Chris
59 posts
Jack Mallers gives the best argument ever on why #Bitcoin is SUPERIOR to Gold against Peter Schiff 🔥
English
Chris retweetledi
Very cool overview of who owns Bitcoin.
English
@jv_finance this is a good lesson of statistics. we’re not dealing with Gaussian distribution :)
English
@jv_finance What exactly did you change? It you share some magic chart without explaining how you calculating it, it’s of lesser value to anyone.
English
The updated version of Bitcoin vs. Global Liquidity is here!
The model has been reworked and is now more robust and structurally more stable.
Let’s look at the new data:
Bitcoin is still oversold, but to a lesser degree than in the old model (−1.4σ vs. −2σ).
Fair value is lower than before and currently sits around $150k.
Bands:
±1σ: $105k–$218k
±2σ: $73k–$313k
Based on this update, additional downside is not as unrealistic as I previously thought.
Liquidity would allow a drop into the $70k range, bringing Bitcoin closer to its cost of production.
Even so, a traditional bear market currently still seems highly unrealistic.
English
@thealepalombo Too much long term geopolitical risks. Last 100 years in Poland : WWI, WWII, 40 years of Communist regime. Now the most prosperous period in PL history. But until when?
English
Poland might be Europe's best-kept arbitrage.
- €70K flat tax covers unlimited foreign income
- Valid for 10 years
- No investment required
One-third the cost of Italy's regime, and nobody's talking about it.
Here's what you need to know:
English
@jv_finance this assumes fitting for an extended time window (several years, looking at your chart). the coefficients are not constant. we saw it in the past. all I'm saying is the FV could be much lower than you think (but still higher than the current price)
English
Current levels:
+2σ: 373k
+1σ: 262k
FV: 184k
-1σ: 129k
-2σ: 90k
Julius@jv_finance
Liquidity is flowing into the market. Fair value has increased from $160k to $184k in just two weeks. Time for Bitcoin to catch up.
Čeština
I don't know how you calculate the GLI but most of the data is on monthly basis, and you have weekly ticks. I took G7-RRR-TGA and found a cubic transformation that has R²=.9622.
I also took BTC market cap instead of BTCUSD as there's increasing amount of BTC in circulation
According to my model we're at z-score: -1.01σ
Why cubic transformation makes sense?
The cubic coefficients are NOT chosen by me, they are DISCOVERED by the optimization algorithm by finding the values that best fit the data.
I applied ORDINARY LEAST SQUARES (OLS) OPTIMIZATION and solved CONVEX optimization problem
The algorithm 'learned' that BTC's sensitivity to liquidity decreases over time, purely from the data, without me telling it to! This is why R² = .9622 is so impressive - it's a data-driven discovery, not a hand-tuned model
English
@jv_finance Have u considered that GLI should take into account different factors for different countries? Like for whatever reason might be china has only 80% sensitivity while US has 100 and Europe 90 and there could be even better shape of that time series?
English
@fudalichris FTX only reached –1σ.
The Covid crash hit the lowest reading at –2.5σ.
English
Currently sitting at –1.75σ.
It’s becoming extremely oversold on a macro level.
Julius@jv_finance
Liquidity is the main driver of Bitcoin. Deviations can last for weeks or even months, but the gaps always close eventually. Bitcoin is currently at –1.66σ, with fair value around $160k.
English
Chris retweetledi
@PeterSchiff I wonder if he really believes what he posts or it it’s a denial, he can’t stand the fact he missed it when it was early or it’s just his way to get some attention
English
While bull markets climb a wall of worry, bear markets slide a slope of hope. Today’s Bitcoin rally gave HODLers false hope that the bear market is over and that a rally to new highs is imminent. As a result, they stay invested or buy more when they should be selling the rally.
English
I’ve been trying to build such indicator for a long time but have not enough skills in that area. Main problem is that GLI correlates very well with BTC but only locally, let’s say few months. One can adjust the scales locally so it matches but then it’s completely wrong in another place. I tried applying some “adoption factor” which changes logarithmically over time, applied some exp factor to GLI but I gave up eventually. If you could share the code what you have so far that would be amazing!!!
English
@fudalichris Yes, I will release them soon — I’m working on it right now.
I’ll announce it when I’m done.
English
This is the full Global Liquidity Model.
It covers more than 10 years of price history — with R² still rising.
It’s not meant to capture every short-term move of Bitcoin, but rather to represent its long-term baseline.
