Up next
Author

📊 Full opportunity report: Week Three — Foundation model vs Brownian motion. Kronos on five-minute BTC. on ThorstenMeyerAI.com — validation score, market gap, and execution plan.

TL;DR

A recent test compared Kronos, a foundation model, to a traditional Brownian motion approach for predicting 5-minute BTC price movements. The results show Kronos does not outperform Brownian motion in out-of-sample tests, challenging assumptions about modern models’ advantages in short-term crypto forecasting.

Recent testing shows that Kronos, an open-source foundation model for financial time series, does not outperform a traditional Brownian motion model in predicting 5-minute Bitcoin price movements on out-of-sample data.

Over two weeks, researchers evaluated Kronos against a Brownian motion baseline using 497 paired trades of Bitcoin on Polymarket’s 5-minute markets. The analysis involved reconstructing market contexts and running simulations to forecast price directions, then comparing the models using Brier scores, log-loss, and hypothetical profit metrics.

The results indicated that Kronos’s predictive performance was statistically indistinguishable from Brownian motion in out-of-sample testing. Specifically, the Brier scores for Kronos and Brownian motion differed by only 0.0011 on 249 trades, well within the margin of noise, meaning Kronos did not demonstrate a clear advantage in short-term predictive accuracy.

While the market-implied probabilities from Polymarket’s order book sat between the two models, the foundation model did not show superior predictive power or profitability in this specific setting, leading to the conclusion that, at least for 5-minute BTC forecasts, modern learned models may not outperform traditional stochastic assumptions.

Polybot Week 3 — Kronos vs Brownian — Thorsten Meyer AI
KRONOS
● RESEARCH SERIES / MAY 2026
THORSTEN MEYER AI · POLYBOT · WEEK 3
POLYBOT · WEEK 3
KRONOS vs BROWNIAN
Research Series · Foundation Model vs Classical Baseline · 2026-05-17

Foundation model
vs Brownian motion.
Kronos on five-minute BTC.

