No Art of the Deal, Straight Blockade, Cybercrash, and oil back up again!

Okay. Uh before we jump in, we are coming up on May 1st. In a couple of weeks, about 3 weeks away from it, a little less than 3 weeks, the uh applied energy analyst uh launches officially. You'll be able to start booking your exams, uh beginning of May. The exams I believe will start anytime after May 21st. There are no set times. It's whenever you feel comfortable. You book your exam. Uh the textbook is done and uploaded to Gumroad um where you can uh purchase it for $50 uh and and you just download it. If you want the PDF, I will stress you don't need the PDF. All of the content is online. Uh but you do have to read it on the screen. But I mean, if you have your tablet in front of you, it's the same as reading a a PDF on the screen. You're just reading you're just reading it off off uh a website screen as opposed to a PDF screen. But if you really do want the PDF, and it is useful if you wanted to cut uh a section out and put it into an LLM and say, could you explain this uh in in in more accessible language or in in a different way? Uh it is useful for that. Um, it is a long PDF because it has everything in here. It's 1,092 pages or 11, 1100 pages. Uh, and it's all clickable. So, uh, you can, uh, click there's oil and gas. If you click on that, it takes you to chapter 6, which then has its, uh, table of contents there. I made the decision to put the table of contents in front of every chapter as opposed to all at the very beginning because it'd be 30 pages of table of contents. So, at the very beginning, you click the chapter six, and then from chapter six, you can go on to what you want, classification, two-tiered market. There you go. Right. And it'll uh it'll get you through whatever it is that you want. And each one, as opposed to uh listing the page number of of where you are in the entire text, each chapter is the page number within just the chapter. I thought that was a little bit easier to use as well. So, the link for this is in the description box if you want the PDF. Uh, but I do stress you do not need the PDF because all of the content is online. uh I believe in uh sometime this week you will see an option that if you don't want to buy applied analysis because this sits inside applied analysis you can just buy the applied uh energy analyst uh course content which will have all of the readings uh and all of the Qbanks associated with this uh you could just buy that one thing if that's all you needed. We'll start with CPI this week. As I do this, uh the news as of late last night, Saturday night, was that after 21 hours of negotiating, uh there is no deal in uh the Iran uh Israel uh conflict, which the US uh is a party to. So, they're going to be there for a while longer, it seems. Or maybe they take the night off. They wake up in the morning and each side says, you know, maybe we can give a little here, give a little there. Uh, as I do this, there is no extra news. Uh, it is it is pretty much done. Uh, Iran has some red lines. It won't cross. The US has some red lines and those red lines do not overlap. And um I think the last headline I saw uh is um about this thing is China is sending uh some military equipment to Iran. What's uh Bitcoin you if you were watching the risk assets, Bitcoin last night on that news just dropped. Not much like about 1,500 bucks, but it dropped and it's still down as of this morning. And I haven't checked Hyperliquid uh this morning to to see what the price of oil is, but as I get to uh further screens maybe in the background, I'll call it up and we'll look at it a little later. Let's look at inflation. Headline was not pretty. 0.9 everyone uh the expectations uh when I say everyone the consensus pretty much got it spot on. Bring it out to three decimal places 865. However, once you strip out energy and food, the core.196 very well behaved year-over-year 2.6%. Here is your energy category in here. That's year-over-year. Look at the size of those numbers. Super core. This is uh taking out food, energy, and shelter. Why do we take out shelter? The three things people have to pay for. Why do we take it out? We take out food and energy because they're volatile to begin with and their prices move for uh all sorts of reasons, not necessarily demand. Uh but usually when we get price spikes in food and energy, it's usually a supply related issue. Usually, not always, but usually a supply related issue. Uh but energy costs feed into uh services and goods because well GDP is uh requires energy. The US is probably one of the most energyefficient countries when it comes to how much energy is required per dollar of GDP. We uh take out shelter for super core because shelter is not market observable prices. It's asking people if you had to pay to rent the house you live in what would you pay? Uh so it's imputed. So we take that out.139.139 on the month over month year-over-year 2.27 you're you're basically in spitting distance of that 2%. If uh we want to look at alternative measures of CPI because some measures are really volatile like sometimes you'll see used cars way up this is the zero line used cars and trucks and then and then you see negative numbers for a while like for instance used cars and trucks we have a negative number you have a negative number here look at this negative uh.1 uh here you have a negative number on medical care commodities uh here you have a positive number so sometimes they they they are volatile um within themselves so there are different measures is one is taking the median and the other one is just trimming off uh the ends getting rid of the ones that are really really low uh because that could just be an anomaly and really really high because that could be an anomaly and looking at the uh fatter part of the curve uh between those two extremes. That's called a 16% trimmed mean. I mean you could have an 8% a 16% a 20% whatever you want but this is a 16% trimmed mean. 8% gone here, 8% gone here. Let's just uh look at the median uh and I've highlighted here July uh of 2025 is 3.65%. By December it was 3.1%. So it had dropped 0.55 over 5 months or.11 per month and then we uh got into the narrative inflation is sticky. inflation is sticky yet from December to January drops another.13 January to February drops another.12 February to March drops another.13 doesn't sound sticky to me if we look at it uh uh using the median so um I figure you're dropping about 0.12 a month we want to get to 2% let's go to the 72 roughly about six months you're at 2% but what happens to the economy what happens to the job market in between there that if uh as a central banker uh and we've heard them say this before, you say these the the interest rate works with a lag one to two quarters. Well, if it's 1 to two quarters, guess what? 6 months is two quarters. So, you can start cutting rates now looking at the median saying, you know, we wanted to see progress towards our target. What more progress do you want? Right? In five months it dropped 0.55 and then uh over the next uh from December uh to March this is one two three more months uh it went from 3.1 to 2.72 dropped another 38. What are we waiting for? Now we're going to use the 16% trimmed mean. Uh it was 3.2 back here. Uh there it is there below the 2.72 maybe about 2.5. So that measure is coming in lower as well. And again, I point to our super core. Um, we're paying attention just to the wrong things. The interest rate won't won't change the price of food. It won't change the price of energy. Energy right now, that's a supply driven problem. The interest rate is ineffective on supply driven problems. The best way to cure a supply driven problem is with more supply, not with higher interest rates. you get more supply by lowering the interest rate to encourage investment. So I I seem to think that all the narration around inflation inflation inflation is the very thing that is building inflation expectations uh into people's uh um assessment of where we're going in the future. And that in itself can create inflation by uh saying well if inflation is going to be 5% I need a raise of 6% and then that feeds in to the core very quickly but so far we're not really seeing that and sort of as an offset don't know if you want to say this is good or bad but uh we are in a jobless uh economy and no jobs being thrown out wholesale but no jobs being added either at least not any degree that you would uh certainly put that in headlines. So without that, you're not going to have wage pressure. Uh and much as there was over the last 40 years the question of should we use automation or should we use labor. Ah there's China where labor is really cheap. Let's send the jobs there. Had China never come along most of those jobs would have been