Ellvate

@optionflys Looks like it aligns x.com/theshortbear/s…

Something interesting caught my attention today. $SPX spent roughly: • 48 days accepting 6000. • 50 days accepting 6900. • It is currently accepting 7500. If this rhythm continues, it would reach roughly 49–50 days around August 11. Give or take a day here and there, the timing has been remarkably consistent. Markets don't only correct through price. They also correct through time. Before every major move, the market first has to accept a new value area. I'm not predicting what happens next. I'm measuring the process. Time is just as important as price. $SPY $QQQ @Optuma

Let go…crying feels good 🧘🏻‍♂️✌️❤️

@RealSimpleAriel @Qullamaggie ⚖️

The people who feel the urge to do something every single day are the same ones who give back a months worth of progress during a pullback/consolidation. "Everyday is a day to lose money, but not everyday is a day to make money" @Qullamaggie

@TheOneLanceB My ego feeds off of it and noticed I hold losers longer…I no longer share PL

END OF THE MONTH REMINDER FOR TRADERS! Every single trader you see on here posting their big months and big wins is subject to selection bias. Most of the traders who underperformed or lost money are invisible. Much of the PNL that gets posted comes from a position of insecurity, seeking the affirmation of others. For most of them, you do not see their losses or their struggles. You do not see their PNL on the bad months. I know from behind the curtain, many of even the most respected traders fall victim to this. This dynamic is part of the toxicity of social media and the comparison game. This reminder is to focus on your own game. Become your best self. And that there is a whole lot more to life when you step away from the screens. PNL should not define you. Get outdoors, spend time with others, and give back. Remember how lucky you are to be pursuing an incredible job, but also remember that once you step off the field, this game doesn’t matter at all.

🤞✌️❤️ $SPY hope the 🤖 is right. x.com/i/grok/share/c…

@TradeDadLog Normal trading journey…long as you make correction you will be fine.

Right now I feel like shit. During Asia I made $4k across 5 accounts ($20k total). I only needed 4 more minimum days to request $2k from each account ($10k total payout). I was genuinely happy. I even told my wife I was done for the day, that I wasn’t going to touch the charts anymore and that I was just going to secure the payout with the minimum profit per day. A few hours later I opened the chart again… and blew all 5 accounts. I knew there was no reason to keep trading But I still chose to open another trade. No one forced me. No one put a gun to my head. I did it to myself. I’m tired of repeating the same mistake. I’m tired of having the discipline for a few hours and then losing it completely. This one hurts more than usual because I was so close… and I threw it away with my own hands. I should’ve just blocked the accounts like I said I would.🤦‍♂️🤡

@TheShortBear Sold rest of my $CLSK this morning

Rotation out of AI today.

Reset and I will double bottom ⚔️

7k > +34k in 5 weeks two bad days 💣 30k > +14k 7 days one bad trade 💣 Back to paper…so hard and frustrating #spy

I am so grateful to be a father to three amazing 🦍🦍🦍✌️❤️

@madaznfootballr I had same issue with TOS, contacted trade desk and asked them to erase fees for the day but they offered $75 credit 🤬 sorry about your loss ✌️❤️

P/L: -$163.5K😰 BLEW THE FUCK UP! New 2nd Worst day of the year and a 3rd Blow up in 3 weeks! Absolutely disgraceful and unacceptable performance. Started the day with a blow up on $CLWT. Didn't realize the charts on TOS were wrong until it was too late and even then DAS, Finviz, Tradingview all showed different levels which furthered the confusion. Correct move was to not even take the trade without knowing which was the correct chart. This admittedly tilted me. Got losses down to just -7K from -20K but then completely lost control and blew up similarly on $ICCM short then flipped long and blew up there too. Then tried to short the pops after the open repeatedly and blew up even more stopping out at the top multiple times. Never really had a good read on $ICCM at all as the price action was just hard to read even after the offering it held up stronger than expected. Overall poor emotional control and gotta get better at staying calm after those type losses due to technicalities. $ATPC $BYAH $ELTX $EZGO $LNKS $LPA $NXTS $UTSI

