Archive/2026-10-08Research

Deep Research: Lighter (LIT)

Lighter (LIT) Deep Research: Fee Burst, Value Capture, Variational, RH Chain

As-of: Thu Oct 8, 2026, 15:42–15:52 UTC (11:42–11:52 AM ET). Latest completed fee day: Wed Oct 7 UTC (8 PM ET Oct 6 to 8 PM ET Oct 7). Partial Oct 8 data is used only as a direction hint and is labeled. Label key: Fact = primary source, on-chain data, or Lighter/DefiLlama API · Estimate = derived or secondary · Speculative = narrative or an unconfirmed path.

Bottom line

Question	Verdict	Confidence
Is the Oct 7 fee doubling and the ≈$85k fee-vs-revenue gap a fee-model change or a one-day burst?	One-day liquidation burst, not a fee-model change. The whole ≈$85k gap is liquidation fees routed to the LLP (insurance/liquidity pool): $84.7k on Oct 7 vs $4.8k on Oct 6. DefiLlama books them as fees but, by its own methodology, never as revenue. ETH alone produced $56.3k of it, on a day ETH fell ≈4.7% (≈5.4% at the intraday low). The take rate on trading fees was 0.54 bp, vs 0.50 bp on Oct 6 and a 0.58 bp 30d average, so there is no schedule change. The adapter's only recent commit (Oct 6) added a ProtocolRevenue field and left fees, revenue and LLP logic alone. The same pattern shows up on 20 earlier days since April.	High (≈90%) (Fact for mechanics; Estimate for the label)
What did Oct 7 actually add?	Parent fees rose +$163.4k. Of that, ≈52% was liquidation fees (+$84.4k to the LLP) and ≈48% was trading revenue (+$79.0k: zkLighter +$47.8k, RH +$30.9k). Trading revenue grew +53%, vs HL fees +43%. That is market beta on a volatility day, not a share gain.	High (Fact)
Token value capture	Live, but narrower than the headline. Buybacks (holders revenue) track zkLighter perps revenue almost 1:1 (97.7% over 30d, 99.1% over 90d). RH perps revenue (41% of 30d revenue) shows $0 holders revenue and is reportedly split 50/50 with Robinhood. Liquidation fees (22% of zkLighter perps fees over 30d) go to the LLP, not to LIT. Holders revenue / fees: 29.9% (1d), 41.6% (7d), 46.5% (30d). The annualized buyback of ≈$31.1M is ≈3.5% of market cap and ≈0.9% of FDV.	High on mechanics (Fact); Medium on the RH split (Estimate)
Dominant risk	The team and investor cliff ends about Dec 27–30, 2026 (≈80–83 days away). After that, ≈456.6k LIT/day vests (≈$1.64M/day at $3.59), ≈19× the current buyback pace, and the first year adds ≈67% to today's circulating supply.	High (Fact schedule; Estimate math)
Research Score	55 / 100: a well-built, verifiable perps venue with live buybacks, but a structurally negative supply calendar and a burst-inflated headline	—
Evidence Quality	High for Q1 (adapter code plus Lighter native API) · Medium overall (RH revenue split, ecosystem-reserve releases and holder labels are unverified)	—

Digest status + narrative leads

Field	Value
Status	DIGEST_OK (browser-verified by a sibling agent's real page load; live morning issue)
DIGEST_ASOF	2026-10-08 08:09 ET (12:09 UTC); about 3.5h old at check, about 3.7h old at this writing
Footer	DIGEST_OK \| as_of=2026-10-08 08:09 ET \| meme=3 \| radar=4 \| airdrop=2

Digest lead	Digest confidence	Verification here	Label
"Lighter announced the pre-market listing of the Variational token" (one recap account)	Med	A VARIATIONAL perp exists in Lighter's API (created Oct 7, 11:18 UTC; 5x). As of 15:52 UTC it has zero trades, zero volume and zero OI, is flagged force-reduce-only, and is quoted only by the LLP. No official Lighter or Variational announcement was found.	Market existence: Fact · "announced" claim: Rumor
Variational "32% airdrop" points / "tomorrow" vs "3 more months" timing (posts carry referral links)	Low	Per Variational's docs: 32% genesis airdrop to points holders, unlocked at TGE, TGE "Q4 2026", no date. The timing posts conflict.	32%: Fact · dates: Low
Perp-points farming spreading to Lighter / Arcus / N1 (single self-report)	Low	Consistent with live weekly Lighter RH and Arcus points programs; does not prove volume quality	Speculative
Overnight liquidations ≈$255.9M (≈$224.6M longs)	Med	Consistent with ETH −4.7% / BTC −2.5% on Oct 8 through 15:45 UTC, after ETH −4.7% / BTC −2.7% on Oct 7 UTC. Supports the liquidation-burst read and suggests the Oct 8 bar may also carry elevated LLP fees (direction hint only)	Fact (prices) · Speculative (Oct 8 bar)
HL vs Lighter competition / perps fee changes / RH Chain perps	—	Not on the digest page. Covered from primary data below.	—

Perps market regime (brief)

