Tim (@timsegewitz) header
Tim (@timsegewitz) profile picture

9+ years in quantitative research. Founder & ML-Engineer building @macroinsight. Prev @hkcm1.

→
Joined July 2019
Iran drones target Bahrain and a third US carrier arrives, yet oil slips on glut outlook Brent fell to $95.30 a barrel as an IEA forecast of a fourth-quarter surplus outweighed the first Iranian attacks on Bahrain since September.
1
1
4
360
Oil rebounds 1.7% as Iran denies US talks, though Gulf crude flows have largely recovered Brent rose 1.7% to $93.08 but is still roughly 23% below its October 30 high, as crude flows recover while fuel exports and shipping remain disrupted.
1
2
8
261
US adds 165,000 jobs in October, triple the recent pace, keeping a Fed hike in play Two-year yields rose as traders kept high odds of a December hike, though those odds sit below late October's, and the 10-year stays lower than a month ago.
4
1
5
139
Nasdaq 100 has worst day since 2020 as chips fall on weak earnings and data-center fears A slide in AI and chip stocks that began in October spread to every S&P 500 sector, pushed the Nasdaq Composite into correction and hit Asian markets on Friday.
1
1
5
168
Oil bounces only modestly after US strike on Iran's Qeshm Island, after a 7.6% fall Brent regained only a small part of Monday's diplomacy-driven drop, as traders weigh a limited strike against talks that have not been called off.
1
1
5
180
Pay attention: Friday's CPI print could decide the Fed's next move, here's what we see. Economists expect headline CPI to rise 0.4% in August after just 0.1% in July, holding the annual rate at 3.4%. Core is seen up 0.2% month on month, with the yearly rate easing to 2.4% from 2.5%. Here's the nuance. The headline bounce is mainly gasoline and energy, so the gap versus core should WIDEN. This is key: core is the number that decides whether the Fed holds or hikes next week. Officials have signaled 0.2% keeps a pause alive, while 0.3% or more reopens a hike. In-line prints should leave policy odds intact. A cooler mix would cut hike odds and boost Treasuries and gold. A hot surprise would firm up September hike pricing and pressure growth stocks.
1
2
211
Bitcoin's MVRV bands are pointing to accumulation, prepare for what might come next. MVRV Statistical Bands wrap bitcoin:native's cost basis in standard-deviation levels, turning one number into a full map. Dark red at +2σ marks historical tops. Purple is fair value. Dark blue at -1σ is where the next cycle's fuel has usually loaded. Here's where price sits now. It pressed the upper red band into the 2024-2025 highs, then spent most of 2026 between the mean and the lower half of the grid, far from both the euphoric top and the dark-blue washout near $16k. To put this into perspective: realized price, the MVRV equals 1 line, sits near $53k and looks like the structural low this cycle can launch from if the unwind stays orderly. The read: a black-swan forced-seller event could still tag the -1σ band in the $40k's, but that's the tail risk, not what the bands are showing today.
1
1
2
130
Pay attention: Ethereum keeps outperforming Bitcoin, here's what we see. ethereum:native has been the leader of the majors in the recent rotation higher, continuing to show strength over BTC. This is key: if crypto sees a dip in the coming weeks, the 2,200 level holds significant liquidity, which should support price. Any absorption there is a great long signal. The backdrop supports it. Crypto markets are heating up, and Saylor's selling pressure is being continually ignored. Sum it all up and Ethereum stands out as a great long idea into the big supply level at $3,000, where we can expect sellers to step in and suppress price.
1
1
2
137
The warning signs copper is showing us right now... Here's the read. Three things stand out. One: copper is battling its previous all-time high with clear ranging price action on the higher time frames. Two: the macro is still bullish, but signs of weakness are emerging after the recent flush. Three: sellers are clearly in control on the lower and medium time frames. In other words, this is a range, not a breakout. This is key: the play is a rotational short on any move higher, aiming for the range lows. Metals also tend to drift sideways with slight drawdowns in September, which favors range trades over trend-following.
2
1
3
223
The Nasdaq is setting up for a dip before a bigger move, prepare for what might come next. After last week's close, the Nasdaq showed local weakness, failing to reclaim key support. And September is working against it. To put this into perspective: September is the single worst performing month for US indices in history. That seasonal drag may be exactly what fills the long idea lower. Here's where it turns. October and November tend to be very positive months, and that's where this idea should thrive, with follow-through toward the previous all-time high. This is key: the higher time frame structure stays constructive, with a potential Head and Shoulders in the making as a base to spring higher.
1
1
4
241
The real driver behind rising Treasury yields is coming into view, prepare for what might come next. The move higher in yields reflects stronger nominal growth, sustained by a 6% of GDP deficit and heavy AI-related capex. Not a failed attempt at yield-curve control. To put this into perspective: the emerging policy mix keeps the front-end policy rate STEADY while delivering liquidity through expanded bill financing, reserve management, and private absorption of short-dated paper. Here's how you know it's working: Stable repo conditions. Contained bond volatility. A steeper curve. This is key: rising yields are a near-term headwind, but the broader backdrop, fiscal stimulus financed at the short end plus a shift toward real-economy spending, still favors liquidity-sensitive assets and real assets.
