Orca volume bot: concentrated depth and where it runs out
Orca Whirlpools concentrate liquidity into price ranges chosen by the people who provide it. Inside those ranges the pool can be deeper than a constant-product pool many times its size. Outside them there is nothing. A volume campaign that assumes smooth depth will execute beautifully until it walks past the edge of the range, at which point every further trade moves price hard and buys very little volume.
Depth exists only where providers put it
This is the defining risk of running a campaign on a concentrated pool, and it is not visible in headline liquidity figures. A pool showing a large total value locked can still be thin at the current price if that capital is parked in ranges far away. Reading the total and sizing trades from it is the mistake that turns a well-funded campaign into a series of expensive, low-volume swaps.
The behaviour to watch for is a run of buys that pushes price toward the edge of the concentrated band. Each one moves price further than the last, so the chart shows an accelerating move while the volume added per SOL spent falls. It looks like momentum and behaves like slippage.
Fee tiers change the arithmetic more than they look like they should
Work it through before choosing a routing target. A campaign routing 300 SOL of volume pays the pool fee on all of it. Moving between adjacent fee tiers can change that line item by multiples, which for most campaigns dwarfs the entire Solana network fee bill. The cheapest tier is not automatically correct, because lower-fee pools often attract less liquidity and therefore offer less depth, but the trade-off should be a decision rather than an accident.
The pool fee is charged by the venue against trade value. The network fee is charged by Solana per transaction regardless of size. They are separate bills with separate drivers, and a quote that merges them into one percentage is hiding which one is doing the damage. The cost breakdown keeps each line apart.
What Orca routing measures at
| Program | Sampled tx | Failure rate | Median fee | Cost per landed swap |
|---|---|---|---|---|
| Orca Whirlpools low sample | 35 | 31.4% | 5,432 | 7,918 |
| Raydium CLMM | 54 | 48.1% | 5,081 | 9,790 |
| Meteora DLMM | 863 | 38.9% | 6,642 | 10,871 |
The clustering in that table is the interesting part. Orca Whirlpools, Raydium CLMM and Meteora DLMM are three separate protocols with different codebases, and they land in a comparable failure band. The common factor is not the protocol, it is the workload: concentrated pools attract routing-heavy and arbitrage-heavy traffic, which contends for the same accounts and produces more losers per block than plain swap traffic does.
Orca appears less frequently per block than Pump.fun programs, so its sample is smaller and the rate carries more uncertainty. When a program falls below fifty sampled transactions we label the row rather than presenting it as precise. Method and current sample size on the measurement page.
Configuring a campaign for a Whirlpool
- Size against the active range, not the headline. Total value locked tells you nothing about depth at the current price on a concentrated pool.
- Stay inside the band. A near-even buy and sell ratio keeps price oscillating within the concentrated region, where every swap gets good execution.
- Prefer many small trades. They also produce more unique maker events per SOL spent, which is what aggregator surfaces count. Makers versus volume.
- Know your fee tier. On a multi-thousand-swap campaign the pool fee is usually the largest single line, larger than anything the network charges.
Volion resolves the routing target from the mint and prices the campaign against the measured failure rate of that specific program, so an Orca-routed token is estimated using the Orca figure rather than a network average. Whether the campaign is a Solana volume bot run on an established pool or a Pump.fun volume bot run against a curve, the fee stays flat and only the network estimate moves.