Venue	Token / price	Mcap	FDV	30d fees	30d revenue	30d holders rev	Note
Hyperliquid	HYPE $83.80 (−5.3% 24h)	$18.63B	$80.0B	$90.55M	$73.41M	$73.41M	Leader. Oct 7 fees $3.53M (+43%) (Fact)
Aster	ASTER $0.686 (−3.6%)	$1.87B	$5.37B	$5.36M (perps)	unavailable	unavailable	Revenue not split on DefiLlama (Fact: missing)
Lighter	LIT $3.59 (+2.2%)	$898.6M	$3.59B	$5.49M	$4.59M	$2.55M	≈6.1% of HL's 30d fees (Estimate)
edgeX	EDGE $0.374 (−7.9%)	$131.1M	$374.6M	$3.83M	$2.76M	$0.61M	Cheapest on multiples; smallest float
Extended	no token	—	—	$1.55M	unavailable	—	—
Arcus (RH Chain)	no token	—	—	$1.08M	$0.99M	—	Points S1 live; geo-restricted
TxFlow	no token	—	—	$0.87M	$0.87M	—	Points farming
Variational	VAR not live	—	—	not on DefiLlama	—	—	Zero-fee RFQ, earns spread; TGE Q4 2026

Regime read (Estimate): this is a risk-off volatility leg: BTC $81,138 (−2.6% 24h), ETH $2,447.77 (−4.5%), long-heavy liquidations. In perps, volatility lifts fees, so HL +43% and Lighter trading revenue +53% on Oct 7 both reflect a down-move rather than demand growth. Third-party trackers put late-September 30d volume at about $204–217B for HL, $65–68B for Aster, $53–54B for Lighter and $37–40B for edgeX. Lighter's native API gives $59.5B over 30d (zkLighter $44.8B + RH $14.8B), consistent with that order.

Snapshot

Metric	Value	Label	As-of
Price	$3.59 (24h +2.2%, 7d −3.9%, 30d −22.9%; ATH $7.86 on Dec 30, 2025)	Fact	15:45 UTC / 11:45 ET
Market cap / FDV	$898.6M / $3.594B	Fact	15:45 UTC
Circulating / total / max	250M / 1B / 1B (25% circulating; CoinGecko figure, not adjusted for staking emissions or the 15.64M burned)	Fact (as listed)	15:45 UTC
Burned	15,638,703 LIT at the dead address (single burn, Jul 10, 2026)	Fact	≈15:47 UTC
Treasury LIT awaiting burn	≈2.62M LIT (≈$9.4M; 1.05% of circulating) in treasury account 0	Fact (balance) · Estimate (that it is earmarked for the next burn)	15:4x UTC
TVL	$762.4M (Ethereum escrow $648.3M + RH Chain $114.1M)	Fact	15:42 UTC
LLP collateral	≈$86.0M	Fact	15:5x UTC
zkLighter OI / RH OI (native)	$1.29B / $0.46B (native metric; likely counts both sides; third parties report ≈$0.63–0.85B one-sided total)	Fact · Estimate (interpretation)	Oct 7 UTC
24h volume (rolling)	zkLighter $2.78B (2.26M trades) · RH $0.70B (1.44M trades)	Fact	15:44 UTC

Fees / revenue (parent `lighter`)	1d (Oct 7)	7d	30d
Fees	$318.8k	$1.208M	$5.490M
Revenue	$227.2k	$1.035M	$4.595M
Holders revenue (buybacks)	$95.3k	$503.0k	$2.554M
Supply-side (liquidations → LLP)	$91.7k	$172.7k	$894.9k
zkLighter perps fees / revenue	$215.6k / $130.9k	$692.8k / $549.8k	$3.349M / $2.615M
RH perps fees / revenue	$101.3k / $94.4k	$501.8k / $472.1k	$2.042M / $1.881M

Score	Value
Research Score	55 / 100
Evidence Quality	Medium overall (High on the fee-model question)

All fee figures are Fact (completed Oct 7 UTC bar). They match the Radar pack to the dollar.

Fee-model vs burst analysis

Mechanism (Fact)

Fees = maker + taker + transfer + withdraw + liquidation fees, read per market from Lighter's own exchangeMetrics.
Revenue = maker + taker + transfer + withdraw fees. Liquidation fees are excluded and booked as SupplySideRevenue, because "up to 1% of notional" goes to the LLP.
Holders revenue = LIT bought by treasury account 0 on LIT/USDC (market 2049), measured from actual trades. It is not a fixed share of fees.
So a fees-minus-revenue gap is, by construction, that day's liquidation fees.

Daily history (zkLighter unless noted; Fact; take rate = trading revenue ÷ native volume, Estimate)

