First live day of surge alerts, 28 Aug 2026: 138 pump.fun surges on 104 mints. Fires with no risk flag were above their fire price an hour later 60% of the time; fires flagged early_buyers_exiting, 6%.
Our token momentum engine went live on 28 August 2026. In its first two hours and 49 minutes (09:29–12:18 UTC) it fired 138 surge alerts on 104 pump.fun mints plus 4 revivals. For the 65 surges old enough to have a one-hour outcome, the result depends almost entirely on the risk flags attached at fire time: fires with no flag were above their fire-time market cap an hour later 60% of the time (6 of 10) and doubled 40% of the time; fires carrying the early_buyers_exiting flag were up 6% of the time (2 of 34). Across all surges the median peak inside the hour was 1.45× the fire price, reached a median 4 minutes after the fire, and the median market cap at the one-hour mark was 0.33× — most surges are a four-minute window followed by a round trip.
This is one morning of data, published on day one because the outcome table is written by the system, not by us, and we would rather show the first numbers than curate a better-looking week. Everything below is as of 2026-08-28 12:19 UTC.
| Metric (28 Aug 2026, 09:29–12:18 UTC) | Value |
|---|---|
| Surge fires / distinct mints | 138 / 104 (30 mints hit more than one tier) |
| By tier: early / strong / breakout | 100 / 30 / 8 |
| Revival fires | 4 |
| Fires with a one-hour outcome | 65 surges, 3 revivals |
| Up at +1h (all surges) | 14 of 65 (21.5%) |
| Median peak multiple inside the hour | 1.45× |
| Median MC multiple at +1h | 0.33× |
| Median minutes from fire to hour-peak | 4 (60% peaked within 5 min, 32% in the same minute) |
| Fires below half their fire MC at +1h | 43 of 65 (66%) |
| Clean fires (no risk flag) | 12 of 138 (8.7%) |
| Fires with at least one tracked KOL buyer | 76 of 138 (55%) |
A surge is a token under 30 minutes old whose market cap runs against its launch market cap — the first sample the engine saw within 90 seconds of birth. On pump.fun a launch is about 28 SOL of market cap by construction, so the multiple is independent of the SOL price; a USD floor keeps micro-launches out.
| Tier | Age | MC floor | Multiple vs launch | Extra condition |
|---|---|---|---|---|
early | ≤ 10 min | ≥ $12K | ≥ 3× | — |
strong |
Building a product on Solana data?
Skip the Geyser pipeline — embed KOL flow, deployer reputation, and the all-DEX firehose over REST, WebSocket, or webhooks. The Business tier (€400/mo) is the self-serve embed license at 500k calls/day — Enterprise adds white-label & redistribution.
Tools mentioned
Live health scores, average ratings, and direct links on MadeOnSol.
Build a deployer sniper bot
Access deployer alerts, bonding stats, and tier data via API. Free tier: 200 requests/day.
import { MadeOnSol } from "madeonsol";
const client = new MadeOnSol({ apiKey: "msk_your_key" });
// Elite deployer alerts with KOL buy enrichment
const { alerts } = await client.deployer.alerts({ limit: 5 });
// Top deployers by bonding rate
const { deployers } = await client.deployer.leaderboard({ tier: "elite" });Building a product on Solana data?
MadeOnSol isn't an RPC or another generic token API — it's Solana memecoin intelligence that's painful to build in-house, pulled from dual-region gRPC shred streams: sub-second from the on-chain event to your app.
Signals you'd otherwise build
KOL & smart-money flow from 2,000+ labeled wallets, deployer reputation, coordination clusters, linked-wallet entity resolution, and an all-DEX firehose.
Embed it in your product
The Business tier (€400/mo, self-serve) licenses you to display MadeOnSol data inside your own product — 500k calls/day, 10 WS + 5 firehose connections. Enterprise above adds white-label & redistribution rights and custom endpoints.
Evaluate first, commit later
Test everything on a free ULTRA key — no commitment. If it fits, we scope volume or white-label pricing.
Keep reading
Analytics
August 1–28, 2026: 1,032,440 launchpad deploys from 223,339 wallets, 18,202 graduations (1.8%), median time to bond 1 minute, median graduate peak $31K. 1,170 wallets minted 40% of the month's tokens.
