Layer 2 / Scaling · Crypto deep dive · 2026-07-04
Celestia TIA
Accumulate on dips; scale in rather than chase. Accumulation zone: dips toward $0.37973–$0.3989 (the 50/200-day support shelf).
Price
At a glance
The four pillars
Each scored 0–100 relative to the Layer 2 / Scaling sector (not absolute). Composite weights are sector-specific.
Valuation
TIA has the only real income statement in crypto — sequencer gross profit (fee revenue − data-availability cost). It trades at a P/Fees of 12724.7×. The decisive question is value-capture: most L2 tokens are governance-only and do not own the sequencer profit — you are buying an option on the fee switch and on Ethereum's rollup roadmap. Large unlock overhangs make Tokenomics the dominant pillar here. On the Value pillar it screens rich versus its sector (26/100 sector-relative).
Tokenomics & dilution
Float is 80% of FDV — a moderate share of supply still to unlock; watch the vesting calendar. Uncapped/emissive supply — net issuance must be netted against any staking yield to judge real yield. Unlock dates are contractual and public — the closest thing crypto has to an earnings calendar, and the market systematically underprices them. (Live unlock-calendar integration is the next data layer.)
Network & moat
Value capture: INDIRECT — value accrues via staking/collateral/gas demand. Moat comes from liquidity, integrations, and switching costs — real but contestable; in crypto, forks and incentive wars erode moats faster than in equities.
How to invest
Accumulate on dips; scale in rather than chase. Accumulation zone: dips toward $0.37973–$0.3989 (the 50/200-day support shelf).
Honest limits
Crypto is one liquidity trade in many costumes — in a liquidity drain every sector correlates toward 1, so this rating is relative selection within crypto, not diversification. The macro regime gate sizes total crypto exposure; the score picks within it. Reflexivity breaks models faster than in equities (usage → price → usage). Regulatory and hack tail-risks are sized with flags, not forecast.