Day (UTC)	Parent fees	Parent rev	zkLighter perps fees	Trading/other rev	Liq fees → LLP	Liq share of perps fees	zk volume	zk take (bp)	Buyback (holders)	RH perps fees	HL fees
08-19	$599.5k	$200.5k	$460.9k	$156.6k	$304.4k	66%	$2.29B	0.68	$122.3k	$135.7k	$5.99M
08-21	$425.2k	$239.6k	$360.8k	$191.6k	$169.3k	47%	$2.85B	0.67	$171.2k	$59.2k	$7.30M
08-22	$347.1k	$146.7k	$309.5k	$117.7k	$191.8k	62%	$1.60B	0.74	$159.8k	$34.7k	$4.66M
09-15	$245.5k	$180.2k	$160.5k	$105.1k	$55.4k	34%	$1.81B	0.58	$123.7k	$81.9k	$4.55M
09-21	$389.0k	$248.6k	$266.8k	$143.6k	$123.2k	46%	$2.23B	0.64	$122.2k	$117.3k	$5.14M
09-23	$276.9k	$198.1k	$184.7k	$114.3k	$70.4k	38%	$2.00B	0.57	$142.5k	$87.4k	$3.28M
09-24	$203.2k	$186.4k	$123.3k	$110.1k	$13.2k	11%	$1.82B	0.61	$112.7k	$75.4k	$4.12M
09-25	$182.4k	$164.3k	$105.9k	$93.6k	$12.4k	12%	$1.54B	0.61	$76.2k	$73.1k	$3.00M
09-26	$83.9k	$81.6k	$42.1k	$40.1k	$2.0k	5%	$0.68B	0.59	$56.1k	$39.1k	$1.70M
09-27	$107.9k	$102.6k	$55.9k	$51.5k	$4.4k	8%	$0.86B	0.60	$48.7k	$47.5k	$2.29M
09-28	$263.2k	$199.9k	$163.7k	$108.8k	$54.9k	34%	$1.88B	0.58	$85.0k	$95.6k	$3.09M
09-29	$185.2k	$158.8k	$110.3k	$90.3k	$20.0k	18%	$1.67B	0.54	$136.9k	$70.8k	$2.98M
09-30	$204.7k	$175.1k	$124.2k	$98.8k	$25.4k	20%	$1.75B	0.57	$77.1k	$73.7k	$2.19M
10-01	$168.0k	$158.2k	$90.7k	$83.8k	$6.9k	8%	$1.51B	0.56	$75.4k	$75.8k	$2.43M
10-02	$250.9k	$200.2k	$143.4k	$104.8k	$38.6k	27%	$1.71B	0.61	$103.3k	$105.8k	$4.06M
10-03	$68.3k	$66.6k	$29.5k	$28.1k	$1.4k	5%	$0.54B	0.52	$49.1k	$36.0k	$1.40M
10-04	$84.6k	$80.6k	$40.1k	$37.9k	$2.3k	6%	$0.77B	0.49	$23.7k	$43.4k	$1.76M
10-05	$162.1k	$154.4k	$85.5k	$81.2k	$4.3k	5%	$1.52B	0.54	$68.4k	$73.6k	$3.03M
10-06	$155.4k	$148.2k	$87.9k	$83.1k	$4.8k	5%	$1.66B	0.50	$87.8k	$66.0k	$2.47M
10-07	$318.8k	$227.2k	$215.6k	$130.9k	$84.7k	39%	$2.42B	0.54	$95.3k	$101.3k	$3.53M

Oct 7 liquidation fees by market (zkLighter, top contributors; Fact)

Market	Oct 7 liq fee	Oct 6 liq fee	Share of Oct 7 total ($84.7k)
ETH	$56.3k	$0.2k	≈66%
BTC	$13.3k	$0.6k	≈16%
PUMP	$1.4k	≈$0	≈2%
HYPE	$1.3k	≈$0	≈2%
ENA	$1.0k	≈$0	≈1%
All others (top-35 sum $76.0k)	—	—	remainder

Evidence scorecard

Test	Result	Points to
Methodology text	Liquidation fees are excluded from revenue and sent to the LLP; unchanged since the Aug 12 "count dssr" commit	Burst (Fact)
DefiLlama adapter commits, last ≈2 weeks	One commit (Oct 6, #9948) added `dailyProtocolRevenue` (revenue minus buybacks). No change to fees, revenue or supply-side. RH adapter: no commits since Jul 22	Not an adapter artifact (Fact)
Has the gap appeared before?	Yes: 21 days with LLP fees >$50k since Apr 10 (e.g., Aug 19 $304k, Aug 22 $192k, Sep 21 $123k), and 20 days with LLP share >35% of perps fees. The 30d average share is 22%; the median day is $13.2k	Recurring burst (Fact)
Take rate on trading fees	0.54 bp vs 0.50 bp (Oct 6), 0.58 bp (30d), 0.55 bp (Sep 30–Oct 7). Slow decline since August (0.76 → 0.55 bp); no step up	No schedule change (Estimate)
Docs fee schedule	Standard 0/0; Premium 0.4 bp maker / 2.8 bp taker, with staking discounts (introduced Feb 2026); Plus 0.5 bp. No Oct 2026 change found	No change (Fact, docs)
Market day	ETH −4.7% (open $2,697, low $2,551), BTC −2.7%; HL fees +43%	Volatility burst (Fact)
Partial Oct 8	Lighter's API and DefiLlama publish only completed days. Direction hint: ETH another −4.7% and BTC −2.5% through 15:45 UTC, and the digest cites ≈$255.9M overnight liquidations, so another elevated LLP day is likely	Speculative

What would confirm or reject the verdict

Bar(s)	Confirms burst	Would indicate a fee-model change
Next 2–3 completed bars (Oct 8–10)	Once volatility fades, LLP fees fall back under ≈$15k/day and fees ≈ revenue again; take rate stays at 0.5–0.6 bp	A fees-minus-revenue gap above $30k/day on calm days (ETH within ±2%), or a take rate above 0.7 bp without a volatility spike
Docs / adapter	No edits to trading-fees.md or `fees/lighterv2`	A new fee tier, an LLP revenue-share change, or an adapter re-classification
Oct 8 bar specifically	Likely elevated again (market fell further), which fits the burst story. Do not read a high Oct 8 as confirmation of a model change.	—

Token value capture

Flow	Destination	Status	Reaches LIT?
zkLighter maker/taker/transfer/withdraw fees	Treasury → daily 24h TWAP buybacks of LIT on Lighter spot	Live (Fact: treasury trades on-chain; buys executing today at ≈$3.52)	Yes (≈1:1 with zkLighter revenue over 30d/90d)
Bought-back LIT	Burned (policy since Jul 1, 2026); one burn of 15,638,702 LIT on Jul 10	Live but periodic; ≈2.62M LIT now sitting in treasury	Yes. Supply reduction lands only when burns execute (next burn date unannounced: Speculative)
RH perps revenue (41% of 30d parent revenue)	Lighter + Robinhood; reported 50/50 split (one secondary source)	Live	Not visible: $0 holders revenue on DefiLlama, and buybacks track zkLighter revenue alone (Estimate)
Liquidation fees (22% of zkLighter perps fees, 30d)	LLP (≈$86M pool)	Live	Indirect only: LLP deposits require staked LIT (10 USDC per 1 LIT staked), so it is a utility link, not a cash flow to the token
Staking APR (fixed 6%)	Paid from the ecosystem reserve since Jul 1	Live	Emissions, not revenue (dilutive; fake-yield flag)
LIT Fee Credits proceeds	Streamed to stakers	Live	Yes; size unavailable
Equity	Nov 2025 round was equity plus token warrants	—	Value is split between Lighter equity and LIT (Fact as reported)