A modern learned model just lost to math from 1900. On 497 paired trades. Stage 2 is not happening.
Polybot’s fair-value strategy uses a 1900s geometric Brownian model to price 5-minute BTC outcomes. The natural follow-up after two weeks of negative parametric results: would a modern learned model trained on millions of real candles do better? The credible candidate: Kronos — open-source MIT-licensed foundation model, 25,000+ GitHub stars, AAAI 2026, four sizes from 4M to 499M parameters, trained on candles from 45 global exchanges. Test design: 497 paired (FILL→SETTLE) trades, Brownian baseline reconstructed line-for-line, Kronos-small (24.7M params) sampled with 16 forecast paths, scored on Brier + log-loss + hypothetical P&L, chronologically split for out-of-sample discipline. On 249 out-of-sample trades: Brownian 0.188 Brier vs Kronos 0.189 Brier. Gap 0.0011. Statistically indistinguishable. Stage 2 is not happening. But the paradox is more interesting than the verdict: when used as a directional signal Kronos fires 28% less often and wins 60.7% vs Brownian’s 49.1% — slightly better trader on hypothetical P&L, even while systematically over-confident in the tails (predicts 2.4% chance → actual 20.4% win; predicts 84% → actual 69.6%). The negative result is the answer. The methodology is what gets published.
Thorsten Meyer AI Polybot · Week 3 MIT-licensed · methodology public Research article · ~2,200 words Forezai · Polybot
This is not financial advice. Nothing in this article should be used to inform real trading decisions. The bot trades simulated money. If you build something like it and run it with real funds, the most likely outcome — by a wide margin — is that you lose those funds. That holds whether you use a Brownian model, a 100-million-parameter foundation model, or any other forecaster.
497
Paired (FILL→SETTLE) trades
all BTC · 5-min Up/Down markets
0.0011
Out-of-sample Brier-score gap
249 trades · statistically indistinguishable
2×
Kronos log-loss vs Brownian
signature of confident wrong predictions
+$538 / +$465
Hypothetical Kronos vs Brownian P&L
the paradox · 60.7% vs 49.1% win rates
POLYBOT WEEK 3· KRONOS-SMALL · 24.7M PARAMS· BROWNIAN BASELINE· 497 PAIRED TRADES · BTC· POLYMARKET 5-MIN UP/DOWN· BRIER 0.193 / 0.211 / 0.213· LOG-LOSS 0.567 / 0.604 / 1.080· OUT-OF-SAMPLE 0.188 vs 0.189· GAP 0.0011 · INDISTINGUISHABLE· STAGE 2 NOT HAPPENING· KRONOS BETTER TRADER · WORSE FORECASTER· 60.7% vs 49.1% WIN RATE· TAILS: 2.4% → 20.4% · 84% → 69.6%· POLYBOT MIT· KRONOS MIT· AAAI 2026 PAPER · 25K+ STARS· 11 MIN MAC M-SERIES · MPS BACKEND· 1,300 LINES OF PYTHON· RESEARCH_PIPELINE.MD PUBLIC· SAME GAUNTLET · DIFFERENT MODEL· POLYBOT WEEK 3· KRONOS-SMALL · 24.7M PARAMS· BROWNIAN BASELINE· 497 PAIRED TRADES · BTC· POLYMARKET 5-MIN UP/DOWN· BRIER 0.193 / 0.211 / 0.213· LOG-LOSS 0.567 / 0.604 / 1.080· OUT-OF-SAMPLE 0.188 vs 0.189· GAP 0.0011 · INDISTINGUISHABLE· STAGE 2 NOT HAPPENING· KRONOS BETTER TRADER · WORSE FORECASTER· 60.7% vs 49.1% WIN RATE· TAILS: 2.4% → 20.4% · 84% → 69.6%· POLYBOT MIT· KRONOS MIT· AAAI 2026 PAPER · 25K+ STARS· 11 MIN MAC M-SERIES · MPS BACKEND· 1,300 LINES OF PYTHON· RESEARCH_PIPELINE.MD PUBLIC· SAME GAUNTLET · DIFFERENT MODEL·
FIG. 01 — THE TEST PIPELINE
Five steps · for every paired (FILL → SETTLE) trade in the running session
~1,300 lines of Python · 11 minutes on Mac M-series with PyTorch MPS · methodology public, specific numbers local
1
Reconstruct OHLCV context of the 60 minutes leading up to fire-time. Pull from the bot’s local Binance recording where available; fall back to Binance’s public klines API otherwise. Cache to parquet so re-runs cost nothing.
2
Recompute the Brownian baseline in Python — a line-for-line port of the bot’s own fairValuePUp(spot, openPrice, secondsLeftFrac, windowVol) formula. Matches scipy.stats.norm.cdf to three decimal places.
3
Read off the market-implied probability from the FILL price — what Polymarket’s order book thought the side was worth at the moment of fire. The market’s view as a reference point.
4
Run Kronos-small (24.7M parameters) on the OHLCV context · sample 16 forecast paths to the window’s end · count the fraction in which the underlying closes above the open price. That fraction is Kronos’s predicted p(Up).
5
Record (p_brownian, p_market, p_kronos, actual_outcome, P&L). Score on Brier + log-loss + hypothetical P&L. Sort chronologically · split into first/second half · report on both halves separately.
The discipline that matters: if a model wins on the first half but ties or loses on the second, that’s the curve-fit-in-slow-motion pattern the previous two articles named, and it doesn’t count as edge. The whole pipeline is reproducible from docs/RESEARCH_PIPELINE.md. Any future candidate model gets a sibling directory in research//, reuses the same Brownian baseline, the same trade-log loader, the same OHLCV fetcher, the same metrics, the same out-of-sample split. Same gauntlet, different model, same discipline.
FIG. 02 — FULL-SAMPLE SCORING · 497 PAIRED TRADES
Three models · two probability-scoring metrics
Brier score and log-loss · the standard scoring rules for probability forecasts · lower is better
Model
Brier ↓
Log-loss ↓
BrownianGeometric Brownian motion · the 1900s baseline
0.193
0.567
Market-impliedPolymarket order book at FILL · reference
0.211
0.604
Kronos24.7M-param foundation model · 16 sampled forecast paths
0.213
1.080
Kronos’s log-loss is roughly twice Brownian’s — the signature of a model that makes confident, wrong predictions in the tails. Polymarket’s order book sits between the two, reasonably calibrated, slightly worse than the bot’s Brownian and slightly better than the foundation model. The 100-year-old math beat the 24.7M-parameter foundation model on both probability-scoring metrics.
FIG. 03 — OUT-OF-SAMPLE VERDICT · 249-TRADE TEST HALF
Chronologically-separated · never seen by tuning
The verdict the test was designed to deliver · noise band of repeated runs with different sampling seeds
Brownian · 249-trade test half
0.188
Brier score (out-of-sample)
lower is better
Kronos · 249-trade test half
0.189
Brier score (out-of-sample)
lower is better
The gap
0.0011
Statistically indistinguishable
inside the noise band
Kronos does not beat Brownian on a held-out chronologically-separated sample. So Stage 2 is not happening.
“Stage 2” was the planned next step: wiring Kronos into Polybot as a live strategy if Stage 1 produced a clear signal. The case is not earned by this data. For 5-minute BTC at the horizons the bot trades, the open Kronos-small checkpoint does not. Stop. The next candidate model — Chronos · TimesFM · Lag-Llama · a Kronos finetune on 5-min crypto · something else — goes through the same gauntlet. Most will fail it. That is the gauntlet doing its job.
