MEV, sandwiches and how Axiom protects your fills

Maximal extractable value is the profit available from choosing the order of transactions in a block. In practice, for a trader buying a fast-moving token, it shows up as a worse price than the one you saw.

The classic form is a sandwich: your buy is seen while pending, a transaction is placed in front of it to push the price up, and one behind it to sell into the move you just made. You still get filled. You just pay for the privilege.

On calm markets this is noise. On a launch with thin liquidity and wide slippage tolerance, it can cost more than the trading fee several times over.

Axiom exposes three MEV protection modes. Off applies no protection: the fastest path, fully exposed. Reduced submits through Jito, which limits the exposure of your transaction in the public mempool. Secure routes through whitelisted validators and is the recommended setting.

The trade-off is the same one that exists everywhere in execution: protection has a cost, usually in speed or inclusion certainty. Most traders should sit on Secure and stop thinking about it.

Traders who deliberately choose Off are usually racing an event where inclusion in the very next block outweighs the risk of being sandwiched. That is a defensible choice made explicitly, not a default worth drifting into.

Slippage tolerance is the worst price you agree to accept. It is not a fee and it is not a suggestion: it is the width of the window an extractor has to work in.

Set it too tight on a fast market and transactions fail. Set it wide on a thin one and you have pre-authorised a bad fill. The honest answer is that it should change with liquidity, not stay at one comfortable number forever.

Slippage and MEV mode work together. Secure mode narrows who can see your transaction; slippage limits what they can do if they see it anyway.

Solana blocks are contested. Priority fees and bribes decide how quickly your transaction is included, and on a busy block they decide whether it lands at all.

Both are adjustable in Axiom, and both come with smart auto-recommendations that balance speed against cost. That matters because the correct amount is a moving target: what gets you into a quiet block is a fraction of what a contested migration demands.

Underbidding is not free. A transaction that does not land during the window you wanted it in is a full loss of the opportunity, not a saving.

Execution protection is worthless if the underlying custody model is weak. Axiom is non-custodial: you control your funds, trades settle fully onchain through direct integration with decentralized protocols, and the security stack uses secure key management and encryption delivered in partnership with Turnkey.

Your self-custody wallet comes with a recovery phrase you can export into Phantom, Rabby or Solflare at any time. Nobody from Axiom will ever ask you for it.

Those two things together are the actual security posture: your keys are yours, and your transactions do not have to travel through the most exposed path available.

For most traders: MEV protection on Secure, priority fee and bribe on smart auto-recommendations, and slippage set per market rather than globally.

Then use the tools that reduce how often you need a fast fill at all. Limit orders execute at a price you chose in advance. Sniper Sell pre-sets an exit at migration. Both remove the panic click, which is where bad fills concentrate.

Protection is not about eliminating risk in these markets, which is not possible. It is about not donating the part of the trade you could have kept.