@TheOneLanceB @wguidara I will take one 😁

One of my favorite books ever is Unreasonable Hospitality by @wguidara. Being featured in Market Wizards has been a special moment for me and I’ve wanted all the important people on my journey to share in it with me. I’m in the midst of signing and mailing across the globe over 250+ copies for mentors, friends, and true fans. You are nothing without the friends and supporters around you. Thank you to all of my followers who take time to listen to what I have to share 🙏🌎

@TheOneLanceB @jackschwager You are inspiration for me to work even harder ✌️❤️

To persevere at any difficult endeavor, like trading, I always talk about how one needs a powerful “WHY”. Love these comments from @jackschwager about how your “WHY” for doing something can be a drive to or drive away from something. What is YOUR “WHY”?

@paxtrader777 Couldn’t hold very long enough but it was enough 🫠

What a day. I am exhausted!!! I am too old for this shit.

@madaznfootballr Sorry that sucks…you will build again ✌️❤️

P/L: -$114.4K😰 BLEW THE FUCK UP! 2nd worst loss of the year. This one really hurts and is absolutely soul crushing. Basically the reverse of yesterday as shorts got fucked on $INHD halts. Today I got fucked on downside halts with longs. As you can see I'd be up pretty damn good on the day but got stuck in not one but two longs that halted down and gapped down nasty including $MTEN which gapped down like -50% while I was in long. $PAVS gap down was also 5 pts or something as well. Have some serious PTSD from halts now. This one will be difficult to recover from especially mentally especially since I was doing so well for awhile now. $AZI $AHMA $CCTG $CHAI $ELPW $EPSM $GMEX $HAI $JZXN $MSW $UK $YOUL

Playing out

@steveruffley 50% fib on $spy? 😁

@SteveUrkelDude 😆😂

Michael Saylor and Tom Lee spotted in LA this morning, after #Bitcoin & #Ethereum make new cycle lows. $MSTR $BMNR

Welp all the $keel feed has stopped…glad I sold week ago, looking at $3.50 next level.

@hitraderjoe I sold all my 12k shares last Friday, my understanding was big hype for 2 or 3 contract while they already have 12 facilities