FIG. 04 — THE PARADOX · BETTER TRADER vs WORSE FORECASTER
By operational standards Kronos wins · by probabilistic standards Kronos loses
The hypothetical-P&L counterfactual replays the same data through “what if Polybot fired on each model’s probability”
Operational view · Kronos as the better trader
Kronos fires less · wins more · nets slightly more.
Hypothetical fires
201
Brownian fires (reference)
279
Win rate (Kronos)
60.7%
Win rate (Brownian)
49.1%
Hypothetical net P&L (Kronos)
+$538
Hypothetical net P&L (Brownian)
+$465
Fires ~28% less often and wins more reliably when it does. If you use Kronos as a directional signal in a broader system that does its own sizing — closer to how TradingAgents uses analyst outputs — the directional accuracy might still be useful.
Probabilistic view · Kronos as the worse forecaster
Systematically over-confident in the tails.
Kronos predicts
2.4%
Trades actually win
20.4%
Kronos predicts
84%
Trades actually win
69.6%
Log-loss vs Brownian
~2× worse
Brier (full sample)
0.213 vs 0.193
If you are building a fully-probabilistic system where the probability feeds an expected-value calculation against the market’s implied price — which is what Polybot does — calibration is everything, and Kronos’s calibration is bad enough to disqualify it. It thinks it knows more than it does at both ends.
Both interpretations are honest. Neither earns the model a place in Polybot. One of them might earn it a place, later, in TradingAgents — as a 5th analyst voice that votes on direction without being trusted for calibrated odds. That experiment is not what this week tested; it is a separate hypothesis for a separate week.
FIG. 05 — WEEK FOUR · THREE POSSIBLE THREADS
Each is a separate article · the pattern across them is the same
Honest measurement · out-of-sample discipline · no rescue narratives when something doesn’t work
1
A second-tier candidate model · Amazon’s Chronos
Same general shape as Kronos · different training corpus · also open-source. Running it through the exact same gauntlet would say whether the negative result is specific to Kronos or generalises to learned models in this regime.
Generalisation test
2
Kronos with a finetune on 5-min crypto data
The Kronos repo ships a finetuning pipeline. Taking the open Kronos-base checkpoint, finetuning on the bot’s own recorded BTC tick history, re-testing. Isolates “is the pretrained distribution wrong for crypto?” from “is the architecture wrong for this horizon?”
Architecture vs distribution
3
A live-trading update on Polybot
The fleet has been running paper trades continuously across these three weeks. A fresh aggregate-P&L view, with the same calibration-style analysis applied to live performance rather than historical replay, is overdue.
Status reset
The contract is “same gauntlet, different model, same discipline.” Specific numbers stay local. Methodology is public on the repo’s docs/RESEARCH_PIPELINE.md. Publishing reproducible parameter recipes for strategies that might be marginally profitable encourages people to copy them with real money, and the prior on real-money outcomes when copying retail strategies is “they lose.” Publishing the methodology lets the next person test their own model honestly without inheriting any of mine.
By probabilistic standards · Kronos is a worse forecaster. By operational standards · Kronos is the better trader. Both interpretations are honest. Neither earns the model a place in Polybot. One of them might earn it a place, later, in TradingAgents.
Thorsten Meyer AI · Week 3 · Foundation Model vs Brownian Motion
Source dossier & methodology notes
  • Kronos — Open-source MIT-licensed foundation model for financial time series · 25,000+ stars on GitHub · AAAI 2026 paper · four model sizes (4M · 24.7M open small · 102M open · 499M closed) · trained on candles from 45 global exchanges · authors are explicit it is a research model, not a trading system · github.com/shiyu-coder/Kronos
  • Polybot — Open-source paper-trading bot · MIT-licensed · runs against Polymarket 5-minute Up/Down crypto markets · fair-value strategy uses geometric Brownian motion · two prior weeks of published research established that most parametric edges are mechanical artefacts
  • Polymarket 5-min Up/Down — Prediction-market windows on BTC, ETH, and other crypto · the test session: 497 paired (FILL → SETTLE) trades, all BTC · order-book at FILL serves as market-implied probability reference
  • Test design — Run for every paired trade · reconstruct 60-minute OHLCV context · recompute Brownian baseline (line-for-line port of fairValuePUp(spot, openPrice, secondsLeftFrac, windowVol)) · read market-implied probability · sample 16 Kronos forecast paths · record (p_brownian, p_market, p_kronos, actual_outcome, P&L)
  • Scoring rules — Brier score (mean squared error of probability vs actual outcome) · log-loss (penalises overconfidence) · hypothetical P&L (counterfactual if Polybot had fired on each model’s probability with the same edge-margin and risk gates) · all reported on first/second half separately
  • Full-sample (497 trades) — Brier: Brownian 0.193 · Market-implied 0.211 · Kronos 0.213 · Log-loss: Brownian 0.567 · Market-implied 0.604 · Kronos 1.080 · Kronos log-loss ~2× Brownian — confident-wrong-in-tails signature
  • Out-of-sample (249 trades, test half) — Brownian 0.188 · Kronos 0.189 · gap 0.0011 · well inside the noise band of repeated runs with different Kronos sampling seeds · statistically indistinguishable
  • Counterfactual P&L — Brownian: 279 fires / 49.1% win rate / +$465 net · Kronos: 201 fires / 60.7% win rate / +$538 net · Kronos fires ~28% less often, wins more reliably, nets slightly more by operational standards
  • Calibration failures — Kronos predicts 2.4% chance → trades win 20.4% · Kronos predicts 84% chance → trades win 69.6% · systematic overconfidence at both tails
  • Compute — Mac M-series with PyTorch MPS backend · ~1,300 lines of Python · 11 minutes clock-time · cached to parquet so re-runs cost nothing
  • Public methodology — docs/RESEARCH_PIPELINE.md on the project repo · same gauntlet runs any future candidate forecast model (Chronos · TimesFM · Lag-Llama · Kronos finetune · other) · sibling directories research//
  • What this does NOT prove — Not that Kronos is bad (one checkpoint · one horizon · one market) · not that Brownian is good (just not worse at this task; week-2 collapsed at higher sample) · not anything about Stages 2 or 3 of the broader pipeline · negative-Stage-1 kills this candidate at this horizon · the gauntlet does its job
  • Week 4 candidate threads — (a) Amazon’s Chronos generalisation test · (b) Kronos finetune isolating architecture vs distribution · (c) Polybot live-trading aggregate-P&L update with calibration-style analysis
Colophon