lost automation. Now should we use labor or should we use AI? AI is a very real thing and I mean we've all seen the stories of Claude Mythos and the company saying a so powerful we we simply can't release it. Brilliant marketing I believe them that it is a it is a better model but brilliant marketing is is this is reminds me of of South Park. I don't know if you ever saw the episode where Cartman bought an amusement park and he only wanted it for himself. So, no one was allowed to go in. But because no one was allowed to go in, everyone wanted to go in. Everyone wanted it. This feels like a South Park marketing uh tactic is, you know, how do we get more customers? I know. Let's deny everybody the use of it and everyone will line up. How many people, Show of hands, come on. How many people can't wait to see Mythos now? that if you weren't a customer of Claude uh or Anthropic, you're probably saying, "Let's let's head over there. I want to see this thing when it comes out." I think everybody wants to see it. Brilliant marketing. But if it is, I mean, I use Opus. It's killer. Uh if this is even more killer than the killer, whoa, talk about an apocalypse on jobs. I you have to start to lean into that a bit. But that will be the decision going forward. So I don't know so much that inflation expectations will bleed into wages if the decision inside the organization is AI or wages. Okay, let's look at uh rates and yields money market uh pulling back a little bit but uh not much action on the long end of the curve uh sort of a week off. Day by day you wouldn't know that but week over week you look at it you say ah long end of the curve took a week off tightening since the beginning of the year long end of the curve is higher that is all consumer credit tighter since the beginning of the year KRE banking index had a good week 4.45% 45% 6.37 year to date. Uh just beating out the small caps, but certainly beating out SPY and Triple Q. Dollar took a dump, but I'm sure it's going to get most of it back. Uh we'll see what happens when uh it opens tonight. Uh for the next week, we get PPI on Tuesday. I don't know if that matters to anybody with uh the backdrop of uh Iran and Israel war uh uh looming. Well, continuing on, I shouldn't say the Iran Israel war. It's the Israel, uh, Yemen, Lebanon, Iran war. A war on three fronts with the US coming along for financial support. Uh, everything had a good week last week. Triple Q 4.46%, IWM 3.98, spy 3.6. A note on uh, Triple Q here. You wouldn't know 4.46 46 uh based on the absolute routing that happened in software and especially cyber security. So I've uh I've um asked Claude uh or got into a debate with Claude and then really pushed back. I asked it to create the worst bare case it can and it is terrifying. uh as you're reading I go these are all good points but I systematically attacked each point and it still held to its bare case although it admitted that I had weakened the bare case quite a bit we'll look at that and uh as I do this I don't know the answer to this uh YouTube tells me now with each video that I can attach up to five files I haven't tried it yet uh but I did create a PDF of the prompt and the answer uh with my dialogue for cyber security. So when I upload this video, I'll see exactly where that feature is. No promises. I haven't tried it yet. It tells me I can upload five files. If I can. I will just upload the file. So I can't tell you right now I've uploaded the file. Just go ahead and download it because I don't know if I can. However, there's your notice. Let's look at ZQ holding its ground. uh zero cuts in eight meetings and the Worsh confirmation apparently is not going to happen uh at the beginning of next week because the um paperwork wasn't filed correctly. That doesn't sound right to me. That at this level you wouldn't have the paperwork filed correctly and that you wouldn't just say, "Well, this is important enough that do you mind if we make an edit right here and right here? Let's fix this right now." or send it back and say, "Can you make these adjustments right now?" But the paperwork wasn't filed correctly. I don't know. Doesn't sound right to me. But we're getting uh no movement on uh on Fed funds futures. One thing that I have noticed over the last week or so is the conversation is shifting towards Worsh. there's more conversation uh about his plan and the battle uh uh and the battle sheet the balance sheet uh so it's nice to see that that's happening and that u ZQ is still being rather well behaved but not well let's uh you know we're not pricing in anything at least at least it's not acting like a meme stock like it was a few weeks back and pricing in rate hike after rate hike every single today. It's a lot more well behaved, but I still believe underpriced. Uh TLT not having a good time. Uh I have been asked numerous times, uh is it time to get into US debt? Is it time to get into it now? My answer is very simple. Uh a year ago, I would have said maybe. Uh you know, with with Worsh coming in the answer was probably not. Now it is a a firm absolutely not. um this to to help finance Israel's uh Israel's war against Iran. The US is is going to uh spend much more money than it has spent so far. Uh or I think over 20 billion on that number. Uh add a zero and maybe another zero if it carries on for 2 three years. Uh but uh going from where you are now to 100 billion. 100 billion the US doesn't have by the way that that gets added to the debt. The taxpayer is going to have to come up at some point in the future whether through higher taxes on your income, higher taxes on your consumption or just just taking assets that if you're going to think that you're going to inherit that $2 million uh estate from your parents, maybe it's more like a million with a 50% tax. There must be some some payment for all of this. um that's just going to add to the debt. I don't see I do not see any concern from any administration about the level of debt. And this administration has floated a $1.5 trillion uh spending package for defense um with an aging population. So there goes Medicare and Medicaid. Social Security is about to get the biggest increase it's had in a long time because of the inflation uh impact. The interest rate if it's if we believe in zero is not coming down. Probably set to head to 1.3 to 1.4 trillion over the next year. So no, I don't think it's time to buy US debt at all. Now, I hold US debt and uh that matures anywhere between 2041 and 2048. Uh and two years ago, I was unconcerned. I'll take it. I got it at 5.6% yield. It it was the lows in October 23 when when uh uh when everything just really spiked. Was it 23 or 24? I forget. I've had it for a number of years now. um it has appreciated since then. So I can always I can always get out now. Now a year ago I wouldn't have thought about getting out of US debt. I would have said well I got what I got and I'm going to hold it. I have abandoned TLT by the end of the year. I completely abandoned TLT in all my accounts everywhere. Uh and move that into treasuries. Now I'm thinking probably by the end of this year it might be time just to leave US debt altogether. Unless somewhere along the way someone with some authority stands up and says, "We can't keep doing this, guys. We we got to deal with this." And for this one, you kind of have to look at the Fed and the Treasury together. Uh as being the adults, hopefully being the adults saying we got to do something about this. This is this is part of the financial stability of the system because if the world loses faith in US debt, everything everything gets more expensive because all consumer debt is priced on the yield curve. And I don't know if there's been a time uh where and I don't want to say it's anti-American sentiment around the world. It's it's not the United States of America. It's Trump's America that has a level of um what's a good word to use here? Uh anti-Trump America sentiment that I think has exceeded almost everything else. Um, I don't know about the willingness to accept the US dollar in in a crisis like we just had and like we have right now. For the US dollar to peak at about a hundred and then really give up its gains is is is not really that great. The US dollar should be setting records at this point. Um but Iran uh that is charging for ships going through the straight of Hermuz is saying pay us in uh yuan and bitcoin not even accepting the US dollar. That is telling. So my answer to is it time to buy US debt? There's got to be another country that where where you can buy the debt of another country. There's got to be something else. Uh because if this continues on, US debt's going to get even worse. If this does not continue on, that US dollar is going to drop, which means the purchasing power of the