Julius@jv_finance
Welcome our new Global Liquidity Model for Bitcoin! This model aggregates the most relevant measures of liquidity into a single framework. It’s ideal for identifying zones of over- and undervaluation. Currently, Bitcoin is oversold (-1.12 SD), with fair value sitting at $168k.
English
Divergences this big almost never happen in bull markets.
Let everyone believe this was the top — they’ll be sidelined during the next leg up.
The liquidity outlook is extremely bullish over the coming months.
Julius@jv_finance
Bitcoin could still drop a bit lower in the short term. Many on-chain metrics aren’t signaling a bottom yet, and liquidations don’t look great either. Still, my outlook for the next few months remains bullish. Bitcoin is extremely undervalued, and liquidity is set to rise in the coming months. I’ll use every dip to deploy the rest of my cash. Don’t get sidelined.
English
Bitcoin could still drop a bit lower in the short term.
Many on-chain metrics aren’t signaling a bottom yet, and liquidations don’t look great either.
Still, my outlook for the next few months remains bullish.
Bitcoin is extremely undervalued, and liquidity is set to rise in the coming months.
I’ll use every dip to deploy the rest of my cash.
Don’t get sidelined.
English
No niestety, nie doceniłem trudności tego ćwiczenia. Okazuje się że zbudowanie fajnej technologii i dodzwonienie się do dużej grupy ludzi to nie wszystko. Nie pomogła mi ostrożność i staranność z danymi i obliczeniami. Wniosek jest taki że prawdopodobnie korelacja wykształcenia i wielkości miejscowości z preferencjami wyborczymi to jest to co totalnie zaburzyło badanie. Ale się nie poddaje! mam jeszcze drugą turę na poprawki
Polski
Czy AI może wreszcie zrobić porządny sondaż wyborczy?
A gdyby tak wykorzystać niesamowitą technologię od ElevenLabs i przeprowadzić sondaż przed wyborami na dużą skalę, tak aby wielkość próby była znacznie wyższa niż w tradycyjnych sondażach?
Conversational AI od ElevenLabs działa jak magia ✨
Dostałem grant od ElevenLabs na stworzenie czegoś fajnego i pożytecznego dla świata
Agent AI nie tylko się skaluje i pozwala zdobyć więcej danych. On jest zawsze neutralny. Każdą rozmowę przeprowadza według tego samego scenariusza. Jest zaangażowany! Pozbywamy się biasów i efektu znużenia.
Gdy połączymy to z rzetelnym modelem statystycznym to jest szansa na zbudowanie czegoś, czego jeszcze nigdy nie było!
Zegar tyka ⏳, do wyborów już niewiele czasu, ale myślę, że uda się dowieźć coś wartościowego w ciągu kilku dni. Pozostaje politycznie neutralny, zajmuje się technologią i danymi, czymś na czym się trochę znam
Posłuchajcie sami jak to brzmi, co sądzicie? 🚀
Polski
@MarcinPalade @MagdaBiejat w sondażu przy użyciu AI wychodzi 5.8%, próba 2K
Polski
W jakim przedziale zamknie się poparcie dla @MagdaBiejat w pierwszej turze wyborów? RT
Polski
Polski
🔔 Już jutro o 9:40 w programie @DDTVN wystąpi nasz analityk Mateusz Cholewa (@matichol96)!
🎙️ Temat? Dezinformacja w sieci – jak ją rozpoznać, unikać i skutecznie się przed nią bronić.
📱 Media społecznościowe to nie tylko źródło informacji, ale też przestrzeń pełna manipulacji. Mateusz podpowie, na co szczególnie warto zwracać uwagę.
📺 Włączcie telewizory i bądźcie z nami o poranku!
Polski
@K_Stanowski Czy jest jakiś wynik (np. zbyt niski albo zbyt wysoki) z którego będziesz niezadowolony? Na ten moment sondaż AI daje ci 1.9%
Polski
@GiertychRoman Ciekawie, bo w mojej ankiecie AI (N = 2000, 12-13 V) Trzaskowski wyszedł 42,4 %. To o 10 pp więcej niż w Twojej prognozie – aż tak duży efekt końcówki kampanii? 🤔
Polski
Moja prognoza.
1 tura.
Trzaskowski 32
Nawrocki 30
Mentzen 12
Hołownia 8
Braun 6
Zandberg 5
Biejat 4
2 tura
Trzaskowski 51
Nawrocki 49
Polski