Set in Source Serif 4 (display, italic accent), EB Garamond (body), IBM Plex Sans (UI labels), IBM Plex Mono (mastheads, ticker, tags). Paper-cool gray-cream #e6e7e4.

Chromatic register: structural-slate dominant (methodology + out-of-sample-discipline analysis), empirical-clay for the data forensics, labor-rose for the probabilistic-failure framing and disclaimers, transition-bronze for the operational paradox, alternative-sage for the open-methodology positive signal and the better-trader half of the paradox.

Key frame:

POLYBOT WEEK 3 KRONOS-SMALL · 24.7M BROWNIAN BASELINE OUT-OF-SAMPLE 0.0011 THE PARADOX BETTER TRADER WORSE FORECASTER STAGE 2 NOT HAPPENING SAME GAUNTLET DISCIPLINE HONEST MEASUREMENTS

Implications for Short-Term Crypto Prediction Models

This finding questions the assumption that larger, more complex models like Kronos automatically deliver better short-term predictive accuracy in volatile markets such as Bitcoin. It suggests that traditional models based on geometric Brownian motion remain competitive, which has implications for traders and researchers exploring AI-driven forecasting tools. Moreover, it highlights the importance of rigorous out-of-sample testing before deploying advanced models in live trading environments, as perceived advantages may not materialize in practice.

Scalp Smart Hacks to Win Big in Forex Day Trading: Welcome to Scalp Smart, the comprehensive guide designed to transform your forex scalping journey from guesswork to consistent profits.

Scalp Smart Hacks to Win Big in Forex Day Trading: Welcome to Scalp Smart, the comprehensive guide designed to transform your forex scalping journey from guesswork to consistent profits.

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background on Model Testing and Market Expectations

Previous research and anecdotal claims have suggested that modern machine learning models, especially foundation models trained on large datasets, could outperform traditional stochastic models in financial forecasting. However, empirical results have been mixed. The author previously ran a paper-trading bot based on a Brownian motion model, which showed limited edge over random chance. This prompted testing whether a more sophisticated model like Kronos could do better, especially given its training on extensive global exchange data and its recognition in academic circles, including an AAAI 2026 publication.

The testing methodology involved reconstructing market contexts from historical data, running multiple forecast paths, and evaluating the models using well-established probabilistic scoring metrics. The key question was whether Kronos could produce more accurate probability forecasts and, consequently, better trading signals than the simple Brownian baseline.