Holders-revenue ratios (Fact inputs; ratios are Estimates)

Window	Holders / fees	Holders / revenue	Holders / zkLighter perps revenue	RH share of parent revenue
1d (Oct 7)	29.9%	42.0%	72.8% (buyback is a TWAP, so it lags a spike day)	41.6%
7d	41.6%	48.6%	91.5%	45.6%
30d	46.5%	55.6%	97.7%	40.9%
90d	57.0%	71.4%	99.1%	25.8%

Read (Estimate): holders revenue as a share of fees is falling because growth has come from (a) RH, which does not feed visible buybacks, and (b) liquidation spikes, which go to the LLP. The buyback machine itself is consistent. Annualized holders revenue ≈$31.1M ≈ 3.5% of market cap / 0.9% of FDV. Quarterly buybacks: Q1 $14.6M → Q2 $7.1M → Q3 $7.4M (Fact). Monthly revenue troughed at ≈$2.2M in Apr–Jul and recovered to $4.5M in September, while holders revenue stayed flat at ≈$2.6–2.7M/month because the gain came from RH.

Variational market on Lighter

Item	Finding	Label
What it is	A prelaunch perp on Variational's not-yet-issued VAR token (Variational: zero-fee RFQ perps DEX on Arbitrum; TGE "Q4 2026"; 32% genesis airdrop unlocked at TGE; US/Canada restricted)	Fact
Market ID / listed	id 4098, created Oct 7, 2026, 11:18 UTC (7:18 AM ET), status "active"	Fact
Leverage	Initial margin 20% → 5x; prelaunch rules are isolated-only, no liquidation fee, and a breach of close-out margin goes straight to ADL	Fact
Activity (re-checked 15:52 UTC)	0 trades, $0 volume, 0 OI. Mark = index = 1.6701 (a seed reference, not a traded price). `force_reduce_only: true`: new positions appear blocked	Fact (data) · Estimate (interpretation)
Order book	Only the LLP quotes: 99 asks ($84.0k, 1.6716–1.7214) and 99 bids ($81.4k, 1.6206–1.6689)	Fact
Fee contribution	$0 to date. Even once live, Standard accounts pay 0, there are no liquidation fees, and Lighter's other pre-IPO markets (ANTHROPIC, OPENAI) do well under $1M/day each	Fact / Estimate
Official announcement	None found from Lighter or Variational; the "announced" claim comes from one recap account	Rumor
Signal	Prelaunch and pre-IPO markets (OPENAI, ANTHROPIC, OURA, VARIATIONAL) are a product line for attention and listings, not fees. Polymarket's VAR FDV odds ($1B ≈64.5% at 06:10 UTC Oct 8; Estimate) show demand for price discovery that Lighter could capture if it opens the market	Speculative

Lighter on RH Chain

Daily bars since the subsidy cliff (Fact; shares are Estimates)

Day (UTC)	Lighter RH fees	Lighter RH rev	Arcus fees	Arcus rev	RH chain gas fees	Lighter share of RH perps fees	Lighter RH ÷ chain gas	RH volume	RH OI (native)
09-22 (pre)	$77.5k	$71.9k	$27.8k	$25.5k	$355.5k	74%	0.22×	$561M	$435M
09-27 (pre)	$47.5k	$46.6k	$49.9k	$46.0k	$68.0k	49%	0.70×	$406M	$456M
09-28 (pre)	$95.6k	$87.2k	$92.2k	$85.6k	$108.0k	51%	0.89×	$652M	$450M
09-29 (cliff day)	$70.8k	$64.4k	$68.5k	$63.2k	$166.1k	51%	0.43×	$548M	$456M
09-30	$73.7k	$69.5k	$70.3k	$64.4k	$97.1k	51%	0.76×	$602M	$454M
10-01	$75.8k	$72.9k	$86.8k	$78.3k	$82.7k	47%	0.92×	$589M	$453M
10-02	$105.8k	$93.7k	$75.0k	$66.3k	$100.9k	59%	1.05×	$676M	$459M
10-03	$36.0k	$35.7k	$34.8k	$30.9k	$59.4k	51%	0.61×	$353M	$454M
10-04	$43.4k	$41.7k	$50.6k	$44.8k	$49.9k	46%	0.87×	$379M	$451M
10-05	$73.6k	$70.2k	$87.1k	$78.7k	$55.1k	46%	1.34×	$627M	$461M
10-06	$66.0k	$63.5k	$79.0k	$71.4k	$49.5k	46%	1.33×	$581M	$471M
10-07	$101.3k	$94.4k	$101.9k	$92.5k	$51.8k	50%	1.95×	$731M	$459M

