---
name: uberm-framework
description: "The UM framework: @0xuberM's method for finding plays early, his sector view and his entry rules, from the Flow State ep.6 X broadcast of 19 Sep 2026, speaker-separated so only his own words are filed. Ben ordered it folded into every DD alongside [[okay-lets-ride]]. Read before any DD, token screen or contract-address paste."
sources: [cowork]
aliases: [UM, uberM, 0xuberM, um framework, flow state, lets ride and um]
---

# The UM framework

- [stated] Ben, 19 Sep 2026: "When I say we do due diligence, you will then apply the 'let's run it' approach as well as the UM approach, and this is how you evaluate coins going forward." Basic on-chain facts and deployer forensics stay as they are. Whether the coin has POTENTIAL, where it is in its life, and whether it fits the narrative is judged by [[okay-lets-ride]] PLUS this file.
- [stated] Ben, 19 Sep 2026: "0xUberM is the same person as UM. He is the focus."
- [stated] Ben, 19 Sep 2026, CORRECTION HE CAUGHT: my first pass mixed up who said what on the broadcast. "The guy who talked about ORBIO is not UM." He also asked for voice detection, five iterations, and for the framework to produce ACTIONS, not just a summary.

## SOURCE AND HOW IT WAS SEPARATED
X broadcast https://x.com/i/broadcasts/1XxygwEPLbNGM, "Flow State" episode 6, 91 minutes. Transcribed with faster-whisper, then SPEAKER-SEPARATED with Resemblyzer voice embeddings and centroid matching against anchor windows. UM speaks from 00:22:51 to about 01:11; before that it is the hosts. Method that worked, reusable: build a centroid per speaker from windows you are certain about, then cosine-match every segment, rather than unsupervised clustering, which collapsed into one blob.
**NOT UM, do not attribute to him:** the ORBIO stance (00:20:38 to 00:22:35), "Stonkfun is Solana Foundation's way of staying in the game" (00:20:08), the "what is the meme paired to, does it have a community outside crypto" question (01:01:27 to 01:03:04), and "the existing winners of long have a minimum 10x, maybe 100x" (01:02:54). Those are a host. UM endorsed the pairing point afterwards and added his own liquidity argument to it.

## WHO HE IS, in his own words
- Started trading crypto 2024, did well that cycle, stayed through the bear, "paid the bear market tuition".
- "Tend to be pretty early for things. Don't execute those ideas the best because sometimes I am way too early."
- Called AI (Artificial Inu) at roughly $3mm to $5mm, up seven figures, sold some on the way up because the move surprised him.
- Wrote the STONK thesis at the start of August at about $2mm, closed most of the bag on the first run up, made about a quarter million instead of seven figures.
- Does not trade perps. Outside on-chain he only owns Bitcoin.

## LAYER 1 - THE NON-DERIVATIVE TEST. This is the whole method.
> "A lot of the time everything I see is just a derivative of something that's already been done before. And a lot of on-chain trading is that. The reason you do that is so you can pay attention to what's happening. But then eventually something is going to appear in front of you that has elements to it that you never really saw before. And whenever that happens, you have to put all your attention in it."

- Daily trading of derivative coins is not the edge. It is the subscription fee that keeps him in the flow so he sees the non-derivative when it appears.
- Inverted from normal DD: **if you can comp it cleanly, it is a derivative and it is not one of these.** The uncomparable thing is the signal.
> "How do you even cap that? How do you even put a ceiling on it? You can't. Because it's just a new thing. There's nothing to compare it to."
- **No invalidation on these.** "When it comes to stuff like this, I'm just willing to dust everything, because the upside is infinity." He keeps invalidations for ordinary setups and suspends them only here.
- **They are rare.** Asked what he liked that day, having looked every day: "there's really nothing right now."
- The cost is being early. Required behaviour: patience plus evangelism. "Be willing to be patient and just sit on your hands, but also do your part, try to convince as many people as possible that this is it."
- Applied to a coin outside his own style, DRV (Derive, options): "it's in its own lane. It's unique. It has unique elements that you can't really comp it to anything else. Therefore it's worth paying attention to." Demand exists outside crypto and top perp traders (he named base16z) are pivoting to options. DRV Ethereum 0xcDD7Df6367994963293B581B5b27a3844be4234a, also on Robinhood Chain at 0x82B152F345cd4f554d90768E975BEbccA3409754.

## LAYER 2 - THE KINGMAKER QUESTION
He does not ask "is this good". He asks **"do the people who own this venue want this trade to work."**
- STONK thesis, his own words: auto-claims 60% of trading fees to buy and burn STONK; you can only pair pre-IPO and private stocks on Solana on stonk; good setup, can get reflexive very quickly, **"and the only bear case is if Solana doesn't care about it."** It did nothing for a month so he sold. Then Solana did care.
- **He reduces the entire bear case to ONE sentence, and that sentence is about the venue owner's attention.**
- His own phrasing of the kingmaking: "you have pons, which was the first big launchpad runner on Robinhood, then Solana started kingmaking stonk, and I think the next one is gonna be paid... because the interested parties in these ecosystems are going to try to king-make these trades."
- On Robinhood: Vlad said in interviews the chain's focus was on-chain stocks and yield, and hinted at memecoins.