“The results show Kronos does not outperform the Brownian motion model in out-of-sample tests for 5-minute BTC predictions.”

— Thorsten Meyer, researcher

Chart Logic - Technical Analysis Handbook (Color Edition): The Comprehensive Guide to Trading Cryptocurrencies and Tools to Outperform Your Basic Buy and Hold Investor

Chart Logic – Technical Analysis Handbook (Color Edition): The Comprehensive Guide to Trading Cryptocurrencies and Tools to Outperform Your Basic Buy and Hold Investor

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Limitations and Unanswered Questions in Model Evaluation

It remains unclear whether different training configurations, larger model sizes, or alternative market conditions might yield different results. The current testing focused solely on 5-minute BTC predictions using a specific dataset and model checkpoint. Additionally, the potential for Kronos to outperform in other time horizons, assets, or in live trading conditions remains untested. The analysis also does not account for transaction costs, slippage, or other practical trading considerations that could influence real-world applicability.

Vastarry Crypto Price Ticker Display - WiFi Bitcoin Ethereum Real-Time Dashboard, Desktop LED Monitor for Cryptocurrency Gold Silver Prices, Smart Investment Gift for Traders

Vastarry Crypto Price Ticker Display – WiFi Bitcoin Ethereum Real-Time Dashboard, Desktop LED Monitor for Cryptocurrency Gold Silver Prices, Smart Investment Gift for Traders

Multi-Market Coverage Supports cryptocurrencies, spot gold, spot silver, forex, US stocks, Hong Kong stocks, and A-shares. Cryptocurrency data…

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Future Directions for Model Testing and Market Application

Further research could explore larger or differently trained versions of Kronos, alternative market contexts, or longer prediction horizons. Additionally, live trading experiments may shed light on practical performance differences. Researchers and traders should continue to rigorously evaluate AI models against simple baselines before integrating them into trading systems. The current findings encourage skepticism about the assumption that more complex models automatically lead to better short-term forecasts in highly volatile markets.

Stock & Crypto Price Display, Real-Time Desktop Ticker with Candlestick Charts, Live Market Data Gadget, Track US Stock Market TSLA NVDA SPY, Watch Bitcoin BTC ETH Doge Web3 (Wood USB-C)

Stock & Crypto Price Display, Real-Time Desktop Ticker with Candlestick Charts, Live Market Data Gadget, Track US Stock Market TSLA NVDA SPY, Watch Bitcoin BTC ETH Doge Web3 (Wood USB-C)

STOCK & CRYPTO MARKET DATA – Track both US stocks (SPY, TSLA, NVDA) and crypto pairs (BTC, ETH,…

AmazonView Latest Price

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Does this mean foundation models are useless for crypto trading?

No, this specific test shows they do not outperform simple stochastic models in 5-minute BTC predictions, but future research may find different results or applications in other contexts.

Could larger or more advanced versions of Kronos do better?

This remains an open question. The current study used a specific model size, and larger or differently trained models might produce different outcomes.

What does this imply for traders using AI models?

It emphasizes the importance of rigorous out-of-sample testing and skepticism about claims that larger models automatically improve short-term prediction accuracy.

Will the results change with different assets or timeframes?

This is uncertain. The current findings are specific to 5-minute Bitcoin predictions; other assets or longer horizons may yield different results.

Source: ThorstenMeyerAI.com

You May Also Like
Every Benchmark Launched 2023-2024 Has Fallen — The METR / SWE-Bench / CORE-Bench / MLE-Bench / PostTrainBench Sequence

Every Benchmark Launched 2023-2024 Has Fallen — The METR / SWE-Bench / CORE-Bench / MLE-Bench / PostTrainBench Sequence

Every major AI research benchmark launched between 2023-2024 has now saturated or is nearing saturation, signaling accelerated AI capability development.
Gemini 3.5 Flash

Gemini 3.5 Flash

Google introduces Gemini 3.5 Flash, a new AI model delivering high-speed, agentic, and multimodal performance for developers, enterprises, and everyday users.
Show HN: Nobie – An Excel-compatible Runtime For Agents And Humans

Show HN: Nobie – An Excel-compatible Runtime For Agents And Humans

Nobie introduces an Excel-compatible runtime designed for agents and humans, enabling easier automation and interaction with spreadsheets. Development announced on Show HN.
Using AI to write better code more slowly

Using AI to write better code more slowly

Developers are exploring a slower, more deliberate use of AI in coding to enhance quality, contrary to the popular fast, low-quality approach.