RH metric	Value	Label
Avg Lighter RH fees, Sep 22–28 vs Sep 30–Oct 7	$70.8k → $71.9k (flat through the cliff)	Fact
Avg RH daily volume, Sep 15–28 vs Sep 30–Oct 7	$493M → $567M (+15%)	Fact
RH take rate	≈1.2–1.3 bp, stable since August (vs ≈0.55 bp on zkLighter)	Estimate
Avg RH chain gas, same windows	$166.5k → $68.3k (−59%)	Fact
Lighter TVL on RH	$106.9M (Sep 29) → $114.1M (Oct 8), ≈10.9% of RH chain TVL ($1.047B); Arcus $46.2M	Fact
Lighter share of RH perps fees	Fell from ≈74–83% (Sep 22–25) to ≈46–59% since Sep 27, as Arcus fees stepped from ≈$13–16k to $50–100k/day starting Sep 27 (cause unverified)	Fact (bars) · Estimate (share)
Lighter vs RH chain gas	Lighter RH fees alone are now 1.95× chain gas	Estimate

Read (Estimate): Lighter's RH perps volume did not depend on the gas subsidy. Trading happens on Lighter's own RH instance, and only deposits and withdrawals touch RH gas, so the cliff hit chain gas, not perps. The real post-cliff change is share loss to Arcus, which now matches Lighter on daily fees.

Organic vs points-farmed (RH)

Signal	Finding	Label
Points program	Live: real-time points plus weekly Friday drops (next Fri Oct 9); 2x via Robinhood Wallet; pool 11M LIT (some reports say "$11M") ≈ $39.5M at today's price	Fact
Incentive vs revenue	The pool, at about $39.5M, is ≈21× RH's 30d revenue ($1.88M)	Estimate
Trade size	Avg RH trade ≈$489 vs ≈$1,229 on zkLighter (rolling 24h); 1.44M RH trades/day	Estimate
Turnover	RH volume/OI ≈1.6×/day (not wash-like extremes)	Estimate
Rules	The terms ban wash trading, Sybil activity and self-referral, with retroactive clawback	Fact
Verdict	Partly incentive-driven; organic share cannot be measured with public data	Estimate

Access and competition

Item	Finding	Label
Geo	RH perps and points restricted for US, Canada, UK, Switzerland, UAE, Singapore and others	Fact
Gas	Chain subsidy ended Sep 29, 11:59 PM EST; Robinhood Wallet swaps >$0.50 are gasless through Dec 31	Fact
Bitstamp U.S. perps	Robinhood Derivatives + Bitstamp; 8 contracts; 10x BTC/ETH; 1 bp all-in through Dec 31; rolling out. LIT fell ≈13% in 24h on the news (secondary). It does not compete for Lighter's non-U.S. RH flow, but it removes the U.S.-flow upside	Fact (terms) · Estimate (impact)

Tokenomics & unlocks

Bucket	%	LIT	Status	Schedule
Airdrop (Seasons 1–2)	25%	250M	Unlocked Dec 30, 2025	Done
Team	26%	260M	Locked	12-month cliff → 36-month linear; cliff ≈Dec 27–30, 2026 (sources differ)
Investors	24%	240M	Locked	Same; reported entry ≈$0.28/LIT (≈$283.5M FDV), $68M raised; ≈12.7× in the money at $3.59
Ecosystem / reserve	25%	250M	No published schedule	Funds the 6% staking APR (≈7.5M LIT/yr at ≈125M staked, July figure), the 11M LIT RH pool, and partnerships

Supply math (Estimate)	Value
Insider vesting after the cliff	≈456.6k LIT/day ≈ $1.64M/day ≈ $598M/yr at $3.59
vs current buybacks	≈$85k/day → vesting ≈19× buyback
First-year insider release vs current float	166.7M ≈ 67% of 250M circulating
Ecosystem emissions (staking + RH pool)	≈7.5M LIT/yr + up to 11M LIT; timing discretionary
Staked LIT	≈125M (July figure); ≈155M LIT sits on zkLighter ($554.7M, 46% of the rollup's secured value)

Holder concentration (Ethereum L1 only; Fact balances, Estimate labels)

Holder	LIT	% supply	Likely identity
0xe554…9d62 (EOA, 4 txs)	234.3M	23.4%	Unlabeled; locked or reserve custody (Speculative)
0x3b4d…5ca7 (zkLighter escrow)	156.3M	15.6%	LIT deposited on Lighter (staking/trading)
0xa9f5…44e8 (EOA, 0 txs)	58.9M	5.9%	Unlabeled, dormant (Speculative: vesting)
0x47fa…b2d6 (EOA)	51.2M	5.1%	Exchange hot wallet (holds BGB; Estimate)
0xe5fd…fa59 (EOA, 0 txs)	42.4M	4.2%	Unlabeled, dormant
Dead address	15.6M	1.6%	Jul 10 burn
Top 10 / top 20	—	64.7% / 74.4%	10,060 L1 holders; L2 balances not visible

Valuation comps (annualized from 30d; Estimate)

Venue	MC / fees	MC / revenue	MC / holders rev	FDV / revenue	FDV / holders rev	Note
Lighter (parent)	13.5×	16.1×	28.9×	64.3×	115.7×	Holders rev ≈$31.1M/yr
Lighter, zkLighter revenue only	—	28.2×	28.9×	113.0×	115.7×	The part that actually feeds buybacks
Hyperliquid	16.9×	20.9×	20.9×	89.6×	89.6×	≈all revenue → AF buybacks
Aster	28.7×	unavailable	unavailable	—	—	FDV/fees 82.4×
edgeX	2.8×	3.9×	17.6×	11.2×	50.2×	Thin liquidity ($3.3M/day)

Read (Estimate): on parent revenue, LIT looks cheaper than HL (16× vs 21×). On the revenue that actually reaches the token, LIT is more expensive than HL (29× vs 21× market cap, 116× vs 90× FDV), and it carries a near-dated insider cliff that HL's float does not.

