ETH quietly did the heavy lifting. XRP just came along for the ride.
ETH entered March 2025 at $2,235.94, still shaking off February weakness. XRP was trading at $2.1422.
Over the next 106 days, ETH climbed to $2,547.61 — up 13.9%. XRP closed at $2.1655 — up just 1.1%.
One asset did almost all the work. The other barely moved.
The question: can a 50/50 rebalancing bot systematically harvest that gap — selling ETH strength back into XRP weakness — without a human flinching at the wrong moment?
Strategy Parameters
How Each Setting Impacted Performance?
Rebalance bots are simple in concept – sell the winner, buy the laggard – but trigger sensitivity decides whether that discipline pays off or bleeds into fees.
Parameter Impact Summary
| Parameter | Impact | The Logic (Why) |
|---|---|---|
| ETH/XRP 50/50 Pair | 🎯 Divergence exposure | Correlated but diverging assets |
| 2% Ratio Trigger | 🔁 Frequent rebalances | Small deviations still fire trades |
| 1m Time Check | ⏱️ Fast reaction | Catches ratio drift quickly |
| 0.1% Fee Rate | ⚠️ Low fee drag | Only 21 trades kept fees small |
| 50/50 Target Weight | ⚖️ Forced profit-taking | Sells winner, buys laggard automatically |
| 106-Day Window | 📈 Captured full ETH rally | Long window let divergence compound |
21 trades. $357.68 net profit. A 1.76% edge over just holding.
💰 The Bottom Line
On $4,000 deployed, the bot generated $357.68 net profit – 8.94% over 106 days. Annualized, simply, that’s roughly 30.8%.
Stay grounded: this window captured one clean divergence cycle, not a repeatable monthly rate.
⚖️ The Rebalancing Edge
A pure 50/50 buy & hold of ETH and XRP would have returned 7.18% ($287.20) — basically the average of ETH’s +13.9% and XRP’s +1.1%.
The bot beat that by 1.76 points, turning $287.20 into $357.68. That $70.48 gap is the entire value proposition of rebalancing.
🛡️ The Fee Reality
21 trades at 0.1% cost just $5.73 total – 1.6% of net profit.
With roughly one trade every 5 days, this trigger stayed lean. That matters when you see what the aggressive variant below paid for 3x the activity.
Here comes our A/B/C strategies quick comparison:
| Variant | Ratio / Time | Trades | ROI % | P&L (USDT) |
|---|---|---|---|---|
| Test A (Conservative) | 5% / No time trigger | 3 | 8.41% | $336.39 |
| Test B (Aggressive) | 1% / 30m | 66 | 8.67% | $346.95 |
| Test COptimized | 2% / 1m | 21 | 8.94% | $357.68 |
Test A’s wide 5% band barely reacted — only 3 rebalances in 106 days meant it missed most of ETH’s rally, leaving the lowest return despite near-zero fees.
Test B overcorrected: 66 trades chasing every 30-minute wobble produced only marginally more profit than Test C, while paying roughly 3x the fees for 3x the activity.
Test C’s 2% trigger with a fast 1-minute check found the sweet spot — active enough to capture ETH’s climb, disciplined enough to avoid fee bleed.
What the results are really telling you.
✅ what worked
The 2% ratio trigger caught ETH’s climb early and often. As ETH pulled ahead of XRP, deviations from the 50/50 target kept crossing 2% — triggering sells like the March 2 rebalance that moved 0.0429 ETH near $2,232 into XRP.
Repeated across 21 points from March to June, this systematically booked ETH gains instead of letting one asset run unchecked while capital sat idle in the other.
⚠️What didn't work
XRP essentially flatlined (+1.1%), so every rebalance that bought XRP was buying an asset going nowhere for months. The log shows repeated small ETH→XRP sells — May 8–10 alone moved roughly $75–97 each time — redeploying capital into a laggard with no near-term catalyst.
Had XRP kept falling instead of holding flat, this logic would have kept averaging into weakness.
💡 The key insight
Rebalancing doesn’t need both assets to rise. It needs them to diverge.
ETH did all the directional work here — XRP simply had to stay different enough from ETH to keep triggering the 2% band.
The edge came from forced discipline: selling strength into weakness 21 times, mechanically, without waiting for a “better” exit. That’s the entire 1.76% advantage over just holding both coins.
🚩 Watch out for - a potential red flag
This strategy assumes ETH and XRP keep moving at different speeds. If they became perfectly correlated — both up or both down together — the 2% trigger would rarely fire, and returns would converge toward flat buy & hold.
Worse, if XRP had crashed instead of staying flat, every rebalance would have kept feeding capital into a falling asset. Before running this setup, confirm the pair still shows genuine price divergence over the last 30–60 days, not just historical correlation.
Overall Performance Score, Strengths and Limitations
Solid Divergence Harvester
8.94% ROI in ~3.5 months, beating a 50/50 buy & hold by 1.76 points on just $5.73 in fees. Works best when one asset trends and the other lags — not when both move together.
🏆 What this strategy does well
- Beat buy & hold on ROI (8.94% vs 7.18%)
- Low fee drag — 21 trades cost under $6 total
- Fast 1m time check kept pace with ETH's climb
- Fully systematic — no discretionary timing
🚫 Limitations
- 1.76% edge is modest relative to ETH's standalone +13.9% move
- Underperforms holding ETH alone in this specific window
- Depends entirely on continued price divergence
- No drawdown metric reported — session-level risk unclear
Quick Takeaways
- Rebalancing profits from divergence, not direction
- Tighter triggers react faster but cost more in fees
- 2% + 1m found the best trades-to-profit ratio here
- The edge over buy & hold was real but modest (+1.76%)
- Works worst when both assets move together
How did passive HODL compare?
If you had simply put $2,000 into ETH and $2,000 into XRP on March 1 and held, here’s how it compares:
Running the bot instead of holding: $357.68 minus $287.20 = $70.48 extra profit in about 3.5 months.
That’s a 1.76-point edge for accepting $5.73 in fees and 21 automated trades – a reasonable price for systematically selling strength into weakness.
Before you run this playbook, check these off.
Use this as your go/no-go checklist before deploying this exact parameter set.
🧠 Market Suitability Matrix
| Market Condition | Rating | Strategic Notes |
|---|---|---|
| Divergent Trend (one leads, one lags) | ★★★★★ Excellent | Exactly what happened here |
| Both Assets Trending Together | ★★☆☆☆ Risky | Trigger rarely fires, no edge |
| Sideways / Range-Bound Both | ★★★☆☆ Moderate | Small drift, modest trades |
| High Volatility, Same Direction | ★★★☆☆ Moderate | Trades happen but edge shrinks |
| One Asset Crashing | ★★☆☆☆ Risky | Keeps buying the falling asset |
| Strong Bull Run (Both Up) | ★★☆☆☆ Risky | Rebalancing caps upside on winner |
| Very Low Volatility / Flat Both | ★☆☆☆☆ Poor | Few triggers, dead capital |
How to tune this playbook for different scenarios.
Disclaimer: All data sourced from CryptoGates Rebalance Backtest Bot. Results are historical simulations using Binance 1-minute OHLCV data. Past backtest performance does not guarantee future live trading results. DYOR.
Battle-Test Your Strategy
Before the Market Does.
Eliminate guesswork with institutional-grade backtesting for DCA, Grid, and Rebalance bots. Real historical data. Real-world results.


