Public record

@0xBumblebee

Posts, receipts and the limits of the evidence.

BumblebeeBlue checkmark (X Premium subscription) @0xBumblebee

3,215 followers

decoding odds | part-time tennis quant | former kid

X (opens in a new tab)

Posts picks, claim cards and app tips. A second account it funded lost about $8,300, and no post shows that account’s trades.

  • Second account’s loss not shown
  • Paid by apps it promoted
  • Says it takes paid deals
  • Tested: no two-sided betting

@0xBumblebee shares picks, claim cards and app tips, and often posts its overall losses. A second account it funded, Uiom, lost about $8,300; no post shows its trades. DG3’s affiliate wallet paid the wallet it posts as its own $800 in the weeks it promoted DG3. Its posts did not mention the money.

In fairness

It often posted its overall losses, and told followers in general terms that its X account was paid.

Betting on both sides was checked across its nine addresses. It did not.

Show the record4 points and their sources

What the record shows

  1. 1Its public profile sent about $6,500 to a second account, Uiom, whose wallet it later posted as its own. Uiom kept betting its whole balance and lost about $8,300. No post shows Uiom’s trades.

    3 more sources

  2. 2Counting every account tied to it, the loss is about $18,100, more than twice the $8,120 its public profile shows. Most of the gap is Uiom; the profile’s own figure is accurate.

    Image @0xBumblebee postedSee it whole
    View on X (opens in a new tab)Archived copy (opens in a new tab)

    It did post big losses, like this one-day card of 2026-04-02.

    2 more sources

  3. 3DG3’s affiliate wallet sent $800 to the wallet it calls its own in the weeks it promoted DG3 with its code. The posts don’t mention pay; what the money was for is unknown.

    Post by @0xBumblebee on X, 9 June 2026See it whole
    View on X (opens in a new tab)Archived copy (opens in a new tab)

    Its DG3 post of 2026-06-09, with its code.

    2 more sources

  4. 4On 2026-06-11, a wallet attributed to Firefly paid its wallet $175 in a run of eight payments. That evening it promoted the festival, without mentioning a payment.

    Post by @0xBumblebee on X, 11 June 2026See it whole
    View on X (opens in a new tab)No archived copy exists

    Its festival post of that evening.

    1 more source

Accounts tied to @0xBumblebee

Payments received

Deleted posts, for example

Bumblebee@0xBumblebee · 5 Jan 2026
Deleted
seems that my strategy was to naive, nobody wants to wait 3 days and probably the huge amount of money will be pledged right after the sale starts in total i lost: > $300 at 1st time when I bought YES shares but sold 1 hour later > $600 if I sell current positions (I won't) the only good thing here - my initial research was good and if you aped there you already can sell in 38% profit I could end this up with $760 profit but lost $900 (only I could do this)
The post on X (deleted) (opens in a new tab)Archived copy (opens in a new tab)