dollar in your own currency, whatever return you think you're going to get on fixed income will just be taken away by the currency. So if you uh have no reason to be in US debt at this point in time, I would say find something else. Okay, real yields. Uh not much changed here and same with break evens. Not much has changed there. Corporate spreads uh tighter. Uh look at high yield tighter by 22. Uh double B by 19. Triple C by 28. Backdrop of private credit stress is not keeping people away from uh stressed debt. There is triple C. April FOMC uh 97.9% on nothing happening. down from 99. It's basically 100%. Uh Powell is not going to move. Uh in June, there is that question mark. Is it going to be wars or is it going to be Powell? Uh but 94% uh probability uh basically the same week over week that nothing nothing much is going to happen. So watching paint dry or uh a better analogy watching golf just as exciting. Uh, I shouldn't say that. Somebody got a hole in one. That would have been exciting, right? Uh, let's have a look at our metals. All of them did well this week. Uh, we're going to take a closer look at copper uh this week. Uh, and uh, uh, zoom in on that one. Uh, gold up, silver up, copper, aluminum, platinum, platium, all up. Platium down for the year. Big move was oil down 13%. look for it to come right back if we don't have some resolution by the time we get to tonight. Uh I think the threats of uh genocide and war crimes uh from Trump will begin again. I mean he did that last week. Threatened a war crime. I will destroy all all your electricity and your infrastructure. Then he threatened genocide. A whole civilization will die tonight. Don't want to do it. Really? Right to genocide. right to genocide and war crimes. How he hasn't got a Nobel Prize yet? Makes you wonder, right? H what's Norway doing? Blame Norway. Blame blame Denmark and then take Greenland. Uh comedians are having a great time with this. Natural gas uh down as well. Now, uh, Trump does say, you know what, he I think he said it, uh, in a press conference sometime in the last 12 hours, there are empty ships heading to the US to fill up with sweet, beautiful oil. Um the travel time from the US to the markets uh that were served by uh oil coming through the straight is a longer voyage that if you think about the charge from Iran, it's a dollar a barrel of oil that you're carrying on the ship. I believe just the charter and the extra days alone would add more than a dollar if you're bringing it from the US and front month is more expensive than Brent already uh uh on a benchmark basis. Now I don't know what the what the benchmark uh what the discount or premium is uh in uh in the Middle East to to that uh to that benchmark because no one really pays the benchmark price. There's always quality adjustments uh and there's always location adjustments as well. So, it's just a benchmark and uh all different blends of oil. There's something like 120 130 blends of oil in the world are all priced at either a premium or a discount to the benchmark. So the benchmark sets the quality uh and the location and then any deviation from that quality up or down and any deviation from that location favorable or unfavorable changes the price. But I think that a a a transport from the US to Asia just the extra days and the charter costs per day would add more to the price of oil than just paying Iran a dollar. So, I don't know that uh that that that would work out too well. I don't think it will. So, I'm not willing at this point to take a ban on export off the table for the US. That would that would drop this massively right away, but I think this is set to uh increase uh going into the week. Uh but you got midterms coming up and they're going to have to do everything they can. everything they can. Um, being that the elections belong to the states, it's going to be hard to just cancel the midterms because you're in a war. That probably isn't going to happen. So, you're going to have to try to rig the election before it happens. Um, voter suppression, gerrymandering, things like that. We're already seeing that going on. Uh, but also, uh, trying to please the people, uh, so that you can get their vote. this is the quickest way to get energy costs uh to crater is to put an export ban on US oil. I'm not I'm not willing to take that off the table yet. He's telling other countries come and buy our oil. He's saying empty ships are heading to the US, but uh in the coal light of reason, I think it's easier just to pay a dollar a barrel uh and and have a shorter trip uh uh a 28-day trip instead of a 45day trip and and pay maybe a premium because it is light sweet crude and perhaps your refineries aren't set up to refine the American oil. So, I don't know if it's that easy. Um wheat and corn uh down WASDY report we had um um uh planting uh planting expectations and the WASD report. So that was negative for wheat and corn soy holding up uh and hogs and cattle more expensive. So from the beginning of the year food more expensive, oil more expensive and money seems to prefer commodities as opposed to US uh US treasuries. uh mortgage rats. Last week I said I would have a hard time uh not buying them uh given the yields they have. Uh they performed well this week in line uh with uh with how SPY performed. So I'm not going to say they outperformed, but uh I still would have a hard time not picking them up just given their uh given their yields. 6.37 on the mortgage rates still ugly. Uh Monday you get existing home sales. Wednesday, National Association of Homebuilders Index. It has been under 50 for a long time. Housing is down for the count. I don't know that it can get much lower. Uh reverse uh repo 507 million repos nothing. Uh reserves uh big increase 3.183 but the TGA had a big decrease. The TGA is not counted as part of reserves. Uh but when um uh expenses are paid from the Treasury general account, it goes into other bank accounts and as soon as it hits other bank accounts, it is now part of reserves. SOFUR, this is interesting. 3.57. Effective federal funds rate is 3.64. So far down to 3.57. I'm going to say that's probably a result of going into a weekend with negotiations. Uh maybe there was uh you know some some uh um uh demand in the market uh or lack of demand in the market. I don't know. But 3.57 I would expect that to move back up to the 3.65 uh for the week. But what a move on so far. Uh we have some Fed speak this week. Uh I hesitate to put Mirren down because he gets no respect. He doesn't move the market a bit and uh no one on that committee respects him because they look at him as Trump's mouthpiece. So we can just cross him out. I don't know he's worth listening to. Tuesday, Baron Collins, Wednesday, Bowman. Thursday, Williams. Uh Friday, we have Waller. Okay, I used perplexity for this. What is the current state of hearings to confirm Worsh for Fed chair? I am of the understanding that the Senate hearings were scheduled to begin April 13th. Your understanding is slightly off on the date hearings are not starting the week of April 13th. The tensitive midappril plan has been pushed back. What was originally expected late March? multiple reports said the Senate Banking Committee was planning to hold Kevin Worsher's Fed chair confirmation hearing as soon as the week of April 13th, 2026. CNBC and other outlets then reported a specific target date of April 16th for the hearing, though it was never formally posted on the committee's public schedule. Current status as of April 9th. as of April 9th and 10th. Political and financial outlet reports that the banking committee is no longer planning to hold the hearing next week. I and that the nomination hearing has been postponed slashcancled for that midApril slot. This is where my uh excitement started to drop considerably. I'm like, oh no. Because my whole ZQ trade is is premised on Worsh getting to speak and the tone of the conversation shifting from the interest rate to the balance sheet. If that doesn't happen, I'm in the wrong trade. Reports point to incomplete paperwork and disclosure forms. That does not sound right to me at this level to have incomplete paperwork and disclosure forms. Although it is this administration that just can't execute on anything rookie hour. Uh forms from wars as a key procedural reason. Committee rules require receiving the full packet and giving at least about 5 days public notice before a hearing can be held. It seems to me he would have had that packet. I mean, once he was confirmed a couple of months or once he was selected that that this would have been happening in the background that this transition wouldn't have been happening. It's not like he got a phone call on Monday saying, "Hey, can you be ready in two weeks?" Uh, yeah, let me pull some things