## LAYER 3 - HOW HE FOUND AI, his worked example, in order
1. **Chain owner's stated intent.** Vlad on the record: Robinhood Chain is for on-chain stocks and yield, plus a memecoin hint.
2. **Appetite proven elsewhere on the chain.** Marscoin running, CASHCAT to a large cap. Demand for stock-paired memes is real, not hypothetical.
3. **Name the primitive and its first mover.** Stock pairs started on Robinhood with long.xyz. "They were the first ones to this primitive."
4. **Rule out the rival variation.** "I don't think reflections are the bullish variation of this primitive", because reflections create centralised-exchange listing problems and regulatory ambiguity.
5. **Pick the best proxy, not the best coin.** The coin that ran FIRST and HIGHEST inside the ecosystem that spawned the primitive. AI had run to $15mm and was back at $3mm.
6. **Check the mechanism does what it claims.** AI uses trading fees to buy NVDA on chain and lock it in a pool, so the token owns NVDA stock.
7. **Check the pair is the biggest possible.** "It's paired to the biggest stock in the world."
8. **Conclusion phrasing:** "this is the best proxy for this trade on chain." Then allocate and evangelise.

**The reusable question is "what is the best PROXY for this narrative on chain", never "which coin is cheapest or has the best chart."**

## LAYER 4 - OWN THE CASINO OR PLAY THE TABLES
- Launchpads are "Ponzis disguised as token factories", fees recycling into the pad's own token, "mini Hyperliquids for a more niche thing."
- On high-volatility, low-liquidity pad ecosystems, **own the pad token, not the coins it launches**: "a very small group of people tends to win those setups", so "you're better owning the house, owning the casino."
- **The exception is long.xyz.** There you own the actual coins, because they are "built for sustainability and more extended lifespans": thicker locks, anti-PVP and anti-sniping mechanisms.

## LAYER 5 - THE TRIFECTA
> "Every chain sort of has this group of coins that intertwine together and they collectively go up."
His examples: SOL, LUNA, AVAX. Perp DEXes: Hyperliquid, Lighter, Aster. Now **PONS, STONK, PAID**.
- PAID: captures the fee share a pump.fun launch would otherwise get; 80% redirected to people on X, 20% buys back and burns PAID. He said it had bought back over a quarter million in four days and that he put six figures in. Solana 98kfF7rmsg1QDUEoCqNE7g7M1FdrTt92TEp2CLzypump.

## LAYER 6 - LIQUIDITY IS THE UNLOCK, not a risk metric
- The bear-market problem was that narrative vehicles were illiquid, so only trenchers could play them.
- long.xyz fixed it. AI absorbed a single $2mm market sell without breaking structure.
- Liquidity attracts a different subset of capital: people who never trench, do not care about being early, and will buy far higher with size.
- And it serves the chain owner. His own point, made to Leo at a $5mm cap: "what Long is doing is using these fees to make the pools of tokenized stocks way more liquid... that directly benefits Robinhood, because now they can offer this to customers as an actual legitimate way of being exposed to equities."
- **This inverts the usual instinct. For UM, deep liquidity on a meme is a bullish structural feature, not a sign the easy money has gone.**

## LAYER 7 - ONE MEME PER STOCK, AND IT WILL BE ON LONG. His strongest claim.
> "When it comes to memecoins paired with on-chain stocks, all the winners of that are going to be on long XYZ. Whatever pairs you think will matter on chain, whatever stocks you think will matter on chain, their respective memecoin counterparts will be on long XYZ. I have full conviction on that."
> "So if you think, oh, I like Apple stock or I like Google, whatever it is that you think will be relevant as a narrative, you just buy that respective counterpart on long XYZ."
**This is the layer that produces actions.** It implies a lane map: one definitive meme per tokenized stock. See [[long-lane-map]] for the live map, the contested lanes and the empty ones.

## HIS SECTOR VIEW, next one to two years
- Bull market. Stock-paired memes are the theme for at least the first half of the cycle, "there's still a lot of juice left on the stocker narrative."
- Three places to be: the token factories (PONS, STONK, PAID); the long.xyz ecosystem coins; and infrastructure that benefits from tokenized-equity activity, plus AI and privacy.
- End state: long.xyz becomes one of the biggest holders of tokenized equities on Robinhood, making Nate effectively a top on-chain holder of NVDA, AAPL and GOOG, which opens the door to institutional capital.
- Rotation: "every crypto cycle we always do some pivot to AI in some shape or form." Late 2023, early 2024, late 2024, again this year. Expect the AI rotation AFTER the stocker mania.
- Vlad's incentive (raised by a host, agreed by UM): chain revenue at about 90% margins, millions a day within two months. "That alone should be a strong enough thesis for you to be allocated to the proxy bets."
- Scale (host's point, not his): RH chain does about $1bn to $1.4bn a day; Dogecoin peaked at $60bn to $70bn.

## WHERE HE DISAGREES WITH COULOU
- **Reflections.** UM: reflections are NOT the bullish variation of the stock-pair primitive, because of CEX-listing friction and regulatory ambiguity. Any reflections-based bull case must answer this.
- **ORBIO.** UM did NOT give the ORBIO take on this broadcast; that was a host. Do not put an ORBIO opinion in his mouth. Coulou is long ORBIO.

## HANDLING HIM AS A SOURCE
- He is long AI, the long.xyz ecosystem, PONS, STONK and PAID. He talks his book. Use the method, verify on chain.
- He is honest about his misses (sold STONK at $2mm, sold AI on the way up), which makes his self-reported entries usable.
- By his own account he is early and executes badly. **Copy his entry logic. Do not copy his exits.**