Analytics
August 1–28, 2026: 533 tracked Solana KOL wallets made 1,133,569 trades across 90,075 tokens, moving 2.11M SOL. Median market cap at first KOL buy was $4,654 and the second KOL followed within 9 seconds.
Analytics
As of mid-July 2026, NOXA was ~75% of Robinhood Chain deploys before dying and relaunching as Pons — now #1 at ~930 deploys/hr. flap.sh leads curve volume, Clanker feeds Uniswap v4, and hood.fun has just 8 lifetime graduations.
Analytics
As of mid-July 2026, only 23% of human alpha wallets on Robinhood Chain are mostly-memecoin — 66% mostly trade non-launchpad assets like stables, tokenized stocks and blue-chips. The chain built for stocks became a meme machine, but the money says otherwise.
Trading
On Robinhood Chain there is no public mempool and gas bidding does not buy ordering — a private FCFS sequencer decides everything by arrival time. Here's why the edge is information and latency, not fees.
Analytics
Updated 28 August 2026: 2,984 spam factories minted 357,378 tokens — 51% of ranked tokens on Robinhood Chain. Biggest factory 109-for-13,799, largest zero 0-for-6,494. Only 252 of 170,440 deployers earn a tier.
Enjoyed this article?
Real-time KOL trades, Pump.fun deployer intel, and 1,200+ ranked Solana & Robinhood Chain tools — free to explore.
Open the KOL Tracker| ≤ 30 min |
| ≥ $30K |
| ≥ 6× |
| ≥ 2× the lowest sample of the last 3 min (still climbing) |
breakout | ≤ 2 min | ≥ $45K | ≥ 8× | — |
Each tier fires at most once per mint and tiers are independent. Every fire must be sustained (hold on the current tick and on a sample at least 10 seconds older; nothing fires before 20 seconds of age) and must pass hard gates that are not flags: liquidity of at least $1.5K and at least 2% of market cap, and buy volume on the tape of at least 3% of the market cap gained. Those gates exist because the first hour of testing produced a same-slot bundle marked to $475K at one second of age with zero parsed buys, and a 22-second-old token showing an $18.3M "market cap" on a few thousand dollars of liquidity. A price mark is not a surge.
A revival is a token with no one-minute trade candle for at least 24 hours that starts trading again, confirmed by the tape (at least 5 buys, $500 of buy volume, and 1.5× the pre-dormancy market cap) rather than by the price.
| Tier | Fires | Median MC at fire | Median multiple | Median age | Median buys | Median unique buyers | Median buy volume |
|---|---|---|---|---|---|---|---|
| early | 100 | $15,073 | 4.2× | 61 s | 94 | 87 | $6,091 |
| strong | 30 | $44,607 | 11.4× | 118 s | 247 | 82 | $15,789 |
| breakout | 8 | $58,008 | 19.4× | 26 s | 21 | 17 | $8,957 |
Breakout is the tier to read carefully: a median age of 26 seconds and 21 buys at a $58K market cap is a token that was bought up before most humans could have seen it. Only one of eight breakout fires was clean.
Risk flags were the norm, not the exception. Of 138 surge fires, 120 (87%) carried sniper_heavy (three or more known sniper wallets in the first-20 cohort), 103 (75%) bundled_launch (three or more early buyers in the same block), 86 (62%) early_buyers_exiting (at least half of a five-plus early cohort already sold at fire time) and 54 (39%) cold_deployer. Only 12 fires had no flag at all.
| Tier | With outcome | Up at +1h | Median peak multiple | p75 peak multiple | Median MC at +1h | Doubled inside the hour |
|---|---|---|---|---|---|---|
| early | 45 | 7 (15.6%) | 1.47× | 1.85× | 0.37× | 10 (22.2%) |
| strong | 16 | 4 (25.0%) | 1.44× | 1.76× | 0.16× | 2 (12.5%) |
| breakout | 4 | 3 (75.0%) | 1.38× | 1.43× | 1.31× | 0 |
| revival | 3 | 0 | 0.93× | 1.19× | 0.50× | 0 |
The tier that fires latest and largest — strong, a median 11.4× launch at fire time — had the worst hour: a median 0.16× at the sixty-minute mark. Breakout's 3-of-4 is four data points. Read the tiers on the scorecard (?stats=1), not on their names.