together. That doesn't happen. So, h on this one here, I'm not on board on that. Some coverage notes that given those rules and the normal Tuesday, Thursday hearing pattern, the earliest practical new date would be around April 21st. That's when I said, "Oh, okay. Well, that's not so bad." But no new hearing date has been officially announced yet. Political backdrop, visible resistance on the committee. Tom Tillis publicly said he will not support Worsh's confirmation until concerns over Fed independence and ongoing investigations are addressed. We haven't heard anything on this yet. So you can think about it in terms of a scale, right? You have a scale on both sides where you have Worsh and lower rates over here and you have Trump's need for revenge over here. That's a tough one. That is a tough one. Um, I don't know which way the scales tip on this one, but uh I seem to think Trump's need for revenge is greater uh than the needs of anyone else. So, I don't feel good about this. But as if if if we get an announcement uh from Trump's America that um oh no, we're going after Powell. We're not letting this go. I'm out of the ZQ trade. Just to let you know straight up because uh it will be a Powell Fed for the next year. Uh and and the trade is over at that point. Uh it was premised on Worsh. If it is a Powell Fed, then I have to base it on the fundamentals that will occur and the willingness of this committee to accept those fundamentals and abandon the air quotes here abundance of caution. In other words, you keep paying higher interest rates because we want an abundance of caution. The White House continues to signal confidence that Wars can be confirmed and installed before Powell's term expires on May 15th. Confidence that he could solve the Russia Ukraine crisis on day one. Confidence that prices would come straight down. Confidence that Iran would fall within a few days. confidence that he'll get a deal and the price of oil will plummet. Do you mean do you mean that kind of confidence? Because you know I think you might be using that word wrong but the delay tightens the timetable increases the risk of slippage or acting chair scenarios. So hearings were aimed at midappril but as of now they have been delayed. No firm public date set. April 13th is not correct anymore. quote the raven anymore. Okay, let's have a look at copper. Interesting things happening here. The backdrop is um at least three, four, five years of structural uh deficits in terms of supply not being able to meet demand. that seems to be well agreed upon uh on almost uh every investment house that's looking at copper and forecasting where copper is going to go including uh industry sources uh industry association sources also predicting the same thing. However, we do have a near-term problem is that we have inventories that have built up because of industrial demand pullback over the last month and being that we have no deal and it's possible that these inventories go even higher. So, we have a short-term overhang of uh some inventory and while we have some supply coming offline in other places you have supply coming on. Uh so, this is a one-year chart of copper. There was uh this over here was uh the removal of the fear of 50% tariffs on copper coming into the country. Um IV 28.6 April 6th um section 232 tariffs, Trump tariffs. They're not US tariffs, they're Trump tariffs. 50% on finished product. Uh and uh 25% on derivative product. uh this is finished uh not copper not not the copper itself but if you're bringing in a product that is primarily copper based or made from copper but in final form 50% tariff if it's a derivative uh 25% but on the copper itself not really. So there is a a rush to move copper into the country before uh this administration decides hey you know what's a good idea let's let's tariff copper move your copper mines to the US he says you can hear him saying it uh inventories have a look at where they are London Metal Exchange 392,000 highest since 2013 uh that was when the cash price was 7200 cash price uh right now uh is 12 let me see if I can get something to write on this uh 12,660. The LME 3-month copper is 12,757. So, it is above the spot. So, that is a uh a telling sign of contango. Um it uh increased inventories increased from April 7th to April 10th uh from uh 12346 to 12757 while the price increases as well. Sort of the wrong direction. Inventory is increasing, price increasing. uh I would expect this to reverse uh and that uh being that we don't have a resolution in Iran that this the price the cash price will come down. Looking at Comx. By the way, if you want these charts, go to West Metal 2L's.com. This one is from ComX. Uh total copper. This is the previous week. This is how much came in. This is how much went out. And this is uh the inventories you have left. 586,000 57. That's of April 9th. This is April 10th. and um Shanghai uh 266,484. You can get that right from uh their uh site as well. The highest ever 433 was on March 13th. This um uh the CSPT China smelter purchasing team uh offered some support. They met back in December and you can see the rise in copper prices at that time. uh they had plans or they announced plans at that time to cut primary smelting in 2026 by 10%. Uh which means 10% less output. It also means 10% less demand for copper concentrate because they bring in co copper concentrate and they put out copper anode. Uh and it was believed that this could remove about 1 million tons of copper concentrate demand which would also remove the downstream product. uh they were citing record low uh um uh TC uh RC costs. This is a treatment costs for uh you give me your copper uh concentrate and I will turn it into something else and I will give you money back for what I get out of it, but less of course the treatment cost. Well, treatment costs were $0 a ton, free uh making their money off of the byproducts that you get from smelting. Lowest ever. Uh but this 10% is not being upheld. Recent uh u output from uh Jiang Xi copper and Yunan copper um showed that they are raising their guidance uh for um uh copper cathode raising their guidance which means they're not cutting back 10% they're actually increasing. So uh this boost in uh prices you got because of that uh quickly came out. The other thing that we cannot ignore is uh you know from a technical perspective you got a head and shoulders and you've had your pullback here. Usually you can get some kind of return back to the neckline but if uh what uh caused the last shoulder if the fundamentals still exist down we go. We've had this little peak over. I think that was a little bit of short covering. I think we're under pressure for a while in copper. The long-term story I think remains in place. However, we have this big inventory hangover that is a huge inventory hangover. And if this uh fiasco is not done, if both sides after a good night's sleep don't say, you know, maybe maybe we can dance, maybe there's something we can do. If they don't give and the bombs start flying Sunday afternoon and Sunday night into Monday and the threats of war crimes and genocide start again, then industrial demand will just keep pulling back, saying, "No, no, let's let's hold off for a while. Those inventories are set to go higher. Copper probably will uh make its way back down to 540, 530, 520, which to be to be clear, I think is a good entry point right now. What I've done, uh, I've, uh, got puts anywhere from $6 all the way down. I got $6 puts. They expire at the end of this month. I got $6 puts, which means I'd have to buy it at six bucks. However, I got, I think, like 3419 for that. I've got 580 puts and 560 puts. As Copper was rallying up here, uh, especially on Friday, I closed all my long positions, but I was selling calls, uh, at the same time. So, I'm basically balanced. I've basically got almost zero copper exposure because of the calls and the puts uh together. It must stay in the range of $6 to 550. I make the maximum amount on those contracts and I think I have 25 20 25 or 26 puts against slightly fewer calls 23 or 24 calls to to keep that zero balance. The theta on this is huge especially with I think uh uh what do we got less than 15 15 days 14 days left uh to expire below six but above 550. I'm not worried about the $6 price. I think it stays below six. Uh if anything it probably my my biggest risk is it drops below 550. If it does that's fine. my average price will be very low when I get the uh I think the June underlying contract or is it the May underlying contract? I think I have the June underlying on this one. Uh but that's uh that's where uh we are with copper. Okay, I'm going to show you something uh that I have Claude Co do starting to use AI agents a bit more. It was bothering me that I couldn't remember my copper trade. So, I pulled up the account here. Here's exactly what I have uh on the calls. 