Competition

Rival	Edge vs Lighter	Threat	Label
Hyperliquid	Depth, ≈10× fees, AQAv2 ≈$193M/yr run rate, options next (no date)	Structural	Fact / Estimate
Aster	Volume #2 with incentives; similar market cap	Medium	Estimate
Variational	Zero-fee RFQ; points into a Q4 TGE pull flow from the same farmers	Medium (until its TGE)	Estimate
Arcus (RH)	Now matches Lighter's RH fees; points S1	High on RH	Fact (bars)
Bitstamp via Robinhood (U.S.)	1 bp, regulated, Robinhood distribution	Caps U.S. upside	Fact (terms)
edgeX / Extended / TxFlow	Smaller	Low	Fact (fees)

Catalysts

Window	Catalyst	Date	Status	Direction
30d	Lighter RH weekly points drop	Fri Oct 9 and weekly	Confirmed	Activity support
30d	Next LIT burn (≈2.62M LIT in treasury)	Unannounced	Speculative (first burn came 10 days after quarter end)	+ small (≈1% of float)
30d	VARIATIONAL market opens to trading	—	Speculative (currently force-reduce-only)	Attention, ≈0 fees
30d	Oct 8–10 bars confirm the burst	Oct 9–11	Data	Removes the "+105%" narrative
1–3m	Team/investor cliff	≈Dec 27–30, 2026	Confirmed schedule (day varies by source)	Strongly negative supply
1–3m	Variational TGE	"Q4 2026"	Confirmed window, no date	Farmers migrate either way
1–3m	Bitstamp 1 bp promo and RH Wallet gasless swaps end	Dec 31, 2026	Confirmed	Mixed
3–6m	Linear insider vesting at ≈456.6k LIT/day	From the cliff	Confirmed	Negative
3–6m	More "Lighter Domains" (broker/chain instances after RH)	—	Speculative	+ if revenue is shared with LIT
6–12m	L2BEAT Stage 1 (security council, exit window)	—	Speculative	+ trust
6–12m	Ecosystem-reserve drawdowns	Discretionary	Unscheduled	Negative

Bull / base / bear (6–12m; Speculative)

Case	Conditions	Holders rev (ann.)	Implied market cap at 25× holders rev	Supply context
Bull	Another Domain launches with a revenue share into buybacks; zkLighter take recovers to ≈0.7 bp; insiders don't sell much	≈$60M	≈$1.5B	Circulating ≈+67% by end-2027, so per-token upside is much smaller than the market-cap move
Base	Revenue run-rate ≈$50–60M; buybacks ≈$30M; Arcus splits RH	≈$31M	≈$0.8B	Flat-to-down per token once vesting starts
Bear	Volatility fades, take rate keeps sliding (0.76 → 0.55 bp since Aug), RH share erodes, insiders sell into thin books	≈$20M	≈$0.5B	Vesting ≈19× buyback

Bear case & fake-metric filter

Aggressive bear case (Estimate): the radar headline (+105% fees) was half liquidations that LIT never sees. Revenue that reaches the token is ≈$2.6M/month and has been flat for six months. The RH growth story pays Robinhood and the LLP, not LIT. In about 80 days, ≈$600M/yr of insider supply (at today's price) starts vesting against ≈$31M/yr of buybacks, and books thin out to ≈$200–560k within ±1% per venue. Staking "yield" is reserve emissions. The contracts are instantly upgradable by a multisig, and L2BEAT flags fast-tracked upgrades and unpublished verifier sources.

Filter	Finding	Verdict
Adapter artifacts	Oct 6 commit added a field only; no change to fees, revenue or LLP	Clean (Fact)
Liquidation inflation	Liquidations are 39% of Oct 7 zkLighter perps fees and 22% over 30d	Strip from "fees" when judging value capture
Zero-fee accounts	Standard accounts pay 0; revenue comes from Premium/Plus (mostly makers and HFT, maker share ≈55–59% on RH). Volume ≠ revenue: 0.55 bp take	Volume overstates economics
Points farming	RH points live (11M LIT pool ≈21× RH 30d revenue); Variational/Arcus/N1 farming chatter	Partly incentive-driven
Wash trading	No direct evidence; turnover 1.6–1.9×/day; the terms ban it	Unverified
Revenue paid in emissions	The 6% staking APR comes from the ecosystem reserve	Yes, flag it
Circular TVL	46% of L2BEAT's TVS is LIT itself (staked)	Flag it: USDC TVL ≈$624M is the real collateral base

Team / legal / security

Area	Finding	Label
Team	CEO Vladimir Novakovski (Harvard; ex-Citadel; Lunchclub founder); team founded 2022	Fact
Backers	≈$90M raised: $21M (2024; Haun, Craft, Dragonfly, Robot) + $68M (Nov 2025; Founders Fund, Ribbit lead; Haun, Robinhood) at about a $1.5B valuation; equity plus token warrants	Fact (as reported)
Jurisdiction	U.S.-founded company; the Terms bar U.S. persons from trading; operating-entity jurisdiction unavailable	Fact / unavailable
Architecture	App-specific ZK rollup on Ethereum; state diffs posted as blobs; validity proofs	Fact
Escape hatch	Forced transactions through L1; desert mode after 14 days lets users exit with ZK proofs. L2BEAT reproduced the state snapshot	Fact
Admin keys	Stage 0. Exit window: none. The 4/7 multisig can zero the 21-day upgrade delay, and every upgrade so far was fast-tracked. The 3/5 governor manages validators, the treasury and markets. Pause/Emergency Guardians added Oct 2 (currently set to 0x). Several verifiers were deployed without published sources. Oracle (Stork primary) signatures are not verified on-chain	Fact
Audits	Nine audit files on the docs page (auditor names not extracted)	Fact (count)
Incidents	Oct 10–11, 2025: 4.5h outage during the $19B liquidation wave, ≈5.35% LLP loss, ≈$50M user plus LLP losses compensated with points. L2BEAT liveness gaps: Sep 8 (≈4h), Sep 13, Sep 20, Oct 1 (≈3h). Third-party probe logged a 5-minute "status incident" Oct 8, 14:00 UTC	Fact / Low (probe)
Dev activity	elliottech repos pushed Oct 5–8 (prover Oct 7, verifiers Oct 8, TS SDK Oct 6); lighter-python 8 commits in 30d; frequent contract upgrades	Fact