As $KEEL crosses $6 this morning here was my original analysis posted on my newsletter. It's become more popular on social media so that is the main update Who is Keel Infrastructure? Earlier this year, Bitfarms rebranded to Keel Infrastructure. The ticker also changed from $BITF to $KEEL on April 6th. Bitfarms was a Canadian-founded Bitcoin mining company established in 2017 that had mining facilities across North America and Latin America. Over the past year they’ve made a number of changes: - Strategic shift to AI infrastructure - Relocated corporate headquarters from Canada to the U.S. - Sold its Latin American mining assets The Bitcoin mining to AI infrastructure play isn’t new. We’ve seen it with Iren, TeraWulf, Cipher Digital, and more. They’ve been on the later side to adopt the trend, only announcing the pivot last year, with further pivot headwinds given their more involved restructuring process. It’s important to specify what Keel is trying to be and what it isn’t. Keel is not the next $IREN. But it’s not trying to be. Keel is aiming to be a “pure-play infrastructure developer and owner”. “[Keel’s] entire focus is on providing the energy-secured sites and facilities that enable [their] customers to deploy AI compute at the pace and scale they need.” $IREN is a hybrid operator. It owns and rents out its own GPUs (like $NBIS, a full-stack neocloud) and also leases raw data center capacity to tenants who bring their own GPUs (like Keel). Keel is pivoting to a business model currently already being run by other major players like Applied Digital ($APLD), Hut 8 ($HUT), Cipher Digital ($CIFR), and Terawulf ($WULF). Power Remains the Constraint, Political Risks Are Rising The move to AI infrastructure is easier said than done. It’s widely known now that power is a leading constraint with data centers getting operational. (For a quick video primer on AI-related power here’s a short IG reel that I made last year: Power is AI Gold.) For those new to looking at this space, Bitcoin mining operations have been well positioned to become AI data centers as they already have access to a large amount of power. Figures from Investor Presentation. Panther Creek (Pennsylvania) - flagship Secured Power: 350MW (potential expansion to 510MW) Operational Target: H2 2027 --- Sharon (Pennsylvania) Secured Power: 110 MW Operational Target: 2027 --- Moses Lake (Washington) - interestingly, this is 1hr from my hometown Secured Power: 18MW Operational Target: H1 2027 --- Sherbrooke + Others (Quebec) 96MW operating across the three Bitcoin mining sites at Sherbrooke, 74 MW additional secured in other Quebec sites No conversion timeline given --- Scrubgrass (Pennsylvania) - in planning Secured Power: None, up to 1.3GW currently under study/planning phase There is technically 123MW energized at this site but there’s no ESA which means they don’t have a full claim or certainty about the power Operational Target: 2028+ --- While most sites have some live power (connected) there is still some work done to build out the full secured capacity. As reminder, the main relevant power statuses are: Secured: signed agreement with utility company committing future power delivery to site, the power doesn’t flow yet and the physical infrastructure may not exist, but the capacity is legally yours Connected: power flowing to the site today Active: connected, lease signed with an AI tenant The Investor Presentation is a bit tricky but it looks like 218MW connected (subtracted 123MW with no EDA) and 430MW secured capacity and a 2.2GW total capacity. The reason why the distinctions are important is because there are build and timing risks between each of the 3 phases. Firstly, getting Secured power is the hardest. It takes 3+ years and is constrained by the fact that the US power grid is ill-equipped to deal with rapid growth. So Capacity does not mean the power will be ready near term. Getting from Secured to Connected is also difficult. This year, half of planned US data center builds have been delayed or canceled. “behind these setbacks is the availability of key electrical components — such as transformers, switchgear, and batteries — that are used both at data center sites and outside of them, as AI companies must expand grid infrastructure to supply enough power to their data centers” Political risks are also rising. There is strong local opposition to data centers, most notably Shark Tank’s Kevin O’Leary’s project in Utah most recently came under heavy scrutiny. Modeling Revenue vs Comps Keel closed on Friday (May 15, 2026) at $4.39. It’s had quite a good run over the last year. It overshot early towards $6 on the AI infra pivot story when it was initially announced. The company also issued convertible debt into the rally, which is something to keep in mind. Based on deals that have been announced by other similar AI infrastructure players we can estimate projected revenue. If we look at the near-term potential contracts with Panther Creek (350MW) + Sharon (110MW) + Moses Lake (18 MW) we get 478MW secured of which we need to apply a discount (20% est.) because there is some power lost to cooling and overhead and compute deals are based on Critical IT which is the power actually delivered for compute. So we can do some back of the hand modeling. Keel has $533m in liquidity ($336m cash, $197m bitcoin) and $588m debt (Jan 2031 maturity) so let’s call that flat. If all 478MW (of which 80% is Critical IT) gets leased for 10 years at the average of $1.48/MW/yr you get $565m per year ($5.65bn total). If we assume a 50% cash margin that gives $311m, which is worth ~$2.6bn PV (10% cap rate). Divided by 604 million shares outstanding, you get about $4.30/share There are of course some assumptions that I made here - Keel could get a higher $/MW/yr deal and they could borrow at better rates and have a higher net margin than in my very rough estimates. On the bear side, this doesn’t take into account the high execution risks that we are seeing clearly leading to delays/cancellations in data center projects. It also assumes all 3 sites get deals. On the bull side, this means that you’re getting fair value for business you believe can come through by year end without even putting a value on the additional 1GW+ of power that’s currently in planning. If Keel executes on the near-term deals the market is likely to put assign a higher risk-adjusted value on the rest of the projects. It is CEO Ben Gagnon’s intention to get all the near-term visible sites leased. “Our 2026 priority is clear. Sign 3 leases by year-end, 1 at Panther Creek, 1 at Sharon, and 1 at Moses Lake” He also implies that they would be able to get better than the median deal: “we believe that the economics are continuing to improve as the scarcity continues to get worse and demand continues to accelerate” The fundamentals back this: “BNP Paribas highlights that consensus estimates for capital spending were for ‘just’ $365bn one year ago, which means this year’s capex estimate of $725 billion is nearly double last year’s estimate.”3 Recent earnings calls across neoclouds and HPC names (Keel’s peers) also reiterate the demand: $NBIS: “we are sold out again in Q1, as we have for several quarters as demand continues to significantly exceed available capacity” “Strong AI cloud demand drives continued pricing power, with several customers vying for each available GPU across all chip generations, and contract durations extending meaningfully" $APLD: “We are seeing a clear acceleration in demand for high-performance AI data center capacity, with hyperscalers as aggressive as we have ever seen them” $IREN: “We are seeing the strongest demand environment to date” And from the end customers: $GOOG cloud backlog rose 93% to $462bn. In a single quarter. $AMZN: "The backlog for Q1 is $364 billion. That does not include the recent deal that we announced with Anthropic for over $100 billion" If we then do more generous calculations, let’s say deal is done on the higher end Revenue: 478 × $1.85m = $884m/yr 50% cash margin = $442m Divided by 10% cap rate = $4.42B equity value Divided by 604m shares outstanding = $7.32 Again, this ignores the power that’s currently in the planning stage at Scrubgrass in Pennsylvania for which the market could add another couple dollars to the stock price on top of multiple re-rating towards $10. Conclusion Execution risks are real and if we get to year end without a deal the price could easily retrace 50% lower and even more so if we get news on delays without clear timeline updates given the infra backlog we see across the industry. Half of planned US data center builds have been delayed or canceled is highlighting a serious issue. Deals can be signed before builds are completed though this is generally with companies that already have pre-existing hyperscaler relationships, which Keel doesn’t have. Overall, I like the name but there are a few paths I see over the next couple of months: Market pulls back on macro, higher energy prices + higher inflation + higher rates story puts the market on the back foot with a previously unthinkable second Fed hiking cycle a genuine risk, the stock retraces towards $3. Project delays and no deals lead to the stock dropping retraces towards $2 to $3. Macro conditions stay favorable, successful execution and a deal on Panther Creek push the stock into the $6 to $8 range. The midpoint will move along with how neoclouds/AI infra in general trades. Relentless demand for neoclouds drives the AI infra complex higher, Keel becomes popular on social media and the market prices Keel for perfection and it goes up 50% to 100% before even announcing a deal. Trades as a high beta to AI infra. Overshoots on deal announcement (like on the pivot last year) up towards $10 I don’t see a lot of evidence that the stock is saturated on social media yet. A search for Keel on Substack doesn’t really return anything, Reddit has some posts but not the number of posts in WSB that give you immediate shock (probably because everything else has rallied so much), and I don’t see any major X accounts covering it yet. A large part of this could be due to the BITF to KEEL rebranding as there are BITF posts from last year. The analysis shows so far that AI infra demand for this year has been relentless, and hyperscaler capex has doubled in a year meaning if Keel executes, finding a customer may largely be assumed a given. Whether or not one chases depends on your tolerance for volatility - a 50% drawdown on execution delays and no deal is possible - and what level of risk adjusted returns you’re seeking. It’s also worth noting that the last time the stock rose aggressively, the company issued convertible senior notes. As for me, this is the type of name I approach with an averaging over multiple clips story. I don’t love the current level but I know I’m not going to time the bottom and it will also bother me if it runs given the projections look overall favorable. I also do assign a high probability that social media turns it’s attention to any narrative that has any stock that has justifiable 2x fundamental story to it even if it means pricing out execution risk. If I end up making a purchase I will post it in the paid subscriber chat for reference. This is not financial advice. Always do your own research. Levels, data, and projections may contain errors and contain assumptions. As always, my aim is not for you to copy my trades but to save you time with research and analyses on interesting ideas.