11 uh calls at 550 uh that I got 19.72 for. Um I have four more at 560 of which I got uh.1899 or 19 cents for. And then I have uh six at the 571s uh for 17 cents. Uh so that gives me 21 on the puts. I got 11 at $6 that I got 3234. I've got five at the 580, but I got 3149 for these. Nice premium on that. And I've got eight at the 560, of which I got. 2246. If we are between 560 and 580, I make the maximum premium on this, which is about 112,000. So on the puts, there's 11 and 5 uh 16 uh and a 24 puts against 21 calls. That was the ratio. Uh and it is the May underlying. So I cannot wait until expiration. uh about two to three days before expiration. I will have to roll these forward uh because uh they're options on futures. You'll get the futures delivered to you. Nothing wrong with that except the futures expire that very day. So in essence, they are cash settled at that point because Interactive Brokers doesn't allow physical delivery. You're going to be cash settled on the option because you get the future and half a second later the futures expire. So uh if you do have options on futures, be aware that when the options expire, if they expire on the day that the futures expire, they are in essence cash settled at that point. So I probably will roll them forward uh before that point. But the theta is so big right now. I keep looking at it going, "Ah, if I could just hold on one more day. One more day. Come on. Let's see what happens this week." Um, so I uh use Claude Co-work uh to uh or I've been experimenting with it as far as its agent capability. It can interact with your email, but it only creates a draft and it drops it in your inbox. And I wanted to see exactly how much usage you get. So I looked at my usage before I run this and I look at my usage after. Uh agents be prepared. Agents use a lot more than a chat. uh a lot more. So, uh just on the rough calculation, if I were paying for tokens on this thing, uh just this one, and I do it every day, I have it send this to me every single day. Uh and I can see that I can increase my parameters on this to eliminate certain things. Uh, and this is basically the best way to do it is you just start, you look at what you get and then you say, "Okay, well, you're giving me this and and you know, you already told me about this the day before. Let me let me put some more parameters in there." If I were going to pay just tokens in, tokens out, roughly, it's about 30 bucks a year. It would be about $30 a year just based on I'm eyeballing the amount of usage when I run this and just this. I'm just eyeballing the usage uh on this. So it I did the calculation for the number of days. It works out to about 20 cents uh a run. 20 cents to get this done. And I get so much information out of this. I think to myself, would I pay 20 cents for this? And the answer is yes, I would. I would pay 20 cents. I have this done every single day, the copper supply and demand briefing. And my prompt is set up such that I'm looking at macrolevel themes. I don't want to hear about one company in the copper mining industry that reported earnings that had this. That doesn't that doesn't change what I I want to do because I'm making a bet on copper futures on the price of copper itself. I want the supply and demand stories. Um so it gives me that and every day 10 10 stories. But sometimes it repeats the story like the Ivanho cuts here. I I already had that earlier in the week. So I know I have to go in and adjust my prompt to say if only update stories, don't repeat stories or something along those lines. Ivanho had cut uh their guidance by about 20% uh from uh DRC uh their copper. uh they now see uh 2026 uh output 290,000 to 330,000 tons down from 380 uh to 420 and 2027 380 to 420 down from 500 to 540. So there's a supply issue that is supportive of the price of copper and uh gives you the uh the source so you can actually uh Ian homemine press release and mining.com Yahoo Finance so you can actually fact check it very quickly. Uh chili copper output hits a 9-year low in February. Structural ore grade decline takes center stage, but we've known that for some time that the grades uh per ton of copper uh per ton of ore, the grade uh the what it's grading has been decreasing over time. That is a supply story that adds to the price that adds a support to the price. They say US section 232 copper tariffs uh full customs uh value now applies as of April 6. And it gives you a a little paragraph here. It gives you the sources for this as well. Goldman Sachs trims the 2026 base case copper target to 12650 lifts projected surplus to 490. That may seem odd. They're lifting uh uh the uh sorry not odd. Uh earlier in the week they they they were forecasting a deficit uh and trimming their price and their argument was that you're going to get a supply overhang but it will disappear. Now they're projecting a surplus and it all is all really based on what is going on uh in um in Middle East and industrial demand for copper uh being put on hold. 2026 copper concentrate uh TRRC benchmark settles at $0 per ton, the lowest on record as smelter margins collapse. Um they're making it on the byproducts, but they're still out outputting. This is what we talked about. about the agreed to 10% 2026 output cut, but Q1 smelter results show little evidence of follow through. This is where I got the uh evidence from the uh raising of the guidance from two uh Chinese smelters. Copper inventories remain remain near multi-year highs. That's for uh London Metal Exchange and Shanghai. China's non-ferris metals association calls for expanded strategic copper stockpiling. Oo, China wants to stockpile copper. There's another demand uh side. So we the the supply side is supportive of price. Uh the inventory overhang is hanging on to the price right now and keeping the price down. But if that is what's keeping the price down, we expect over the next few years prices going up. That's not a bad idea to begin stockpiling, right? So that may overcome just inventory stockpiling may overcome the overhang uh and what's going on in the Middle East right now that maybe the inventories are masking things that are going on under the surface. Uh AI data center and grid demand uh reinforce the structural deficit thesis. Uh Zambia's uh local sourcing law kicks in. Uh Zambia DRC royalty regime keeps African supply in the 10% bracket that holds down supply. So the theme you get out of this is that we have some real supply constraint issues and we have long-term structural demand building. So long-term I'm still bullish on copper. I think I think we will not only see the $6 uh level but the seven and the $8 level on copper but not now. Not with this inventory overhang and not with the political the geopolitical uncertainty uh right now. And this little thing ends with uh, you know, a nice little snapshot very quickly at the end. So, if you're using Claude Co-work, you're going to need Co-work, which means you're going to need at least the 5x or the 10X um um uh subscription. Uh, I have the 5X. And, uh, based on my usage, I can see that it won't be long before I have to go to the 10X, especially if I start putting more agents out there. This one here does does add because I do it daily. Maybe I don't need to do it daily. I don't know. Uh but if you are using it, be prepared for uh your usage. You need the bigger model to begin with. And your usage for agents uh is way more than your chat window. Okay, let's take uh some time and think about cyber security. They got beaten up last week um after Mythos announced with Project Last Wing that PaloAlto and cyber security were part of it that they were getting access to this first to harden up their code base. Um up they went in price and then they gave it right back the very next day because the market goes, "Hey, wait a minute here. We still don't believe this." And down they went. So I uh decided to make Claude my enemy and I said uh construct the strongest bare case that you possibly can. So I gave it its its scenario. Uh uh and I laid out what I thought. Uh and I say I may have a blind spot, but I think the cyber security selloff is massively wrong. And names like Service Now uh I because I'm talking about both SAS and cyber security. Names like Service Now will not just survive AI, but thrive and increase their value propositions. Uh SAS companies will not just be an IT line item, but they will begin to go after the labor budget of companies. I need you to push back hard. I need you to construct a bulletproof bare case for cyber security and SAS companies. Scare the hell out of me. In other words, give me the worst you possibly can. And you read the the points he gave. And if you'd stopped