| Fire carries… | With outcome | Up at +1h | Median peak | Median MC at +1h |
|---|---|---|---|---|
| no flag at all | 10 | 6 (60%) | 1.83× | 1.28× |
| any flag | 55 | 8 (14.5%) | 1.44× | 0.28× |
early_buyers_exiting | 34 | 2 (5.9%) | 1.45× | 0.23× |
without early_buyers_exiting | 31 | 12 (38.7%) | 1.39× | 0.54× |
bundled_launch | 39 | 4 (10.3%) | 1.45× | 0.28× |
without bundled_launch | 26 | 10 (38.5%) | 1.44× | 0.43× |
sniper_heavy | 50 | 8 (16.0%) | 1.44× | 0.25× |
without sniper_heavy | 15 | 6 (40.0%) | 1.61× | 0.50× |
cold_deployer | 22 | 2 (9.1%) | 1.44× | 0.38× |
without cold_deployer | 43 | 12 (27.9%) | 1.45× | 0.28× |
The peak is the same either way — a flagged surge and a clean surge both put in a ~1.45× high inside the hour. What differs is what is left at sixty minutes: 1.28× for clean fires against 0.28× for flagged ones. A surge whose early cohort is already selling at fire time is the cohort's exit, and the alert is telling you so. early_buyers_exiting is the single strongest separator on day one: 5.9% versus 38.7%.
early at 58 seconds, $43,011 (4.4× launch), 220 buys, five tracked KOL buyers, no flags. Peaked at $419,607 inside the hour (9.8×) and was at $21,705 sixty minutes after the fire. The strong tier fired a minute later at $64,276 and saw the same peak.early and strong in the same second at 122 seconds of age, $253,652 (22.1× launch), 192 buys from 59 unique buyers, flagged bundled_launch, sniper_heavy, early_buyers_exiting. Peaked at $1.17M, then $11,432 an hour after the fire (0.045×).early at 248 seconds and $954,705 (83× launch), 407 buys from 60 unique buyers, flagged sniper_heavy and early_buyers_exiting. Peaked at $1.47M and was $11,528 an hour later. The five worst one-hour outcomes were all this shape: a 51–83× multiple at fire, sniper_heavy, and the early cohort already selling.token_surge_events, one row per fire, written by the surge-tracker service off the pricing bus; risk_flags[], buys, unique buyers and deployer tier are captured at fire time in one round-trip and never revised.mc_usd_1h_after, peak_mc_usd_1h_after, low_mc_usd_1h_after filled by a cron every 10 minutes from one-minute candles for fires at least 65 minutes old. Only rows with outcome_computed_at set count toward any outcome figure; all 65 surge outcomes were priced (no token stopped being priced inside the hour). Time-to-peak is the candle minute holding the highest market cap in the 60 minutes after the fire.On the first live day (28 August 2026), 14 of 65 surge fires (21.5%) were above their fire-time market cap sixty minutes later. Fires with no risk flag were up 60% of the time; fires flagged early_buyers_exiting were up 5.9% of the time.
The median fire peaked at 1.45× its fire-time market cap inside the hour, reached a median 4 minutes after the alert; 60% peaked within five minutes. 12 of 65 (18.5%) doubled. The median market cap at the sixty-minute mark was 0.33×.
early is at most 10 minutes old, at least $12K and 3× launch market cap; strong is at most 30 minutes, $30K, 6× launch and still climbing over the last three minutes; breakout is at most 2 minutes old, $45K and 8× launch. Each must be sustained for 10 seconds, pass liquidity and volume-backing gates, and fires at most once per mint.
early_buyers_exiting on day one: surges where at least half the early cohort had already sold at fire time were up an hour later 5.9% of the time, against 38.7% for surges without it. bundled_launch (10.3% vs 38.5%) and sniper_heavy (16.0% vs 40.0%) split the same direction.
GET /api/v1/tokens/surges returns fires newest first with filters (?tier=, ?exclude_flags=, ?only_clean=1, ?min_buys=) and ?stats=1 for the live hit-rate scorecard; the token:surges WebSocket channel and webhook events push them as they fire. PRO tier and above.
The scorecard in this post is regenerated continuously from the same table — GET /api/v1/tokens/surges?stats=1&days=7 prints the current version — and the flags that separate outcomes are filterable at the API (?exclude_flags=early_buyers_exiting,bundled_launch) and on the token:surges stream. Related reading: how much of pump.fun graduates and how fast, the first three minutes of a pump.fun launch and why pump.fun tokens fail in the first hour. Plans are on the pricing page.