ETH Entry Band

ETH 15-Minute Entry Band Narrowing: Kalshi Crypto Research Report

Short disclaimer: This is historical simulation research conducted on Kalshi crypto markets. It is not financial advice, and nothing here should be read as a promise or expectation of future results.

Intro / thesis

This report looks at one narrow question on the Kalshi ETH 15-minute series (KXETH15M): does pulling the entry price band closer to the middle of the market, nearer 0.50, improve simulated results compared with the current 0.35–0.70 band, holding every other entry condition fixed?

The reasoning behind the question is simple. A market trading near 0.50 is genuinely uncertain, so if the directional confirmation signals are informative, entering in the middle of the book could capture more upside per win than entering at an extreme where the contract is already near its resolution value. But a tighter band around 0.50 also means fewer qualifying trades and possibly a worse mix of setups. The batch tested 11 band variants over a 30-day window to see which way the tradeoff actually leaned.

Variant and strategy explanation

This was an explicit batch of 11 completed variants, each with its own complete rules, market and window. The base configuration used for context is the same across candidates except for the entry price band:

• Venue: Kalshi, market series KXETH15M, evaluated every 10 seconds.
• Position limit 60 contracts; price floor 0.01 and ceiling 0.99.
• Trading window: America/Los_Angeles, Monday–Friday 09:00–17:00 for entries. Exits are not gated by the schedule.
• Context feeds: Coinbase ETH-USD (price, VWAP 1h, 5m and 15m change, EMA 12 1m, 15m high/low, 1m velocity, SMA 20 1m, MACD histogram 1m), plus BTC-USD and SOL-USD 5m change, refreshed every 5 seconds.

Rules are evaluated top to bottom, and the first rule that matches fires and ends that tick:

1. time_to_expiry <= 10s → sell_all.
2. yes_position_size > 0 and yes_best_bid <= 0.25 → sell_yes.
3. no_position_size > 0 and no_best_bid <= 0.25 → sell_no.
4. unrealized_pnl < -7 → sell_all.
5. position_size == 0, time_to_expiry <= 12m, and all of: edge.eth.change_5m > 0.001, edge.btc.change_5m > 0.001, edge.sol.change_5m > 0.001, edge.eth.price > edge.eth.vwap_1h, edge.eth.price > edge.eth.ema_12_1m, edge.eth.macd_histogram_1m > 0, plus the band check on price and spread <= 0.015 → buy_yes size 50.
6. Mirror of rule 5 on the short side: the same fields flipped negative (< -0.001, price below VWAP and EMA, MACD histogram below 0), same price band and spread <= 0.015 → buy_no size 50.

The only thing that changed across the variants was the price band inside rules 5 and 6. Tested bands were 0.30–0.70, 0.33–0.67, 0.35–0.65, 0.35–0.70 (baseline), 0.38–0.62, 0.40–0.60, 0.40–0.55, 0.42–0.58, 0.44–0.56, 0.45–0.55, and 0.45–0.60. Each successful variant from this batch is saved as a runnable Turbine strategy, so the configurations below can be inspected or re-run directly.

Top results

Ranked by ROI over the 30-day window:

The broad pattern is that narrower bands closer to 0.50 generally produced higher ROI than the baseline, though the two tightest variants (0.45–0.55 and 0.44–0.56) landed mid-table rather than at the top. The 0.40–0.60 band is the best overall result and also posts the highest total PnL. Several tighter bands show better win rates (65.7% for the 0.45–0.55 and 0.44–0.56 variants) and materially smaller drawdowns (-52.68 for both) than the winner, which is a reminder that the ranking metric is doing a lot of the ordering work here.

One detail worth flagging honestly: the 0.57 ROI spread between 0.40–0.55 (rank 2) and 0.40–0.60 (rank 1) comes with only 96 vs 120 trades. That 0.40–0.55 does not produce a lower ROI than several wider bands despite fewer trades is a bit counterintuitive, and the relationship between trade count and ROI across the middle of the ranking is not fully consistent. The source of that discrepancy is unresolved; I won't attribute it to a particular accounting or fill mechanic, because we don't have the per-trade data to establish that.

Bottom results

The weakest variant is the 0.42–0.58 band, which lands at the bottom despite a decent 55.8% win rate and fewer trades (86), with the deepest drawdown in the batch at -115.75 and the lowest Sharpe at 0.40. The wider-than-average 0.35–0.65 and 0.33–0.67 bands also underperform the baseline, which is consistent with the general story that pulling the band inward from the very wide 0.35–0.70 setup helped in most tested variants. The 0.42–0.58 result runs against that grain, and again the mechanism behind why that particular tighter band underperformed is not something I can pin down from the summary data alone. I'm leaving that unresolved rather than guessing.

Conclusion

Within this 30-day simulation on KXETH15M, narrowing the entry band toward the middle of the market generally did improve ROI versus the 0.35–0.70 baseline. The 0.40–0.60 band is the top-ranked variant by ROI and total PnL, with the 0.40–0.55 and 0.38–0.62 bands close behind. Tighter bands around 0.45 also raised win rates noticeably (65.7% for 0.45–0.55 and 0.44–0.56) and cut drawdown nearly in half relative to the baseline, at the cost of fewer trades.

Not every tightening helped. The 0.42–0.58 band came in last, and the wider 0.35–0.65 and 0.33–0.67 variants also trailed the baseline. The ranking is not monotonic in band width, and the reasons for the notable underperformers, as well as for why trade count doesn't track ROI cleanly across the middle of the table, are unresolved with the data available. Treat these rankings as descriptions of one historical window,

About this Kalshi strategy backtest

What question does ETH Entry Band investigate?

Does narrowing the ETH 15-minute entry price band toward the middle (nearer 0.50) improve results versus the current 0.35-0.70 band, when all other entry conditions are unchanged?

Read the analysis alongside the individual variant results. Check which entry, exit, price-band or risk settings changed, which settings stayed fixed, and whether the report documents enough historical coverage to support its comparison. A different parameter value defines a different simulation; it does not establish a new source of live trading returns.

How many strategy variants are included?

This report lists 11 completed and 0 failed strategy variants. Failed comparisons do not count as successful tests with zero profit. Review trade counts and drawdown alongside net P&L, win rate and Sharpe, and read any coverage or execution caveats in the analysis. A result based on very few trades needs particular care, even when its return looks high.

Does a profitable Kalshi backtest predict live performance?

These are historical simulations. Live results can differ because of liquidity, fees, slippage, latency, market resolution, outages and data quality. Selecting the best result after comparing many variants can also overstate how reliably it will perform on new data. Compare nearby settings, inspect losing variants and test on a separate period before considering a live run. The report is research, not a promise of profit or investment advice.

How can I inspect or test a strategy from this report?

Use an available Open strategy link to inspect that variant. Review its rules, market, date window, fill assumptions, fees and trade history before changing parameters or rerunning it. Confirm which data and execution assumptions the selected market actually supports; a replay does not establish that every historical order could have filled live. For the workflow, read the Kalshi and prediction market backtesting guide. The Kalshi backtesting software page explains the product's supported testing workflow.