📊 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 found that the open-source foundation model Kronos does not outperform the traditional Brownian motion baseline in predicting 5-minute BTC price movements. The results suggest current learned models may not offer a trading edge over classical assumptions.
Recent testing shows that the open-source foundation model Kronos does not outperform the traditional Brownian motion baseline in predicting 5-minute Bitcoin price movements, challenging expectations for modern AI models’ trading edge.
Researchers conducted an out-of-sample evaluation of Kronos, a 25,000-star open-source foundation model trained on global crypto exchange data, against a geometric Brownian motion baseline. Using a custom Python tool, they analyzed 497 BTC trades recorded by a paper-trading bot over a recent period. The models’ predicted probabilities of price increases were scored on accuracy and hypothetical profit. Results showed that Kronos’s predictive performance was statistically indistinguishable from Brownian motion, with both models performing slightly better than market-implied probabilities. Specifically, on the test sample, Kronos’s Brier score was 0.189, nearly identical to Brownian’s 0.188, with no significant outperformance. This indicates that, at least for 5-minute horizons, Kronos does not provide a measurable trading advantage over the classical assumption of independent, normally-distributed returns.KRONOS
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
all BTC · 5-min Up/Down markets
0.0011
Out-of-sample Brier-score gap
249 trades · statistically indistinguishable
249 trades · statistically indistinguishable
2×
Kronos log-loss vs Brownian
signature of confident wrong predictions
signature of confident wrong predictions
+$538 / +$465
Hypothetical Kronos vs Brownian P&L
the paradox · 60.7% vs 49.1% win rates
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
lower is better
Kronos · 249-trade test half
0.189
Brier score (out-of-sample)
lower is better
lower is better
The gap
0.0011
Statistically indistinguishable
inside the noise band
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
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 AI-Driven Crypto Trading Strategies
The findings suggest that even advanced foundation models like Kronos may not currently offer a meaningful edge in short-term crypto trading, raising questions about the practical benefits of deploying large learned models in high-frequency contexts. This challenges the assumption that modern AI can easily outperform traditional stochastic models in financial markets, emphasizing the importance of rigorous out-of-sample testing. For traders and developers, it underscores the need for cautious evaluation before integrating such models into live systems, especially given the high costs and risks involved. The result also highlights the persistent relevance of classical models like Brownian motion, which continue to serve as reliable baselines in market prediction tasks.
The Only Bitcoin Investing Book You’ll Ever Need: An Absolute Beginner’s Guide to the Cryptocurrency Which Is Changing the World and Your Finances in 2021 & Beyond (Cryptocurrency for Beginners)
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, a researcher has been running a paper-trading bot called Polybot, which uses a geometric Brownian motion model to predict 5-minute BTC price movements. The bot’s performance was compared against actual market outcomes and a more sophisticated foundation model, Kronos, trained on extensive global exchange data. Previous attempts to find an ‘edge’ in short-term trading using various strategies have largely failed, with most models producing no consistent advantage. Kronos, developed as a research tool and not a trading system, was tested to see if it could do better than the traditional Brownian assumption. The testing methodology involved reconstructing market contexts and simulating forecasts for each trade, then evaluating the models’ predictive accuracy and hypothetical profitability.“Kronos does not outperform Brownian motion in our out-of-sample tests for 5-minute BTC predictions, indicating no current practical edge.”
— Thorsten Meyer, researcher
Polymarket Profits – Build AI Trading Bots in a Weekend: The Step-by-Step System for Investing in Prediction Markets Without a Finance Degree
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Uncertainty About Long-Term and Different Horizons
It remains unclear whether Kronos or similar models might outperform traditional baselines over longer timeframes or different market conditions. The current test focused solely on 5-minute horizons, and the model’s performance in other contexts has not yet been evaluated.
Ready to use BTC/USDT Price Analysis Bot: Generating Buy and Sell Signals with Python Code for Crypto Traders: Based on the Ichimoku strategy conditions
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Next Steps in Model Evaluation and Market Testing
Further research will explore whether larger or differently trained models can demonstrate a measurable advantage in other trading horizons or under varying market regimes. Additionally, testing in live trading environments and across different assets may provide more insights into the practical utility of foundation models in finance.
CryptoTicr Bitcoin Ticker – Real-Time Crypto Price Display (1.54" Diagonal) for Top 1,000 Coins Like Bitcoin, Ethereum, Dogecoin, and 997 more. Compact Size, Ideal for Desk or Nightstand. Uses Wi-Fi.
CryptoTicr is a passion project of a dedicated USA-based creator focused on crafting the finest cryptocurrency ticker available….
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?
Not necessarily. The current results indicate that, for 5-minute BTC predictions, Kronos does not outperform classical models. Future research may find advantages in other contexts or with different models.
Could larger or more specialized models perform better?
It’s possible. The current test used a specific size of Kronos; larger or differently trained models might yield different results, but this remains to be tested.
What does this mean for traders using AI models?
It suggests caution. Even advanced AI models should be rigorously evaluated out-of-sample before deployment, as traditional models can perform just as well in certain short-term trading scenarios.
Will the results change with different market conditions?
Potentially. The current evaluation was based on recent market data; different regimes could impact model performance, but further testing is needed.
Source: ThorstenMeyerAI.com
Nothing in this article is financial or investment advice. Cryptocurrency and precious-metal investments carry significant risk — do your own research and consider a licensed advisor.
