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

📊 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 comparing Kronos, a foundation model, against a Brownian motion baseline for 5-minute Bitcoin predictions shows no significant advantage. The study used historical data and out-of-sample testing, finding Brownian motion remains competitive.

Recent testing indicates that Kronos, a large open-source foundation model trained on global crypto data, does not outperform the traditional Brownian motion model in predicting 5-minute Bitcoin price movements, challenging assumptions about the superiority of modern learned models in short-term trading signals.

Over two weeks, an open-source paper-trading bot called Polybot, which uses a geometric Brownian motion model to estimate Bitcoin price probabilities, was tested against Kronos-small, a foundation model trained on millions of candlesticks from 45 exchanges. The evaluation involved 497 historical trades, with out-of-sample analysis showing that Kronos’s predictive accuracy, measured by Brier score and log-loss, was statistically indistinguishable from Brownian motion. Specifically, on the last 249 trades, Kronos’s Brier score was only marginally better, but the difference was within the margin of error, indicating no significant advantage.

The experiment was designed to determine if a modern, learned model could beat a century-old mathematical approximation in short-term crypto prediction. Despite expectations, Kronos did not demonstrate superior predictive performance in this context, leading to the conclusion that it is not yet ready to replace simpler models for five-minute BTC trading signals.

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.
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
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

Implications for Crypto Trading Strategies

This finding suggests that, at least for short-term predictions at five-minute horizons, complex foundation models like Kronos do not currently offer a measurable edge over traditional stochastic models like Brownian motion. For traders and algorithm developers, this indicates that reliance on sophisticated models may not translate into better performance without further advancements or different application scopes. It also underscores the importance of rigorous out-of-sample testing before deploying new predictive models in live trading environments.

Bitcoin Pillow – Embroidered Round Stuffed Plush BTC Crypto Cushion – Soft Bitcoin Decor for Traders & Crypto Fans

Bitcoin Pillow – Embroidered Round Stuffed Plush BTC Crypto Cushion – Soft Bitcoin Decor for Traders & Crypto Fans

Embroidered Iconic Bitcoin Design – Features a bold, embroidered BTC logo in a soft round plush—perfect for traders,…

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background on Model Testing and Market Predictions

Over the past two weeks, the author tested various predictive models against a simulated trading bot operating in Polymarket’s 5-minute crypto markets. The bot’s core strategy relies on a geometric Brownian motion model, a mathematical assumption dating back to the early 20th century, which treats log-returns as independent and normally distributed. The question arose whether modern, data-driven models trained on extensive historical market data could outperform this simple baseline. Kronos, an open-source foundation model trained on millions of candlestick data from global exchanges, was identified as a promising candidate for this purpose. The testing involved reconstructing market conditions around each trade and evaluating the predictive accuracy of Kronos against Brownian motion, using metrics like Brier score and log-loss.

“Despite the hype around foundation models, our tests show they currently do not outperform traditional stochastic models like Brownian motion for short-term BTC predictions.”

— Thorsten Meyer

Crypto Seed Cold Storage Wallet with Engraver Pen Kit - Metal Plate and Etching Tool for Cryptocurrency Password Phrase Backup and Recovery

Crypto Seed Cold Storage Wallet with Engraver Pen Kit – Metal Plate and Etching Tool for Cryptocurrency Password Phrase Backup and Recovery

All Inclusive Kit for Crypto Seed Key Storage – Comes a Stainless Steel Plate & Tungsten Steel Engraving…

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Limitations and Unanswered Questions in Model Performance

While the current results show no significant advantage of Kronos over Brownian motion for 5-minute BTC predictions, it remains unclear whether different model configurations, longer evaluation periods, or other market conditions could yield different outcomes. The test focused solely on short-term prediction accuracy and did not assess other potential benefits such as robustness or adaptability. Additionally, the models were evaluated offline, and real-time performance may differ.

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…

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Future Testing and Potential Model Improvements

Further research is needed to explore whether larger or differently trained foundation models can outperform simple stochastic models in various market regimes or time horizons. Developers may also investigate hybrid approaches combining traditional models with learned components. In the short term, traders should remain cautious about assuming that modern AI models automatically provide an edge in short-term crypto trading. Continued testing, both offline and in live environments, will be essential to determine the practical value of these models.

Candlestick Trading Beyond Fake Signals: A Structure-First Framework for Reading Charts the Way Institutions Tend to Move Price Across Stocks, Forex, Crypto, Futures and Commodities

Candlestick Trading Beyond Fake Signals: A Structure-First Framework for Reading Charts the Way Institutions Tend to Move Price Across Stocks, Forex, Crypto, Futures and Commodities

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Does Kronos outperform Brownian motion in predicting Bitcoin prices?

Current tests show no statistically significant difference in predictive performance between Kronos and Brownian motion for 5-minute BTC predictions.

Can foundation models replace traditional stochastic models in crypto trading?

Based on recent results, foundation models like Kronos do not yet demonstrate a clear advantage over traditional models in short-term prediction accuracy.

What are the limitations of this testing approach?

The evaluation was offline, based on historical data, and focused solely on short-term prediction metrics. Real-time performance and other market conditions remain untested.

Will future versions of Kronos perform better?

It is uncertain; larger or differently trained models may yield different results, but current evidence does not support their superiority in this context.

What does this mean for traders using AI models?

Traders should be cautious in assuming that advanced AI models automatically confer an edge in short-term crypto trading without thorough validation.

Source: ThorstenMeyerAI.com

You May Also Like

The bank account in the chat. How personal finance became an agentic on-ramp.

OpenAI launched a preview of personal-finance tools in ChatGPT, connecting bank accounts for Pro users, marking a shift toward agentic consumer finance.

The CFO’s new operating system. Anthropic, OpenAI, and the consulting margin that just got compressed.

Anthropic’s $1.5B joint venture and OpenAI’s parallel funding reshape enterprise AI, integrating models into CFO workflows and disrupting consulting margins.

The gigawatt gap. Why China is structurally positioned for AI power and the US is engineering around its grid.

China’s centralised renewable grid and power capacity enable large-scale AI deployment despite lower chip performance, contrasting US fragmentation and grid constraints.