I spent $500 on Jito tips over 30 days so you don't have to.
Here's the raw data.
The Setup
I ran 200 test trades across 30 days using a dedicated script that logged:
- Jito tip amount (SOL)
- Priority fee (microLamports)
- Time to first confirmation (ms)
- Success/fail
- Pool liquidity at time of trade
The Results
Normal Market Conditions (mempool below 60% capacity)
| Tip (SOL) | Priority Fee | Success Rate | Avg Confirm Time |
|---|---|---|---|
| 0.0005 | 1000 | 72% | 2.4s |
| 0.001 | 5000 | 94% | 0.8s |
| 0.002 | 10000 | 98% | 0.4s |
| 0.005 | 25000 | 99% | 0.3s |
Sweet spot: 0.001 SOL + 5000 microLamports. Above 0.002 SOL, returns diminish hard.
Pump.fun King of the Hill
When KOTH is active and pumping daily volume > $2M:
| Tip (SOL) | Priority Fee | Success Rate |
|---|---|---|
| 0.001 | 5000 | 41% |
| 0.005 | 10000 | 68% |
| 0.01 | 25000 | 82% |
| 0.03 | 50000 | 89% |
| 0.08 | 100000 | 91% |
Sweet spot: 0.01 SOL + 25000 microLamports. Beyond 0.03 SOL you're tipping more than you save on failed tx fees.
Raydium Migration (first 60 seconds)
This is where real money is made and lost.
| Tip (SOL) | Priority Fee | Success Rate |
|---|---|---|
| 0.005 | 10000 | 22% |
| 0.01 | 25000 | 45% |
| 0.03 | 50000 | 67% |
| 0.05 | 75000 | 78% |
| 0.08 | 100000 | 83% |
Sweet spot: 0.03-0.05 SOL. Anything above 0.08 is waste.
The Hidden Variable Nobody Talks About
Tip size matters less than your terminal's base latency. Here's why:
If your terminal takes 400ms to submit a transaction (looking at you, standard RPCs via Phantom), even a 0.08 SOL tip can't save you. The block slot is already taken by the time your tx reaches the validator.
Compare:
- 400ms terminal + 0.08 SOL tip: 67% success | ~$5.60/tip waste on fails | effective cost $8.40 per successful trade
- 127ms terminal + 0.01 SOL tip: 89% success | ~$0.70/tip waste | effective cost $0.79 per successful trade
The faster terminal wins by 10x, not because of tips but because it submits before the slot fills.
My Recommendation
For normal trading: Use a terminal with sub-150ms execution and let it auto-adjust tips. Manual tip setting in Phantom is gambling.
For sniping migrations: You need both speed AND tips. Don't trade a hot migration through a browser wallet. You will lose.
For KOTH: Watch the mempool, not the chart. When pending tx count drops below 200, that's your window. Tip 0.01 SOL, not 0.005.
Raw Data
Full dataset with all 200 trades including timestamps and pool addresses is on my GitHub: solana-trading-terminal-benchmarks
More detailed guide with dynamic fee strategies: Optimal Jito Tip Settings Guide
Disclosure: Some links in this post are affiliate links. I earn a fee if you sign up through them at no extra cost to you. All benchmarks are real and timestamped.
Top comments (0)