How Pulse finds runners before they migrate

New Pairs shows brand new launches. Final Stretch shows tokens approaching 100% of their bonding curve, which means migration to Raydium is close. Migrated shows tokens that have completed that move and are trading on Raydium.

Reading them as three separate lists is the most common mistake. They are three stages of the same object. A token you first see in New Pairs may be in Final Stretch eight minutes later and in Migrated a minute after that, and your job across those three windows is different each time.

In New Pairs you are triaging. In Final Stretch you are deciding size and whether to arm a migration snipe. In Migrated you are trading an actual market with real liquidity and a real book to read.

Almost everything in New Pairs is untradeable. That is not pessimism, it is arithmetic: hundreds of tokens launch in an hour and the overwhelming majority never build liquidity worth an exit.

So the filters here should be brutal. Set a floor on liquidity so you never look at a market you cannot leave. Cap dev holding % and top 10 holders % to remove launches where a handful of wallets can end the trade at will. Watch snipers % and insiders % for supply that was allocated rather than bought.

What survives is a shortlist, not a signal. The point of a hard filter here is that you spend attention on ten rows instead of two hundred.

Final Stretch is the highest-value table in Pulse because it is the only one that tells you what is about to happen rather than what already did. A token near 100% of its curve is minutes or seconds from migrating, and migration resets the liquidity picture.

This is where bundle % earns its place on the screen. If coordinated buys have been landing four or more to a block, the run up the curve was probably not organic demand, and migration is more likely to be an exit than a beginning. The Bundle Checker gives you the detail behind the number.

It is also where you arm migration sniping. Sniper Buy fills the instant migration completes, executed on Axiom's custom nodes, and Sniper Sell can pre-set an exit that fires at migration in case the move is fake. Both decisions are easier to make here, with time, than in the two seconds after the event.

Once a token is live on Raydium you have something to analyse. Volume is real, transactions are real, and the holder base has behaviour rather than just allocation.

This is where Trader Scan takes over from filters. For any token it shows each wallet's total bought, total sold, current balance, realized PnL and holding duration, which is enough to tell an accumulating book from a distributing one. A market where the largest wallets have not sold anything behaves very differently from one where they are trimming into every push.

Explore and Trending are the natural next stop. Once a token has history, ranking it by trading activity over a chosen timeframe compares it against the rest of the market instead of against its own first ten minutes.

The fourteen filters are age, transactions, number of buys and sells, top 10 holders %, dev holding %, snipers %, insiders %, bundle %, holders, liquidity, volume, market cap and pro traders %. Nobody uses all fourteen aggressively at once, and trying to produces an empty screen.

A workable approach is three layers. Layer one is exit safety: liquidity and volume floors. Layer two is supply integrity: top 10 holders %, dev holding %, snipers %, insiders % and bundle %. Layer three is interest: transactions, buys and sells, holders and pro traders %.

Set layer one and two once and leave them alone. They encode your risk tolerance and should not change with your mood. Layer three is the dial you actually turn during a session, because how much activity counts as interesting depends entirely on how busy the market is that hour.

Discovery only pays if execution is attached to it. Every Pulse row carries a lightning-bolt Quick Buy button that fires at your preset size with your slippage and MEV protection mode already applied.

If you would rather not use the mouse at all, hotkeys buy a preset SOL amount or sell a percentage of holdings while your eyes stay on the feed. On a fast tape, that difference compounds across a session.

That is the whole design intent behind Pulse. It exists to replace laggy launchpad sites where the feed and the order live in different places, because in migration markets the handoff between them is the trade.