The three amounts, and why they stay separate
People ask for one number and then fund a wallet with it, which is where the trouble starts. The wallet needs to hold three things that behave differently and must not be allowed to borrow from each other.
| Amount | What it does | What happens if it runs out |
|---|---|---|
| Position capital | Goes into trades | You stop trading, which is fine |
| Fee buffer | Pays priority fees, tips, failed attempts | Entries fail for no strategic reason |
| Rent reserve | Funds new token accounts | Entries fail before they begin |
Why starting too small fails mathematically
This is the part that deserves to be said plainly, because the usual advice to start small is well meant and, below a point, wrong.
Fees in sniping are largely fixed per attempt. Base fee, priority fee and tip do not shrink because your position was small. Returns, on the other hand, are proportional: a 40 percent move on a small position is a small number.
| Position size | Fee cost per attempt | Fees as a share of the trade |
|---|---|---|
| Very small | The same as any other | Dominant |
| Small | The same | Large |
| Moderate | The same | Noticeable |
| Large relative to pool | The same | Small, but slippage takes over |
Two things follow. First, there is a floor below which a strategy with genuine edge still loses, purely to fee drag. Second, the floor is not fixed: it moves with how many attempts you make, because fee drag scales with attempts rather than with fills.
Sizing a position against depth
The correct anchor for position size is the depth of the pool you are entering, not the balance of your wallet. A position that looks modest against your account can still be large against a curve holding a few SOL.
- 1Look at the pool, not the balanceYour own order is part of what moves the price against you. The pool decides how much.
- 2Remember the exit is the harder sideYou have to leave through the same depth, usually at a moment when others want to leave too.
- 3Keep any single position a small share of capitalNew launches fail often enough that concentration is not a strategy, it is a single-event bet.
- 4Let the venue change the answerA venue with a long curve and a high graduation threshold supports different sizing than one that completes in minutes.
Sizing the fee buffer
The buffer has to cover attempts, not trades, and that distinction is the whole point. If most of your attempts fail, the buffer is being consumed by transactions that produced nothing.
- Estimate attempts per week honestly, which is usually higher than the number of positions you expect.
- Assume contested launches cost several times what quiet ones do, because priority fees are an auction.
- Add tips only if you intend to use private submission, and only on the launches where it is justified.
- Top the buffer up on a schedule rather than when an entry fails, because by then you have already missed something.
Working out your own number
There is no universal figure, but there is a method that produces yours in about ten minutes, and it is more useful than any number someone else publishes.
- 1Decide your position size from pool depthPick the venues you intend to trade and look at the depth a typical launch has when you would enter.
- 2Decide how many positions can be open at onceMultiply. That is your position capital.
- 3Estimate attempts per weekMultiply by your expected cost per attempt, including tips if you will use them. That is a week of fee buffer.
- 4Hold several weeks of buffer, not oneA single bad week should not stop you trading for operational reasons.
- 5Add a small rent reserveSmall, fixed, and easy to forget until it causes a failure.
- 6Check the floorIf fee drag per attempt is a large share of the expected move on your position size, the answer is not to trade more carefully. It is that this size does not work.
The numbers that feed this calculation
- What a sniper bot costsThe five cost lines the buffer has to cover
- Are sniper bots profitableWhy fee drag hurts small positions most
- How to start snipingWhere funding sits in the seven steps
- Sniper bot settingsPosition size is the setting that controls the others
Questions people ask
These come up in the ticket queue often enough to answer once, properly. If yours is not here, send it over and it probably ends up on this page.
Three amounts kept separate: position capital sized against pool depth, a fee buffer covering several weeks of attempts including failures, and a small rent reserve for new token accounts. A single number does not describe the requirement.
Key takeaways
- Budget three amounts, not one: position capital, fee buffer and rent reserve.
- A wallet funded to exactly the position size will fail, because fees and token accounts are charged on top.
- Fees are fixed per attempt while returns are proportional, so small positions suffer disproportionate drag.
- There is a floor below which a strategy with real edge still loses to fee drag alone.
- There is also a ceiling, where slippage against pool depth replaces fees as the dominant cost.
- Size positions against pool depth and sweep profits out on a schedule.