BTC Price Clamps
Test how sensitive the saved BTC High Activity Continuation 1000 strategy is to its global Kalshi buy-price clamps. All other settings stay exactly as saved: the Coinbase BTC-USD volume>=10000 BTC units gate, the 5m/15m momentum thresholds, flat-position and order-count guards, the under-3-markets cap, the 180-600s expiry window, the 3-cent spread limit, the 0.30-0.70 purchased-side ask band, 5-contract immediate-or-cancel entries, the 5m-turn exits, the -25 dollar P&L exit and 25 dollar loss halt, max position 5, spend cap 5 dollars per market, one entry per market, max loss 25, and the 30-second loop. Only the global price floor and price ceiling vary, across 10 x 10 = 100 cells, including the saved baseline 0.05 / 0.95. These clamps move buy limit prices and can change whether and how orders fill; they are not pure signal filters. The 1,000 dollar reference portfolio is for reporting only.
Historical research only. Not investment advice.
Top Strategy Variants
Bottom Strategy Variants
BTC High Activity Continuation 1000 — Kalshi Buy-Price Clamp Sensitivity
Short disclaimer
This is historical simulation research on Kalshi crypto markets. It is not investment advice, not a trade recommendation, and not a claim about future results. All figures are backtest outputs from a fixed reference portfolio of $1,000 used for reporting only.
Intro / thesis
The saved BTC High Activity Continuation 1000 strategy on Kalshi is a momentum continuation play on the KXBTC15M series. It reads Coinbase BTC-USD closed-candle feeds, waits for a 5-minute change of at least +0.10% and a 15-minute change of at least +0.20% with 24-hour volume of at least 10,000 BTC units, and then buys YES on the Kalshi market only if the ask sits between 0.30 and 0.70, the spread is 3 cents or tighter, expiry is 180 to 600 seconds out, and fewer than three markets are open. It has a mirror-image NO entry for downside momentum and exits on a 5-minute turn, a $25 unrealized P&L loss, or via a $25 max-loss halt.
The question here was narrow: how much does the strategy care about the two global Kalshi buy-price clamps, price_floor and price_ceiling? Everything else — the Coinbase volume gate, the 5m/15m thresholds, the flat-position and order-count guards, the under-3-market cap, the 180–600s expiry window, the 3-cent spread limit, the 0.30–0.70 purchased-side ask band, the 5-contract IOC entries, the 5m-turn exits, the -$25 P&L exit and $25 loss halt, the 5-contract position cap, $5 spend cap per market, one entry per market, and the 30-second loop — stayed exactly as saved. Only the floor and ceiling moved, across 10 × 10 floor/ceiling values, for 100 cells in all. The saved baseline of 0.05 / 0.95 is one of those cells.
One framing point matters here: these clamps are not signal filters. They move buy limit prices and change whether and how orders fill. A tighter floor or lower ceiling can prevent a fill entirely, or fill at a different price, which changes both the trade count and the P&L, even though the entry logic itself never changed.
Variant and strategy explanation
Ten floor values (0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45, 0.49) were crossed with ten ceiling values (0.50, 0.55, 0.60, 0.65, 0.70, 0.75, 0.80, 0.85, 0.90, 0.95). Every one of the 100 cells completed with no failures. The reference $1,000 portfolio is a reporting convention only, so ROI figures are scaled P&L against that notional, not a return an account would have actually booked.
Each successful variant is saved as a runnable Turbine strategy, so any cell in the grid can be pulled up, inspected, and run against the same DSL that produced its results. That is the practical output of a sweep like this: not just a ranking, but a reproducible set of strategies keyed to specific floor and ceiling values.
Top results
The best cell was floor 0.35 / ceiling 0.60, with $34.92 total P&L over 133 trades, a 39.4% win rate, a max drawdown of -$4.15, and a Sharpe of 0.45. On the $1,000 reference that shows as 698.4% ROI, but the dollar figure is the honest one for a strategy with a $5-per-market spend cap and a 5-contract position limit. This cell is saved as btc-price-clamps-b39d0736d469.
The next tier clusters tightly below it. Floor 0.30 / ceiling 0.60 came in at $30.98 over 139 trades (Sharpe 0.46, drawdown -$4.31), and floor 0.25 / ceiling 0.60 at $30.27 over 141 trades (also Sharpe 0.46, -$4.31). Floor 0.35 / ceiling 0.65 gave back a bit at $34.48 over 155 trades with a slightly better win rate of 40.3% but a deeper drawdown of -$5.27 and Sharpe 0.44. Ranks 5 through 8 all sit near $28.50–$29.50 on ceilings of 0.60, with floors ranging from 0.05 to 0.20.
The marginal averages tell a coherent story. Mean P&L by floor peaks at 0.35 ($30.44) and falls off sharply in either direction, down to $4.89 at the 0.49 floor. Mean P&L by ceiling peaks at 0.60 ($26.42), with a rough plateau from 0.60 to 0.75 before dropping off toward 0.50 ($14.56) and settling near $22 across 0.80–0.95. The ceiling of 0.60 shows up in every one of the eight best cells.
