MEV on Solana is the value a block producer or a bot extracts by choosing what order transactions execute in. Solana has no public mempool, so the Ethereum-style scramble over pending transactions does not exist — instead MEV flows through private transaction feeds shared with block producers, which makes it less visible without making it less real. Sandwich attacks, where a bot buys immediately before your swap and sells immediately after, extracted somewhere between $370M and $500M from Solana users over a recent sixteen-month period, with wide multi-slot attacks accounting for around 93% of that activity.
Key Facts
- Solana has no public mempool — transactions go straight to the scheduled leader, not to a shared waiting area.
- Jito ran a public mempool from November 2022 and shut it down in March 2024 over sandwich concerns.
- Sandwich bots extracted roughly $370M-$500M from Solana users over a recent sixteen-month window.
- Wide, multi-slot sandwiches account for about 93% of sandwich activity, taking over 529,000 SOL in a year.
- A 2025 academic study documented over 500,000 sandwich instances on Jito infrastructure.
- Your slippage tolerance defines the exact ceiling on what a sandwich can profitably take.
- Jito and Marinade have blacklisted validators identified by their sandwich rate.
How a Sandwich Actually Works
Three transactions in one ordering. The bot buys the token first, pushing the price up. Your buy executes at that inflated price, pushing it higher still. The bot sells into the price your trade created, pocketing the difference. You receive fewer tokens than quoted and nothing appears broken — the swap succeeded, just at the worse end of what you permitted.
The bot’s profit is bounded by exactly one variable: how far the price can move before your swap reverts. That is your slippage tolerance, which makes the setting a direct statement of how much you are willing to lose to this. A 20% tolerance on a thin pool is an open invitation with a number attached.
Why No Mempool Does Not Mean No MEV
| Ethereum model | Solana model | |
|---|---|---|
| Where pending transactions sit | Public mempool, visible to all | Sent directly to the scheduled leader |
| How bots see your trade | Read the mempool | Private orderflow shared with block producers |
| Visibility of MEV | High — anyone can measure it | Low — happens inside private feeds |
| Main attack shape | Single-block sandwich | Multi-slot wide sandwich, ~93% of activity |
| Who profits | Searchers and builders | Bots plus a minority of validators |
Jito’s public mempool was the closest Solana came to the Ethereum model, and it was closed in March 2024 precisely because it made sandwiching too easy. The result was a shift to private feeds — less sandwiching in aggregate, but far harder to observe. The 2025 measurement study that documented over 500,000 sandwich instances exists because researchers had to reconstruct the activity from block data, since there is no public queue to watch.
The Validator Angle
A leader chooses the order of transactions inside its own slots. That discretion is what makes sandwiching possible at all, and a minority of operators have run modified software or shared orderflow with searchers in exchange for a cut. The ecosystem’s response has been social rather than protocol-level: a “sandwich rate” metric identifies suspect operators, and Jito and Marinade have removed or blacklisted stake from validators found doing it, which hits the offender where it matters since validator economics depend on delegated stake.
That is a meaningful deterrent and not a solution. As long as ordering is a leader’s prerogative, the incentive persists, and treating sandwich risk as reduced rather than eliminated is the correct posture.
What Actually Protects You
Four things, in order of effectiveness. Keep slippage tolerance as low as reliably fills — this caps the extractable amount directly. Route through an interface offering MEV protection: Jupiter and similar send transactions through channels designed to avoid exposure to sandwich-friendly feeds. Trade deeper pools, since thin liquidity makes the price easier to push. And avoid announcing large trades in public channels before executing them, which is the lowest-tech form of frontrunning and still works.
Checking afterwards is also worth doing. On Solscan, open the block containing your swap and look at the transactions immediately before and after it — a buy and a sell of the same token by the same wallet wrapped around your trade is a sandwich, plainly visible once you know the shape.
MEV Exposure When Trading $DOLAN
DOLAN Duck ($DOLAN) is an ordinary SPL pair with a fixed 98.3M supply and roughly 10,700 holders, which means it carries exactly the same MEV exposure as any Solana token — no more, no less, because sandwiching targets the pool and the transaction, never the project. What varies between tokens is only pool depth and how much slippage traders habitually set, both of which are under your control rather than the token’s. The practical rule holds across every pair on the chain: low tolerance, protected routing, and size relative to depth. That is the same operational discipline as keeping a memecoin position safe generally, applied to the one risk that shows up on every single swap.
MEV is the value extracted by controlling the order transactions execute in. On Solana it flows through private orderflow shared with block producers rather than a public mempool, making it harder to observe.
A bot buys the token immediately before your swap, letting your trade push the price higher, then sells into that price. You receive fewer tokens than quoted and the bot keeps the difference.
Yes. Bots receive transactions through private feeds shared with block producers instead of reading a public queue. Sandwiching still extracted hundreds of millions of dollars over a recent sixteen-month period.
Between roughly $370 million and $500 million over a recent sixteen-month window, with wide multi-slot attacks making up about 93% of activity and over 529,000 SOL taken in a year.
Keep slippage tolerance as low as reliably fills, route through an interface with MEV protection, trade deeper pools, and avoid announcing large trades before executing them.
Open the block containing your swap on an explorer and check the transactions immediately before and after. A buy and a sell of the same token by the same wallet surrounding your trade is a sandwich.
A minority. Ordering is a leader’s prerogative, so some operators have shared orderflow with searchers. Jito and Marinade now track a sandwich rate and have removed stake from validators identified this way.