there, you'd say, "Get me out. Get me out. Get me out, man. This is Opus 4.6. It is good and it is almost bulletproof. Uh, and this is where the value of the human comes in is because you can see weaknesses in its argument. Uh, because it does argue linearly and so you can get weaknesses in there. You can get in there and and and mess around with it. But let's see if you can find the weaknesses as we go through the first one. The code moat is evaporating. You're right that incumbents have data and relationships, but you're underweaiting how dramatically AI compresses time to parody for code. What took PaloAlto or Crowdstrike years of engineering to build detection engines, correlation logic, response playbooks, a well-unded startup can now replicate in months with a small team and foundation models. The experienced delta you're counting on assume assumes the delta is hard to close. AI makes it trivially easy to close. On the software side, the question is whether the data modal load justifies current multiples. But for most SAS companies trading at 10 to 15 times revenue, the answer is probably no. The seatbased model is dying and nothing proven replaces it. Now, for each one of these, see if you can find the weakness because I found the weakness in each one uh and then challenged them. Uh nothing proven replaces it. Service Now charges per seat. So does Salesforce. So does Crowdstrike. When AI agents start doing the work of three to five analysts, those seats disappear. You say SAS will go after the labor budget, but that's exactly the problem. If now sells an AI agent that replaces four service desk employees, the customer saves 300K in labor, but now can only capture a fraction of that in pricing before the customer says, "I'll build it myself with an LLM and some glue code." I don't know what glue code is, but there we go. The pricing power shifts violently toward the buyer. Consumption-based pricing sounds like a solution, but it's inherently less predictable, lower margin, and the market will punish the transition with brutal, brutal multiple compression, just like it did to every company that tried shifting models midstream. Can you see the weakness? Cyber security specifically the attack service argument cuts both ways. Your bull case is AI creates more threats so you need more security. But AI also makes security cheaper to deliver. Automated SOC's, AIdriven threat hunting, autonomous remediation. These dramatically reduce the human capital needed and therefore the dollar spend. Do you see the immediate weakness in that one? The TAM doesn't grow the way you think. If the per unit cost of security collapses, you end up with more coverage but less revenue, which is a nightmare for incumbents priced for perpetual growth. Multi-tenant data mode is thinner than you think. You're assuming proprietary data is defensible. But consider most cyber security telemetry is structurally similar across enterprises. Endpoint logs, network flows, identity events. These aren't unique. A new entrant that onboards 50 mid-market companies has a data set that's 80% as useful as crowd strikes for training detection models. The last 20% has diminishing returns. Meanwhile, open- source threat intelligence uh feeds uh MITER, AT&T, and CK frameworks and shared IoC databases mean the knowledge layer is increasingly commoditized. The data advantage is real but narrower than the current premium suggests. How many people are thinking, "Oh boy, I better sell my cyber security and SAS uh stocks right away." There is a weakness in this one as well. A big glaring weakness in this one. Switching costs are eroding, not strengthening. You say, "No one switches mission critical systems." That was true when switching meant 18-month rip and replace projects. AI powered migration tools are collapsing that timeline. LLMs can map API schemas, translate configurations, and automate data migration at a pace that was unimaginable two years ago. I believe that. I do believe that the lock in that justified premium multiples is weakening precisely when these companies need it most. Glaring uh error, not error, but glaring uh um um opening in this one. The labor budget thesis is a trap. If SAS companies go after the labor budget, they're now competing against a $4 trillion global line item with ferocious procurement scrutiny. CFOs who demand hard ROI proof and critically they're selling automation where the buyer has every incentive to commoditize the tool. When you sell it software, the buyer doesn't know how to build it. When you sell labor replacement, the buyer's own engineering team starts asking, "Why can't we do this ourselves with Claude or GPT and save the vendor fee entirely? The internal build versus buy calculus shifts catastrophically towards build." You know, I told it to scare the hell out of me, but look at the words catastrophically, violently repricing. Uh, we're almost done. It did it like it just kept going and going. Margin compression is structural, not cyclical. AI inference costs money. You see the weakness there already. I see the weakness there already. Every SAS company bolting AI features onto their platform is eating into their own gross margins. Crowdstrike, now Salesforce, they're all spending heavily on compute to deliver AI features while facing pricing pressure from customers who say, "If your AI is doing the work, why am I paying the same per seat?" you get squeezed from both sides. Higher cogs, lower realized pricing. Gross margins go from 75 80% to 60 65. And at that point, the market rerates uh these as infrastructure companies, not software companies. That's a 40 to 50% haircut to the multiple. Do you see the weakness in that one? I mean, that one is glaring. And this is why AI is really good for helping you think out ideas, but certainly you wouldn't take this and say, "I better sell my stock." Terminal value scenario you should fear the most. Imagine it's 2030. Foundation model providers, Anthropic, Open AI, Google offer turnkey security agents, service management agents, and CRM agents as features of their platforms. not standalone products but capabilities baked into the AI layer that every enterprise already pays for. Service Now uh Service Now's workflow engine becomes a feature of the AI platform, not a product. CrowdStrike's detection logic becomes a module. The entire SAS layer gets absorbed into the foundation model layer the same way standalone utilities got absorbed into operating systems in the 1990s. The SAS companies become the Lotus 123 and Word Perfect of this era. You have to go back to the 90s to understand Lotus 123 and Word Perfect. Everyone used Word Perfect. Then Microsoft Word killed it. Everyone used Lotus 123, but Microsoft Excel killed it. Dominant right up until the platform swallowed them whole. your blind spot, your pattern matching to previous tech transitions where incumbents adapted. But previous transitions didn't simultaneously attack the delivery cost, the pricing model, the switching cost, and the competitive mode all at once. AI does. The bare case isn't that these companies go to zero. It's that they become mature, low growth, compressed margin utilities trading at four to five times revenue instead of 12 to 15x. That's a 60 to 70% draw down from here. Even if revenue stays flat and the market is starting to price that in, how many people are scared and how many people saw the flaws in the argument? Not so much the flaws, but the weaknesses. The place where you can get in there with a crowbar and say, "Whoa, hang on now. Now, I constructed that as ruthlessly as I could, but I'll note the strongest counterargument you already identified. The gap between technically possible and enterprise adopted is measured in years and trust and trust is the one thing AI can't shortcut. The bare case above assume assumes enterprises behave rationally and quickly. They almost never do. So, um, I had to push back on this and I did have, uh, some points of rebuttal as you can see here. I had a bunch of them. Let's see if you can find the weaknesses that I found. Number one, you state a startup can use AI to achieve parody. But that implicitly assumes the incumbent does not also use AI to maintain its delta. Going from zero to the incumbent only works if the incumbent stands still. More than likely all incumbents have a war room to determine how they are going to use AI from their starting point which is uh which is months and a good it away from even catching up on just the code. The market for their product must still accept them. This these are the new ones I want you to think about. You've been with your provider of cyber security for 10 years and a new startup comes along and says why don't you just switch to us we have 80% of the data set and you've never heard of them before right the market for their product must still accept them you state the seatbased model is dying but then state that LLMs may be able to offer a layer to replace some services like cyber security however LLMs are subscriptionbased and if using an API tokens in tokens outpriced You state that nothing proven replaces seatbased, but LLM's subscription and token fees have proven it. Switching from a known seat cost to an unknown API cost seems to defeat your argument. Did anybody see that when when it was challenged? You assume that if an LLM can offer security, organizations will lean in that direction. that assumes a cost first analysis versus a risk first analysis. One attack, one event of a customer data leaking, one ransomware event from a company that that left its provider to go alone with an LLM and a build it myself attitude and a team of lawyers will rip that company apart. Incumbents using AI can reduce their headcount to maintain margins as a corollery to your argument that a lean startup can replicate their code fast. It did say a lean startup with a smaller team can get there faster. Well, these large organizations, if they can use AI to show their customers how to reduce headcount, they can also reduce their headcount. 