Bottom results
The worst cell is floor 0.49 / ceiling 0.50, which lost $2.96 on just 16 trades with a 0% win rate and Sharpe -0.27. That is a near-pinched clamp window — only prices between 49 and 50 cents can fill — so very few entries went through. This cell is saved as btc-price-clamps-1f73c712bfea.
Nearly every other poor cell shares the floor 0.49. Ranks 93 through 99 in the bottom list are all floor 0.49 with ceilings from 0.55 up to 0.95, with P&L between $2.36 and $8.32. The floor-0.49 row of the grid carries a mean of $4.89, the lowest of any floor, and it is the only floor row where some cells are barely profitable.
There is a discrepancy in the reported metrics worth flagging: the floor 0.49 / ceiling 0.50 cell shows -$2.96 total P&L with a max drawdown of exactly -$2.96, which is consistent, but the floor 0.49 / ceiling 0.80 cell shows $3.69 P&L with a max drawdown of -$8.98, meaning the strategy lost far more at some point than it ended down. That is possible in principle for any strategy, but the specific path-level accounting that produced this pair of figures is not resolved by the data we have, so we are not going to invent an explanation for it. The same caution applies to the floor 0.49 / ceiling 0.90 versus 0.95 cells, which produce identical P&L, trade count, and win rate.
Conclusion
A few things come through clearly. The sweep ran cleanly (100 of 100 cells), the money-losing corner is narrow (almost entirely floor 0.49), and the ceiling of 0.60 shows up in every top cell. The best cell at 0.35 / 0.60 is not a dramatic outlier — it sits about $4 above the next three — which suggests a modest, fairly stable surface around the 0.30–0.35 floor and 0.60 ceiling region rather than a lucky peak. The neighborhood degradation statistic is 0.156, i.e. the best cell's neighbors give back about 15.6% of its edge, which is mild but not zero, and it argues against treating 0.35 / 0.60 as a precision target.
The permutation test is the check that matters most for the feed-timing claim. It scrambled the Coinbase BTC-USD edge feed while leaving the market price series intact, and it reports p = 0.010 across 100 permutations, with the real best cell's $34.92 beating 99.0% of the re-swept scrambled versions. That is favorable for the edge feed's timing specifically. It does not test the price-based conditions, because prices were not permuted, and the deflated-Sharpe check (0.96 against an expected-max-Sharpe benchmark of 0.23) is a favorable selection-noise probability but does not establish profitability on its own and cannot substitute for out-of-sample evidence.
Taken together: the strategy is meaningfully sensitive to the clamps, the saved baseline of 0.05 / 0.95 is on the softer end of its own surface (roughly $24.84 vs the $34.92 best), and the profitable region is a real but narrow band. Anyone running this should treat the specific 0.35 / 0.60 cell as the best of 100 candidates on one historical sample, not as a forecast. Each winning variant is saved as a runnable Turbine strategy for reproducibility, and future readers should run those saved variants out of sample before drawing any conclusion beyond what the historical grid supports.
Long disclaimer
This document is research and commentary only. It does not constitute investment advice, a recommendation, a solicitation, or an offer to buy or sell any security, contract, or prediction-market position, and it is not directed at any particular person or jurisdiction. Nothing here should be relied on in making an investment or trading decision.
All results described are historical simulation outputs produced by a rule-driven backtest engine. The strategy was not traded live. The $1,000 reference portfolio is a reporting convention used to express P&L consistently across variants; it is not a real account, not a capital allocation, and not a statement about returns an investor would have realized. Actual trading outcomes depend on liquidity, order-book depth, queue priority, fills, fees, settlement rules, and exchange-level mechanics that a backtest necessarily simplifies. The price_floor and price_ceiling parameters affect
About this Kalshi strategy backtest
What question does BTC Price Clamps investigate?
Test how sensitive the saved BTC High Activity Continuation 1000 strategy is to its global Kalshi buy-price clamps. All other settings stay exactly as saved: the Coinbase BTC-USD volume>=10000 BTC units gate, the 5m/15m momentum thresholds, flat-position and order-count guards, the under-3-markets cap, the 180-600s expiry window, the 3-cent spread limit, the 0.30-0.70 purchased-side ask band, 5-contract immediate-or-cancel entries, the 5m-turn exits, the -25 dollar P&L exit and 25 dollar loss halt, max position 5, spend cap 5 dollars per market, one entry per market, max loss 25, and the 30-second loop. Only the global price floor and price ceiling vary, across 10 x 10 = 100 cells, including the saved baseline 0.05 / 0.95. These clamps move buy limit prices and can change whether and how orders fill; they are not pure signal filters. The 1,000 dollar reference portfolio is for reporting only.
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 100 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.
This report is generated from historical simulations. Backtests can be wrong or incomplete, and live trading can differ materially because of liquidity, fees, slippage, latency, market resolution, outages, and data quality. Do your own review before running any strategy.