Investability

Item	Finding	Label / as-of
24h spot volume	≈$121–132M	Fact, 15:45 UTC
Venue concentration	OKX 25.5% (USDT 13.1%, USDC 12.4%); DigiFinex 11.0%; MEXC 11.0%; BitDelta 9.3% (spread ≈1.96%: low-quality volume); Bybit 7.3%; Gate 6.1%; Coinbase 4.2%; Lighter spot 3.5%; Kraken 0.9%. About 37% sits on lower-tier venues	Fact (shares) · Estimate (quality)
U.S. access	Spot on Coinbase and Kraken; Lighter's perps are not available to U.S. persons	Fact

Book (15:46 UTC)	Spread	Depth ±1% (ask / bid)	Est. slippage, buy $50k	Buy $250k	Sell $50k	Sell $250k
OKX LIT-USDT	5.6 bp	$198k / $335k	≈26 bp	≈69 bp	≈12 bp	≈39 bp
OKX LIT-USDC	0.3 bp	$114k / $156k	≈63 bp	≈102 bp	≈55 bp	book exhausted at ≈$246k
Coinbase LIGHTER-USD	14 bp	$281k / $305k	≈24 bp	≈48 bp	≈21 bp	≈49 bp
Lighter spot LIT/USDC	4.2 bp	$201k / $178k	≈13 bp	≈51 bp	≈15 bp	≈77 bp
Lighter LIT perp	2.5 bp	$565k / $555k	≈11 bp	≈23 bp	≈14 bp	≈40 bp
Binance / Bybit	—	unavailable (geo-blocked from the research box)	—	—	—	—

Slippage is measured from best bid/ask (Estimate). Lighter books are capped at 250 resting orders per side, so depth may be understated. Read: $50k fills in one venue at ≈10–25 bp. $250k needs splitting across venues (or the Lighter perp) to stay under ≈50 bp.

Portfolio notes

Pair	30d corr (beta)	60d	90d	Read
LIT vs BTC	0.22 (0.58)	0.43 (1.08)	0.50 (1.37)	High beta, loose coupling
LIT vs ETH	0.30 (0.62)	0.44 (0.79)	0.49 (0.96)	Settlement-layer link is weak in price
LIT vs HYPE	0.54 (0.77)	0.50	0.51	Highest overlap: same perps-sector factor
LIT realized vol	≈108–112% annualized	—	—	≈2× majors (Estimate)

Holding LIT alongside HYPE double-counts the perps-DEX factor. LIT also carries idiosyncratic Robinhood and unlock risk. Its fees rise on volatility days, so revenue partly hedges market stress, but the token price has not.

Metrics to monitor / invalidation

Metric	Healthy	Invalidation / warning
LLP fees share (zkLighter)	<15% on calm days	Persistent fee-vs-revenue gap >$30k/day on calm days → re-examine the model
zkLighter take rate	≥0.55 bp	<0.45 bp for 2 weeks (fee compression)
Holders revenue, 30d	≥$2.5M	<$2.0M (buyback weakening)
Holders / revenue, 30d	≥50%	<40% (growth not reaching the token)
Lighter share of RH perps fees	≥50%	<40% for a week (losing to Arcus)
Next burn	Executed and visible at the dead address	No burn by end of November (policy slippage, Speculative threshold)
Insider wallets (top dormant EOAs)	Dormant	Transfers to exchanges before or at the cliff
VARIATIONAL market	Opens with real OI	Stays at zero (listing was cosmetic)
Security	Stage-1 progress; verifier sources published	Outage >1h during volatility; unannounced fast-track upgrade

WHAT COULD I BE WRONG ABOUT?

The burst label could hide a slow share gain. Trading revenue rose +53% vs HL +43%. If that outperformance repeats on calm days, Lighter is gaining share, which this report treats as beta.
The RH 50/50 split comes from one secondary source. If Lighter keeps all RH revenue and simply has not routed it to buybacks yet, value capture could step up sharply with a single policy change.
Buybacks may not be strictly tied to zkLighter revenue. The ≈1:1 match over 30d/90d could be coincidence or discretionary sizing, and the team can resize at will.
The cliff may matter less than the math says. Insiders bought in at ≈$0.28 but may hold, stake or extend locks. Unlocked does not mean sold.
The cliff date and amounts could differ. Sources say Dec 27 or Dec 30. The ecosystem reserve may already be releasing more than CoinGecko's 250M circulating figure shows.
Native OI is ambiguous. It may be two-sided; per-market OI turnover reads would change if so.
Holder labels are inferred. The 234M EOA could be an exchange or custodian rather than locked supply.
The Arcus step-up from Sep 27 might be an adapter or points artifact, which would overstate Lighter's RH share loss.
Oct 8 could print an even bigger fee bar, again from liquidations. That supports the verdict but will look like "momentum" in headline screens.
Slippage estimates are a single snapshot and can be off by 2× within an hour on a volatile day.