80% of a data set is not 100%, much like 80% full self-driving is not 100%. Accidents happen in the 20%. Breaches happen in the 20%. Attackers don't win with brute force attacks on the strongest part of the fort. They look for vulnerabilities in the wings of the defense. They attack in the 1% or the 0.1% not in the 80%. Further 80% is a moving target. What is 80% today is 79% in an hour given the speed at which bad actors can develop just malware alone. As an analogy, if you have two robo taxis, one with 80% of the data set of the other, how much of a discount would you need to take the robo taxi that will self-drive on a smaller data set at dusk, in the rain, with a bit of fog, or even on a clear day if you are willing to assume that the route will not have a tail case. The 20% may have low probability events, but the cost of each event is unforgiving. This is why the 80% argument, I don't like that. Yeah, you can get to 80% quickly, but 80% is not 100%. Switching costs, not just technical, they are behavioral. Employee resistance, inertia, good enough is good enough. People in organizations, as a result, really only switch when the case is compelling, not incremental. Incumbents are not sitting still. They are moving targets. So getting to a compelling case against an incumbent that is using the same tools you are uh is a fastmoving target, a fastforward moving target. I'm almost done here. Um LLMs have gone after my labor budget with ease, but I'm not about to start to build verticals. I'm not getting rid of HubSpot just because an LLM can code workflows. for the 50 bucks a month I spend. I'm not about to dedicate time to replacing that. It is my belief that HubSpot will use AI to develop a value proposition faster than I can use an LLM to reduce the value proposition of their service. I think it would be the same way for organizations. Incumbents will not stand in one place. They run and AI will allow them to run faster. A heightened red queen effect. uh you state that AI inference costs money, but that is true of organizations that would contemplate eliminating the incumbent services. They would then be the ones incurring inference. You also state that have now develops an AI agent that allows its customers to reduce headcount by four and save 300K, now would only be capturing a small portion of that. But that is exactly what seat subscription was doing. assuming 2,000 a seat per year if now can reduce the customer's labor cost by 300,000 and the AI agent may be worth uh more than $8,000 a year in saved seat revenue. What if I could reduce your labor cost by 300K and your seat cost by 8K and that AI agent would only cost 20K. The customer saves a net 288,000 a year and now raises revenue from 8K to 20K a year. I do not see the labor budget as a trap. if I see it as a gateway to higher revenue. And finally, by assuming organizations will engage in a buyer build analysis due to AI assumes two things. First, that the organization has both the bandwidth and the skill to in-house the build and to maintain the build over time. And second, that the organization won't dedicate its IT resources to doing more of what it already does. Every IT team in every organization has a need to do and a nice to-do list. AI will allow them to hit their niceto-do list. This is also true of the incumbents. Customers pay for the need to-do list from the incumbent. Imagine what the niceto-do list will do for the value of the service they receive. Now, at this point, after having gone through uh the first part uh its response, you might have thought, "Holy they're right." But in reading the push back, did you see all the weaknesses in the argument? Here are all the weaknesses. A few points, right? Uh what do you think now before we look at its reply? What do you think now? It's programmed, not programmed, but directed to push back. You made me work for it. Those are strong rebuttals, but let me go after the cracks on the incumbent moves to argument. You're assuming incumbents deploy AI efficiently. History says otherwise. Large organizations are structurally terrible at self-disruption. I can't disagree with that. The innovator's dilemma isn't just a book title. It's an observable law. Crowd Strike is 8,000 employees, legacy architecture decisions baked into their platform, and a board that will not tolerate a quarter of margin destruction to rebuild from scratch. The startup has no legacy, no technical debt, no installed base to maintain backward compatibility with. The incumbent's war room is full of people protecting the existing revenue stream, not cannibalizing it. Yes, they'll use AI, but they'll use it incrementally, bolting features onto existing architecture while the startup builds natively. That's exactly how PaloAlto displays legacy firewall vendors, how Crowdstrike displays semantic, how Snowflake displays terod data. Every single time the incumbent was also using the new technology, they still lost share because architectural advantage beats feature edition. Okay, that's powerful that that that the architecture that you're building on and I do know one company that is in this situation where they're trying to adapt to a a world of AI and they're using their existing architecture to do it. And I pointed out saying, you know, it's it's the layer underneath that has to change first. You can't just put these things on top. You must build from the bottom up an AI experience on your seat to agent pricing math. Your now example is the optimistic scenario and even it contains the bare case. That got my attention. You said now goes from 8 to 20k per customer, but you eliminated four seats, four humans to get there. Those humans weren't just using now. They were also using Slack, Salesforce, Zoom, Octa, Crowdstrike, Endpoint Licenses, and dozen other perceived tools. You just destroyed demand across the entire SAS ecosystem to give now a modest revenue bump. Uh, I didn't think about that. That's a powerful push back is that it's not just Service Now that's going to replace seats. Other companies are going to replace seats that maybe I don't want replaced. So, while I've eliminated four seats, I've eliminated four seats for a whole bunch of other companies. They're also eliminating seats. They'll eliminate seats for a whole bunch of other companies as well. I got to give it to uh Claude on this one. You just destroyed demand across the entire SAS ecosystem to give now a modest revenue bump. Second order effects are devastating. And here's the deeper problem. Your 20k number assumes now has pricing power. But if Anthropic or Microsoft offers a comparable agent for 5K bundled into an existing platform agreement, now has zero leverage. The customer doesn't care who provides the agent. They care about the 300K in savings. Now's brand means nothing to an AI agent the way it means something to a human who learned the Now interface over years. Agent loyalty is zero. on your risk versus cost argument. This is the strongest point and I'll concede it holds for two to three years. Thank you for the window, but play it forward right now. We left our provider and built it ourselves is reckless. In 5 years, when AI native security is battle tested across thousands of deployments, the legal calculus flips. The plaintiff's lawyer will instead ask, "You were paying 10x market rate for legal cyber security that missed this breach when cheaper, more capable AI native alternatives existed. Isn't that a failure of fiduciary duty?" The risk argument protects incumbents temporarily, but it's