1
1
4
163
Tim retweeted
This is *not* how you use AI in trading. People actually doing this will get rekt. AI has made backtesting fast and accessible to anyone with data. What it has not removed is your need to judge the output and catch the mistakes made along the way. And that's the trap, because not all friction is the same. Using AI well is mostly one skill: telling apart friction that only costs you time from friction that is actively training you. The key question is which friction you need to keep and go through yourself to be able to judge the output. This way you learn about lookahead bias, overfitting, etc. That's the intellectual debt you accumulate when you let AI handle the whole process. People who outsource all of it make AI torture historical data until it confesses to anything. I've literally watched people backtest a strategy, find that it underperforms on certain days of the week, and simply delete those periods from the sample. To them, it looked like refinement, but it's quite the *opposite*. Which is exactly what Step 4 tells you to do. "Analyse this trade history and tell me how to improve this strategy" is an instruction to overfit. There's very little fitting the average person can do before fucking things up big time. In practice, AI is just amplifying the same old biases with a lot more computational power and speed. You just make them at scale and get rekt quicker. That being said, backtests are in themselves very tricky, mainly due to path dependence, which means your results aren't just about whether your strategy works.
I built a trading bot with Claude that printed +$168,236. I have zero technical background, and I used Fable 5 to backtest and build the entire strategy. If you inject this prompt into Claude Code, your AI agents will help you build a profitable trading strategy from scratch:
25
13
4
309
35,426
The stock market looks calm, but there’s a real risk building underneath, and almost no one is protecting against it. Investors have stopped buying “insurance” against a drop, right when they should be buying more of it. Everyone assumes the good times keep going. To be clear: the danger isn’t company earnings. It’s that too many investors have borrowed heavily to bet on the same AI stocks. If they’re forced to sell at once, prices could fall fast. And the recent dip? People didn’t panic-sell, they bought more. So all that borrowed-money risk is still sitting there. The bottom line: the bets are still on, but the safety nets are not.
1
3
193
bitcoin:native 's patient money is signaling a floor, prepare for what might come next. Adjusted Long Term Holder MVRV asks whether veteran holders are in profit or in pain. It measures spot price against the cost basis of coins aged six months to ten years, filtering out fresh speculators and lost Satoshi-era supply. Here's the read. Above 10 is Extreme Profit and heavy distribution. At 1 the cohort breaks even, and below it the blue Extreme Underwater bars fire. To put this into perspective: the peak keeps decaying each cycle, from 48 in 2013 down to 3.5 last cycle, but the blue prints keep firing at EXACTLY 1. In 2012, 2015, 2018, 2022-2023, and now in 2026. The read: each window lasts months, marking accumulation zones rather than timing signals. With price sitting on the cohort's cost basis near 65K, staggered buys beat one bet on the exact low.
1
4
161
The hidden risk under the S&P 500 rally is coming into view, prepare for what might come next. Record-high valuations have left US stocks hypersensitive to liquidity. A small contraction could now swing the market by roughly 10%. To put this into perspective: the real risk isn't earnings. It's the yen carry trade sitting beneath the whole rally. If that trade unwinds, a modest rise in the yen could drive a sharp drawdown, even with strong tech and NVDA numbers. This is key: this exact setup played out in 2024, when the unwind took the Nikkei down 28%.
1
3
111
Pay attention: Nvidia is flashing renewed strength, here's what we see. $NVDA is showing fresh signs of momentum with the recent swing up on the Nasdaq and S&P 500. Here's the tell. Last week's strong close above the 215 RESISTANCE was the first sign of momentum in NVDA since the high in early May. In other words, the leader is waking up. As an industry leader with positive fundamentals and high market share, NVDA is one of the FIRST assets to look for longs on signs of strength. It's often the leader of the pack when momentum resumes. This is key: the idea is continuation into fresh ALL-TIME HIGHS, targeting the 250 to 260 area where the first resistance zone sits.
1
1
5
210
The global liquidity cycle is entering a new phase, prepare for what might come next. Liquidity was stable last week near 193.6 trillion dollars, but 3-month annualized growth has slipped to 5.1%. The momentum is fading. To put this into perspective: fast-reacting assets like crypto and some emerging markets are ALREADY responding to the slowdown. Here's the Fed angle. New chair Kevin Warsh is dropping the old habit of steering markets through press conferences. Instead, he's letting bond swings and a steeper yield curve cool the economy while keeping short-term borrowing cheap. This is key: he judges 10-year Treasury yields to be 1 to 2 percentage points too low for how fast the economy is growing. They have room to RISE, which could eventually pressure stocks. And the latest data reinforces the shift: the U.S. economy LOST 23,000 jobs in July against expectations of an 80,000 gain, cutting the probability of a September Fed rate hike from 60% to 30%.
1
3
124