SIMULATED
Why SlowCook
On a bonding curve every buy raises the price, so the first buyer pays the least. In a sealed uniform-price auction every winner pays the same price.
Same 200 wallets through both mechanisms, 1,000 runs
- Bonding curve (pump.fun-style)
- SlowCook
- SlowCook, shaded bids
SIMULATED: synthetic agents, 1,000 runs. Bars show the median, whiskers the 5th to 95th percentile. In SlowCook the insiders get the strongest strategy we built: they know all retail demand and bid one tick above the price retail alone would set.
All results
| Median (5th to 95th percentile) | Bonding curve (pump.fun-style) | SlowCook | SlowCook, shaded bids |
|---|---|---|---|
| Insider share of supply (bots and bundler) | 40.3% (39.0% to 41.6%) | 10.7% (6.9% to 14.3%) | 11.2% (7.5% to 13.9%) |
| Bundler share | 32.4% (31.7% to 33.2%) | 8.0% (5.1% to 10.7%) | 8.3% (5.5% to 10.3%) |
| Insider return after dumping at T+1 | 437.3% (384.9% to 500.9%) | -9.2% (-19.3% to -0.1%) | 7.6% (-3.3% to 21.1%) |
| Insider profit (NEAR) | 59.02 (51.91 to 67.42) | -0.92 (-2.38 to -0.00) | 0.70 (-0.38 to 2.02) |
| Retail aggregate profit or loss (NEAR) | -37.93 (-43.36 to -33.31) | -7.00 (-15.79 to -0.16) | -2.10 (-9.00 to -0.59) |
| Median retail return | -57.6% (-64.7% to -48.3%) | -8.3% (-19.3% to -0.0%) | -0.7% (-9.3% to -0.0%) |
| Price drop caused by the dump | 81.0% (79.3% to 82.6%) | 17.8% (9.9% to 28.0%) | 28.5% (21.3% to 37.6%) |
What it fixes
Being first, splitting into 20 wallets, or bidding higher buys no better price. Unsold supply is burned instead of flooding the pool. The creator's tokens stay locked for at least 7 days, then vest, and are released only while the price holds above clearing.
What it does not fix
Deposits are public: they cap a bid without revealing it. One person can still use many wallets. The treasury slows a sell-off and cannot stop one. Not revealing is an option priced by a 5% penalty.