a melting ice cube, not a permanent moat. Every year that AI native solutions operate without catastrophic fault, the risk premium erodess. On the 8020 data argument, your robo tax analogy is vivid but misleading. How dare you, sir. In cyber security, the 20% gap isn't static. It's constantly refreshed by new threat data that both players collect going forward. The incumbent's historical advantage is the 20% depreciates rapidly because yesterday's zero day is today's known signature. That's a good point. The attacker who exploited a novel technique 6 months ago is now in every threat feed. The startup that launched 6 months ago has captured essentially the same forward-looking threat landscape your robo taxi analogy would only hold if the roads never changed. Cyber terrain changes hourly. Incumbent's historical data set is like last year's map. Useful but dangerously incomplete on its own. Both players are equally dependent on today's telemetry. I see a major weakness in that argument. But I didn't push back on. I said, "Okay, but I can I can easily beat this argument." Switching cost and behavioral inertia. Actually, we don't have to read the rest. Let's go right to the uh the Red Queen effect. That's an interesting one here. On incumbents running faster, the red queen effect has a crawlery you're ignoring. In biology, the red queen doesn't guarantee survival. It guarantees exhaustion. If incumbents must perpetually reinvest all AIdriven efficiency gains back into the product just to maintain competitive parody, margins never expand, the market currently prices these companies for margin expansion and revenue growth. If AI forces them into a permanent treadmill where every gain is competed away, you get revenue growth with flat margins and the market rerates them violently. The Red Queen doesn't produce winners, she produces survivors running in place. Uh, and then the the Meta Bear case you haven't addressed. Uh, I'm going to let this one go. And the very end, I'll be honest with you, you've weakened my bear case considerably. Your strongest arguments are three risk over cost, five, the 20% that matters, and eight the labor budget as revenue expansion. If you're right that incumbents can price AI agents as a fraction of labor savings rather than a markup on software seats, the revenue opportunity generally does expand. The bare case depends heavily on platform bundlers like Microsoft commoditizing that value before the specialists can capture it. That's the real race to watch. Now with its with its rebuttal to my rebuttal, I could have gone another round. I'm sure that if I kept going, I could have beaten the AI. So if you feel AI is just going to replace everything, it is very very good at bringing you up to speed very fast that if you were sitting in a room with the investment committee saying, "Okay, is this is this the right move or the wrong move? Let's throw ideas around. Nothing is off the table. Let's let's get let's put it up on the whiteboard and let's see what we got. It is very good at throwing things up on the whiteboard. But you still must be an expert in this so that you can see the weakness in the argument. And the LLMs do not construct bulletproof arguments. It's there's usually it's they're very compelling. Uh but because it's linear, I don't really see the metaccognitive thinking in here. I see that it has very good cognitive capabilities at the lower levels but at the metacognitive level it can't say but on the other hand there's this I've left this opening and then construct an argument to close all those openings. Uh so it leaves enough openings that you can uh construct a counterargument which I have done in each and every time and it has tried its best to hit back but on each one of these I can also uh open them up a little bit more. So there we are. Now apparently I can upload this now. We will find out when I upload the video if I can upload it. If I can, I will put it there and you can just cut and paste the prompts into whatever LLM you want and see uh what answer you get. And then you can continue on prompting it and see where you go from there. Okay. SPY forward multiple 20.09 up from 19.45. uh forward four quarter operating earnings expanded week overw week. I uh I don't know uh if I agree with that. And closing S&P 6816 gives us 20 times forward earnings and volatility has basically just drifted away. IV last 15.4% from over 25% at one point last week. Uh and we are uh in spinning distance of all-time highs. Look at this. uh the only pattern that matches for all those people who spent two three years getting their CMT that's the chartered market technician that's where you learn to draw lines on charts um it must be humbling after spending all that time to know that there is only one pattern uh that matters in markets there is only one pattern and you see it everywhere everywhere everywhere look at this it's everywhere by the V. It's the V. The only pattern that matt that matters is the V. So for those of you thinking about a chartered market technician, uh, what is the 23rd letter of the alphabet? That is your question. If you say V, there you go. You pass. You can start using the letters CMT. Congratulations. It is the only pattern that matters anymore. Unbelievable. But here we are. Uh we'll see what happens uh Monday when uh the American market uh opens. If uh if the US uh market participants want to continue to ignore uh what's there and say uh you know uh buy by buy the dip buy the dip or if uh we get uh another V pattern uh spy has recovered the 50 the 100 and the 200 day moving average. However, the 50 is below the 100 right now. We'd need that to get above to uh sort of hit a trifecta there. 2 and 12% below its all-time high. IWM has uh recovered the 50, the 100, and the 200. The 50 is above the 100. The 100's above the 200. Nothing to see here, folks. Down 3.24% from its all-time high. Triple Q 611 down 3.72% off its all-time high. uh it has recovered the 50, the 100 and the 200 but like spy the 50 is below 100. So we don't have a clean bill of health across all of these. Um I don't I I I am not of the camp that thinks we're going on to all-time highs and then uh throughout the year printing 52- week highs going forward. you don't have a resolution uh to the Iran issue uh which means you don't have uh a resolution to the energy issue. We still have no resolution and no real uh insight into the private credit issue. There is still the AI uh rerating of anything that has code within it. There's still the idea of well how much of the labor market is AI and is that going to accelerate as these tools get better? We just keep layering on one thing after another after another. We never really solve them. But the V is so powerful because for the last five, six years, it has worked. There is a whole generation of financial market participants that only know the V, buy the dip, buy the dip, buy the dip. That's going to be a hard habit to break and that provides a floor uh on pricing uh because uh all too eager to buy that dip. Uh we have 28 companies reporting next week uh and I am relying of course on LSG data uh mostly banks but we do have Netflix in there. So let's take a look at uh who we have reporting for the week and uh we'll wrap it up. So Monday, uh, before the bell, Fast and All and Goldman Sachs going into Tuesday, Wells Fargo, Kinder Morgan, Black Rockck, Cityroup, Johnson and Johnson, and the Big Boy on the Street, JP Morgan, uh, all before the opening bell. So if you're playing these, keep in mind you're going to have to do something on Monday, uh, before the close. Going into Wednesday, uh, only one after the bell, JB Hunt before the bell, Bank of America, Morgan Stanley, Mnt, PNC Financial, Charles Schwab, all the financials. Uh, going into Thursday, we start to get a little uh, more uh, spread out here. PepsiCo, Bank of New York, Abbott Lab, Citizens, Keycorp, Marshia, Mlullenum, Prolis, uh, Travelers, US Bank, and there is Netflix. I am simply just taking uh, uh, taking their word for it. Thursday, April 16th. I am not fact-checking any of these, but there's Netflix. Their guidance will probably be very good. Keep in mind, they already announced the price increases. Significant as a percentage of what they charge, but insignificant in the aggregate as a total amount. So, I don't know that uh that uh subscriber losses will outweigh the increase in revenue that they'll get from that. So, their guidance should be very nice. And going into Friday, all before the uh opening, State Street, Fifth Third Bank, regionals, and uh truest uh once you get the financials out of the way, the week after usually follows with the regionals and a lot of the tech companies for the first two weeks, they tend to be front end loaded. So, buckle up. Here we go.