He also
10:53implied that Naruto would need more
10:55power than the one he used to fight Pain
10:56in order to defeat a character like
10:58younger Obito which Hokage Minato had
11:00defeated. Therefore, using a narrative
11:02chain, we can connect Pain to be under
11:05that young Obito. And finally, for
11:06argumentative purposes, we can say that
11:08part two Itachi is relative or parallel
11:11to that Pain. Some people will argue
11:13this and saying that part two alive
11:15Itachi is weaker than Pain, but that's
11:17fine. If we put them relative to each
11:18other, the conclusion still stands. Via
11:20this logical line of steps, it would
11:22show us that Hokage Minato is more
11:24powerful than part two alive Itachi.
11:26Now, before we get into it, pitchforks
11:27down, pitchforks down. This is a power
11:29scaling graph, not a versus battle
11:32graph. Earlier we made the distinction,
11:34and we need to make the distinction here
11:35again.
11:36Power scaling is not the same thing as a
11:38versus battle. It does not always follow
11:40that someone higher on the power graph
11:42will defeat someone lower on the power
11:44graph. It can be the case, and there are
11:46even instances of people who are lower
11:48on the power graph being able to defeat
11:50characters who are higher on the power
11:52graph. The reason for this is because
11:54battles are more complex than pure
11:56scaling. It depends on context,
11:58characteristics, or even match-ups,
12:00something akin to rock, paper, and
12:01scissors. We can see this play out in a
12:03cross verse power graph. For example,
12:05Goku and Light Yagami. Goku is much
12:08higher than Light Yagami on the power
12:09graph, but does that necessarily mean he
12:11would be able to kill him in a versus
12:12battle? The thing is Light Yagami is
12:14quite smart, and Goku is quite
12:16dim-witted when it comes to intellectual
12:18battles. Light Yagami could easily learn
12:20the name of Goku and write it within the
12:22Death Note, which Goku has no means of
12:24countering, and thereby leading to his
12:26death. Meaning, just because Goku is
12:28more powerful than Light Yagami, Light
12:30Yagami does have the probabilistic means
12:32in order to kill Goku nine times out of
12:3510. This goes to show that even though
12:36Goku is way more powerful than Light
12:37Yagami, Light Yagami has the tools and
12:39the characteristics in order to win a
12:41battle against Goku using his arsenal
12:44and his intellect. Therefore, from our
12:45power graph, even though Hokage Minato
12:47is stated to be more powerful than part
12:49two alive Itachi, it does not
12:51necessarily guarantee that Hokage Minato
12:54could win a versus battle against part
12:56two alive Itachi. Such a conversational
12:58requires more arguments about more
12:59contextual events like powers,
13:01abilities, strategies, and so on.
13:04However, utilizing this exact same
13:05drawing methodology we have just
13:07derived, we can utilize it to answer
13:09these more deep conceptual questions.
13:11Let's say we did want to analyze this
13:12versus battle against Hokage Minato and
13:14part two Itachi. And let's say that I
13:16believe that Itachi is able to kill
13:18Minato with his Amaterasu. I believe
13:20that it is a probable conclusion. Well,
13:22we can analyze my claims based on the
13:24evidence presented. We can start with
13:25Hokage Minato being faster than the
13:26Fourth Raikage. The Fourth Raikage, even
13:29in his adulthood, does not compare to
13:31Hokage Minato's speed when he was alive.
13:33Additionally, the Fourth Raikage was
13:35able to dodge Sasuke's Amaterasu. And
13:38there is no narrative indication that
13:39Sasuke's Amaterasu is any different from
13:41Itachi's Amaterasu. Therefore, we can
13:44form this graph here. And therefore, I
13:46can reform my previous conclusion to say
13:48that Itachi's Amaterasu is not actually
13:50as decisive of a win con as I initially
13:52believed it to be in order to defeat
13:54Hokage Minato. And we can continue this
13:56with other data points like strength,
13:58combat type, intellect. And with enough
14:00data points, we can come to a more
14:02decisive conclusion on who is more
14:04probabilistic to win a versus battle
14:06against the other. Now, there's a big
14:07issue that might come out of this type
14:08of graphing. One is that media is quite
14:11big and complex. There may be lines of
14:13logic that you can construct which are
14:15contradictory. The way to counteract
14:17this is by evidence convergence. The
14:19more paths you can create to lead to one
14:21conclusion over the other will show that
14:23that conclusion is probabilistically
14:25more true than the other. For example,
14:27if you're able to describe 10 different
14:28logic chains that shows that A is
14:30greater than B, but you're only able to
14:32show one logic chain that shows that B
14:34is greater than A, then it is more
14:36probabilistically true that A is greater
14:38than B in terms of evidence convergence.
14:40That, of course, I would be lying to you
14:42if I said this was foolproof. There are
14:44a lot of pitfalls that we need to be
14:45careful when constructing these logic
14:47chains. The first is just because you're
14:49able to create a chain to connect two
14:51characters does not necessarily mean
14:53your logic is foolproof. Fortunately,
14:55there are ways to discern this. One way
14:57we can discern this is using what I like
14:59to call the water flow test. Imagine you
15:01start at the top of the graph, and you
15:02pour water at the top of the graph.