Sources / technical notes

Pull times (UTC / ET): DefiLlama 15:42 UTC (11:42 ET) · Lighter API 15:44–15:52 UTC · CoinGecko 15:45 UTC · order books 15:46 UTC · Ethplorer ≈15:47 UTC · GitHub ≈15:43–15:50 UTC · L2BEAT ≈15:5x UTC. Completed fee day = Oct 7 UTC.

DefiLlama (Fact):

https://api.llama.fi/summary/fees/lighter?dataType=dailyFees (also dailyRevenue, dailyHoldersRevenue, dailySupplySideRevenue, dailyProtocolRevenue); child slugs lighter-perps, lighter-spot, lighter-robinhood-perps
Comps: hyperliquid, aster-perps (Aster revenue and holders not available: "not found"), edgex, arcus-perps, txflow-perps, extended-perps, robinhood-chain; variational not listed
https://api.llama.fi/protocol/lighter (TVL by chain); https://api.llama.fi/v2/historicalChainTvl/Robinhood%20Chain; https://api.llama.fi/protocol/arcus
summary/derivatives/* (volume) is paywalled ("Upgrade to the paid API plan"); native Lighter volume used instead

Adapter code / commits (Fact):

https://github.com/DefiLlama/dimension-adapters/blob/master/fees/lighterv2/index.ts
https://github.com/DefiLlama/dimension-adapters/blob/master/fees/lighter-rh/index.ts
Commits: fees/lighterv2: 2ceb4009 (2026-10-06 08:28 UTC, #9948, adds ProtocolRevenue only), a40739fd (09-07), 6aa0b5e2 (08-12, "count dssr"); fees/lighter-rh: 53d7fde8 (07-22); fees/arcus-perps: d7a28434 and 453335f0 (10-01, chain re-keying)

Lighter primary (Fact):

API: https://mainnet.zklighter.elliot.ai/api/v1/exchangeMetrics?period=all&kind={liquidation_fee,maker_fee,taker_fee,volume,open_interest} (per-market via filter=byMarket); /orderBookDetails; /exchangeStats; /orderBookOrders (market 4098, 2049, 120); /account (index 0 treasury; 281474976710654 LLP)
RH API: https://api.rh.lighter.xyz/api/v1/ (same endpoints)
Treasury buyback logs: https://explorer.elliot.ai/api/accounts/0/logs?pub_data_type=Trade
Docs: https://docs.lighter.xyz/trading/trading-fees · /about-lighter/lit-utility · /trading/trading-fees/lit-fee-credits · /trading/prelaunch-markets · /trading/liquidations-and-llp-insurance-fund · /points-program/lighter-on-robinhood-chain-points · /security/security-audits · /about-lighter/technical-architecture-lighter-core
GitHub org: https://github.com/elliottech (repo push times via GitHub API)

Market data (Fact): CoinGecko /coins/lighter (tickers), /coins/markets, /coins/{id}/market_chart (90d daily for correlations; 2d intraday for BTC/ETH Oct 7–8). Order books: OKX /api/v5/market/books (LIT-USDT, LIT-USDC), Coinbase /products/LIGHTER-USD/book?level=2, Lighter /orderBookOrders. Binance and Bybit returned geo-restriction errors.

On-chain (Fact): Ethplorer getTopTokenHolders, getTokenInfo, getAddressInfo for 0x232ce3bd40fcd6f80f3d55a522d03f25df784ee2. L2BEAT: https://l2beat.com/layer2s/projects/lighter

News / secondary (Estimate unless noted):

Burn and staking: CryptoTimes (2026-07-01), KuCoin flash (2026-07-01), CoinDCX / cryptonews.net (Jul 10 burn)
Tokenomics: tokenomics.com, tokenradar.ai, tokenomist.ai (Dec 27 cliff), TokenIntel memo (Dec 30 cliff, ≈456k LIT/day)
RH: Robinhood newsroom (11M LIT pool, restricted regions: primary), Robinhood support article, CryptoBriefing (Jul 11, 2026; 50/50 revenue split, single source), Bankless (Lighter Domains)
Team/incidents: Fortune (Nov 11, 2025), CryptoTimes (Oct 13, 2025), Odaily/Beosin, Phemex
Bitstamp: Yahoo Finance, coin360, Cryptonomist (Sep 30, 2026), BlockBeats
Perps share: egoncoin, Gate Learn, BuiltOnBulk (Sep 2026); Castle Labs (take rates, OI share)
Variational: memcool.app (Polymarket odds at 06:10 UTC Oct 8), perpdexguide.com, cipher-intelligence.io (pre-IPO list), bykaranteli.com (listing time, status probe)

Internal packs: /workspace/crypto-packs/2026-10-08/radar.md (Llama ≈09:21 ET; all Lighter numbers re-verified and matched) · /workspace/crypto-packs/2026-10-08/airdrops.md (Variational docs re-check 9:20 AM ET) · /workspace/crypto-packs/2026-10-08/digest-warm.md (browser-verified DIGEST_OK, as_of 2026-10-08 08:09 ET).

Technical notes:

Take rate = DefiLlama zkLighter perps revenue ÷ Lighter native volume metric (which may include small spot volume).
Annualization = 30d × 365/30.
Ratios use parent lighter unless stated.
"Liq share" = SupplySideRevenue ÷ zkLighter perps fees.
Per-market liquidation fees were queried for the 35 highest-volume perps (sum $76.0k of $84.7k).
Correlations use daily log returns from CoinGecko daily closes.
Holder labels are inferred from balances and contract status.
Raw pulls are saved in /workspace/crypto-packs/2026-10-08/lighter-raw/.