Everything that must be said with these points

  • The handle posted its losses in aggregate and often: -$2,704.06 (‘PnL is down only ... By the way someone is me’, 2026-02-12), -$4,501 (2026-03-10), -$6,300 (2026-03-28), a day of -$23,061.46 and down $25,000 from a peak of about $45,000 (2026-04-02), ‘i took a needed break from trading on Polymarket due to big losses’ (2026-04-19), ‘-$3,200’ for a losing streak (2026-06-13), ‘i lost the whole bankroll’ (2026-07-17) and ‘still net negative’ (2026-08-04). What it did not show is this account and the size of its loss.
  • On 2026-05-29 the handle posted a market link carrying the referral name r=Uiom, and Uiom’s profile is public.
  • The winnings on the two claim cards the handle posted (‘Claim $22,358.24’ on 2026-03-31, ‘Claim 37,597.79’ on 2026-04-01) were real. The public figure of the visible profile is accurate for that account (-8,112.34 dollars by cash).
  • A comparison of single positions (113 positions shown one by one made +34,478.31 dollars, the 442 others lost 52,623.83) leaves out the aggregate loss cards and is not a measure of what was disclosed; 36 of its matches rest on a market name without a figure.
  • The two credits of 292.802 dollars that the handle asked followers about on 2026-04-19 (‘i found $585.6 on my account however i didn’t make any deposits’) were labelled ‘Deposited funds’, with no source, in the app screen it posted. They are best read as the referral payout for Uiom’s largest fee day, 2026-04-04 (30% of 950.62 = 285.19, plus about 7.6 from other referrals), paid two weeks late and twice. Asked whether it matched referral volume, the handle answered ‘nope, no referral volume yesterday’, which was true; nothing shows that it knew where the money came from.
  • That Uiom is the handle’s referral is inferred from the two exact 30% matches; the referral relation itself is not public. Polymarket’s referral help page (archived 2026-03-24, before Uiom existed) reserves the right to disqualify self-referrals; nothing shows that Polymarket did so.
  • Referral income is ordinary programme income and is context here, not the accusation: of the 894.51 dollars of referral rewards, up to 861.27 arrived on days tied to Uiom’s trading if the payout that came twice is counted twice.
  • Komacsi’s -787.96 predates the handle’s Polymarket posts. Two-sided betting between the accounts was tested and not found.
  • The test covers the accounts that were found; four deposits from BNB chain (1,077.25 dollars) and five small deposits of unread origin are where a further account could sit.
  • Inside single accounts both outcomes were bought in 18 markets and held at the same time in one (3.50 against 319.43 dollars, for 40 minutes on 2026-07-13).
  • The comparison with other accounts made earlier is not a chance rate (those accounts trade both sides of most markets) and is not used.
  • What the payments were for is not written on the chain; payment for a post or per post is not shown.
  • The posts show the handle’s personal code and link, so a reader can see a referral relation, and DG3 publishes a referral programme (50 dollars per referred trader). Against that reading, the first payment predates the code by eleven days and was the first of five transfers by that wallet within 41 minutes (150, 55, 60, 75, 75).
  • In the same weeks the handle said in public, in general terms, that it was paid for activity on X (‘earned mid 3 figs from gigs’ on 2026-05-31, ‘i get $1k/month for doing 50-200 replies a day on my own account’ on 2026-06-03, ‘june: 4 figs’ on 2026-08-13), without naming DG3.
  • X’s paid-partnership label cannot be read for posts before 2026-08-09; on the 88 own-timeline posts of the handle read after that date it is not set.
  • The DG3 account the handle used staked 125.09 dollars in all (result +32.66), and the handle posted its lost bets as well (‘so i lost $25’).
  • Being paid by an app is ordinary; what is stated here is promotion in the handle’s own voice by an account the promoted app was paying, with nothing in the posts saying so. For this handle H3 holds only in that form, next to its general statements that it was paid.
  • This is a general disclosure: none of its single posts promoting DG3, Firefly or Share carries words of disclosure.
  • Only 1,201.84 of the 3,251.84 dollars comes from wallets tied to named apps (DG3 800, Firefly 389.95 including a 14.95 contest prize, Bagel 11.89 in contest prizes) and 362 from payers the handle named itself; the remaining 1,688 comes from round-amount payers that are not identified and may not be pay at all.
  • The list is loose in both directions: it says ‘april: 0 figs’ although the handle had announced paid deals in April, and ‘july: 0 figs’ although 250 and 200 arrived in July (which may be counted by the month the work was done).
  • ‘i get paid nothing to post on X’ (2026-06-23) follows by three days a post about X’s own revenue sharing, which showed the account as ineligible, and reads as being about that; it is not a denial of paid deals.
  • In March 2026 the handle wrote about X’s enforcement against undisclosed sponsored posts (‘restricted to do undisclosed sponsored posts’, 2026-03-14), so it knew of the rule.
  • Polymarket programme income (1,389.38 dollars of referral rewards, liquidity rewards and rebates, and a 500-dollar contest prize) is ordinary.
  • The record gives the order of events, not their purpose: the replies did not trigger the payments; both belong to the launch of one campaign. What the money was for is not on the chain.
  • Firefly’s own posts advertise social rewards for sharing its events and creator campaigns, so a reward under a published campaign is possible, though 170 and 175 are not amounts of its published prize tables.
  • Three of the eight transfers of the run are small odd amounts (38.23, 17.55 and 8.00), so the run is five round payments and three other transfers.
  • @_kate_lv replied under the same post at 10:33 UTC and was paid 750 USDC by the same wallet a day later (2026-06-12, 11:25:31 UTC), not within minutes.
  • @vadim_kesha1 replied under Firefly’s posts three times in June (11, 14 and 24 June); only the first was followed by a payment, and Firefly’s second payment to the CryptoKesha-side address (180 USDC on 2026-06-14) came thirteen hours before that day’s reply.
  • The wallet that received the 200 on 2026-07-02 has four transfers in its whole history (200 in from Firefly’s hub and 25.2 in from the Firefly account; out to that account and to the deposit of the DG3 account @0xBumblebee showed); who holds its key is not shown.
  • The 14.95 PUSD of 2026-07-13 to @0xBumblebee’s Firefly account is one of 107 equal prizes of a public contest and is ordinary programme income; the Firefly account lost 249.38 dollars.
  • @0xBumblebee said in public, in general terms, that it was paid for activity on X in May and June 2026. X’s paid-partnership label cannot be read for June and July 2026.
  • That the paying wallets are Firefly’s is graded strong, not proven: the Polymarket profile Firefly-EverythingApp, which names @thefireflyapp, declares the loader as recipient of a withdrawal on 2026-10-01.