15:05Well, water can only ever flow
15:07downwards. Therefore, if you ever find a
15:09location where the water has to flow
15:11upwards in order for your argument to
15:12work, you have found a logical
15:14inconsistency. For example, let us say I
15:16wanted to compare Hokage Minato and
15:18Itachi as well. Let's say I start from
15:20Hokage Minato, who is greater than young
15:22Hiruzen, whom is greater than old
15:24Hiruzen, whom is relative to part one
15:27Orochimaru, whom is weaker than Pain,
15:30whom is relative to Itachi. Regardless
15:32however I set up to be higher, whether
15:34it be Hokage Minato or part one Itachi,
15:37in both instances the water needs to
15:39flow upwards in order to reach a
15:41conclusion. Although the characters are
15:42connected, no question is actually
15:44answered. That is because there is a
15:45totally separate question that is being
15:47asked now that has no answer within the
15:49power graph. The question is, how do
15:52Hokage Minato and Pain relate to each
15:54other? The graph that was just
15:55constructed does not give any answers on
15:57that conclusion, and therefore the
15:59entire statement has no conclusion. The
16:01last big issue is assumption stacking.
16:04Each logical chain you have has a little
16:06bit of implicit uncertainty. Therefore,
16:08the longer the chain is, the more the
16:09uncertainty compounds, thereby making a
16:12less reliable conclusion. If you have a
16:13graph that is 10 chains long versus a
16:15graph that is two chains long, the one
16:17that is two chains long will have a
16:18higher predictability power because of
16:20the less uncertainty within it. We can
16:22see this visually, too. If I have this
16:24two-chain power system, one of them
16:25needs to fall for the logic to fail, but
16:28I have a higher probability of defending
16:30the two versus a 10-chain link with 10
16:33logical chains. Only one of them needs
16:35to fail in order for me to fall, and it
16:37is unlikely for me to be able to justify
16:38them all as strongly as my two-link
16:41chain. Therefore, the 10-link chain has
16:43a higher probability of failing. But
16:45there are a lot of benefits with using
16:47this visual system. The first is that
16:49arguments are in a literal chain. You
16:51can literally follow your line of
16:53thinking instead of just going blind in
16:55the dark trying to grab onto whatever
16:56you can. There is a line of logic which
16:58you can follow, and you could argue
17:00against someone's line of logic
17:02directly. This goes against the whole
17:03derailment point we mentioned earlier.
17:05People are no longer talking about
17:07unnecessary characters, and if they are,
17:09you have a way to call them out. If they
17:11cannot connect that character into the
17:13chain any logical manner, then it is a
17:15worthless conversation. It does not need
17:17to be had. A little bit of glazing here,
17:19but Six with his high intuition is able
17:21to see these happen in real time without
17:23physically needing a drawing to see
17:24this. He notices in his debate with
17:26Kabuki. He one-shots Shin with wood
17:28style chakra. Kabuki, I want you to go
17:30ahead and try to link Shin to Madara in
17:32any sort of way. Go ahead. What do you
17:34mean? Many other debaters would have bit
17:36on that point, have bit on it. To be
17:38honest, I would have as well, but he was
17:40able to discern that, "Hey, this
17:42character is worthless to our line of
17:43argumentation. I can't use it in my line
17:45of logic, and he hasn't demonstrated his
17:47ability to use it in his line of logic.
17:49Therefore, we don't even need to discuss
17:51it." That is high intuition, and without
17:54a graph, many people may not have been
17:56able to discern it. The second benefit
17:58is the benefit of realization of where
18:00your arguments are the weakest. The
18:02thing is, many people, like I said
18:03earlier, have a conclusion and will find
18:06the evidence to pick it. But your brain
18:07is really powerful at gaslighting
18:09yourself. You're able to make yourself
18:11believe that your evidence is stronger
18:13than it is because your brain will hide
18:15all these intimate assumptions that you
18:17just are not privy to. When you're
18:19forced to draw them out and list out
18:21your reasons of why these logical chains
18:22happen, you're both validating your
18:24arguments that are strong and finding
18:26out where your arguments are weak. For
18:27example, within my power graph of Hokage
18:30Minato and part one Itachi, I can see
18:32that the main flaw that people would
18:34attack is the young Obito and Pain
18:36comparison. If I am not up to the task
18:38to defend that point, then my whole
18:40point falls apart. I should be able to
18:42defend that point and justify that point
18:44with the narrative evidence at hand. But
18:46let's say you want to do power scaling
18:48like this, and let's say you want to
18:49analyze two characters, and then after
18:51every angle you attack it from, you
18:53realize that no chain can be made. One
18:55may believe that this is a flaw, but
18:57it's actually not. Arguments can only be
18:59made out of lines of reasoning from the
19:01narrative, whether it be feats,
19:03statements, data books, or whatnot. If
19:05there is no evidence available for you
19:06to make any lines of logic, then there
19:09is no logical conclusion that you can
19:11come to. This is actually a strength of
19:13the system, not a defeater of it. If no
19:15narrative links can be made, then
19:17narrative scaling theory falls apart,
19:18and we shift our focus to a different
19:20type of scaling, quantitative scaling.
19:23Unfortunately, the way we currently do
19:24quantitative scaling has some serious
19:26problems, and fixing those problems may
19:28open up a whole new world of conclusions
19:30which we currently take for granted. But
19:32that is something we will discuss in the
19:33next video. As for this one, I hope
19:34you've learned something, and I hope
19:35you're able to apply this into your
19:37future discussions and your future ways
19:39of power scaling characters through a
19:41narrative lens. But that's all I have
19:43for today, so I'll see you in the next
19:44quantitative power scaling video.

About this transcript

This page contains the full transcript of A Better Way To Power Scale (And Why Most Debates Fail) by Sokol, generated from the public captions YouTube serves with the video. The transcript has 4,115 words across 616 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.