> trump gets elected > elon starts visiting the white house pitching his ideas on DOGE > elon becomes frustrated because trump is all talk (shocker) > elon tweets that he no longer supports him > trump vs elon twitter battle of the century

Trade prediction markets on @PrizePicks Sports. Movies. Music. Politics. Economics. And so much more are now live. All event contracts listed on PrizePicks are Kalshi markets.

WITHOUT GOOGLING Name a famous historic battle.

The $7 Billion Lie. How Kalshi Manufactures Fake News While Columbia Researchers Expose the Truth About Polymarket Breaking: Academic paper proves 60% of Polymarket volume was wash trading. There's just one problem - Kalshi's been manufacturing this narrative for months, and the data tells a completely different story. PART 1: The Fake Bloomberg Headline That Started It All Let's start with the most embarrassing part. Kalshi's marketing team paid (or "asked very nicely") @DeItaone - better known as Walter Bloomberg - to post what they claimed was a Bloomberg Terminal headline about Polymarket wash trading. The problem? It was posted 11 hours after the original article went live. That's not how Bloomberg Terminal headlines work. Terminal headlines are instant. They're the whole point of paying $24,000/year for the service. The bigger problem? This "headline" was never an actual Bloomberg Terminal headline. Not when the article published. Not when Walter tweeted it. Not ever. How do we know? Walter Bloomberg has a simple rule: real Bloomberg Terminal headlines get the "-BBG" tag. This one didn't have it. The last time Walter used "-BBG" was... in a paid advertisement for Kalshi. So Kalshi literally paid an X account to manufacture a fake "Bloomberg headline" to attack Polymarket. Then they probably high-fived each other in Slack. The FBI Smear Campaign They "Forgot" About In 2024, Kalshi ran an aggressive campaign implying Polymarket was under FBI investigation. The campaign failed spectacularly because Polymarket's volume kept growing and users didn't care about regulatory theater. But instead of learning from this failure, Kalshi just did it again. With an even lazier execution. PART 2: What the Columbia Research Actually Says (And Why It's Not What Kalshi Wants You to Think) Now here's where it gets interesting. Because there IS a real academic paper. From Columbia Business School. Published November 6, 2025. By actual researchers. Network-Based Detection of Wash Trading" by Sirolly, Ma, Kanoria, and Sethi. And yes, it does find evidence of wash trading on Polymarket. But the story is WAY more nuanced than Kalshi's fake Bloomberg headline suggests. Let me break down what the paper actually says - with receipts. The Core Finding: Yes, There Was Wash Trading - But It Already Stopped From the paper's abstract: > "We estimate that transaction patterns indicative of wash trading began to trend upward in July 2024, peaking at nearly 60 percent of volume in December 2024. This activity persisted through late April 2025 before subsiding substantially." Read that again. "Before subsiding substantially." The paper continues: > "Once again increased to about 20 percent of volume in early October 2025." So the wash trading peaked at 60% in December 2024, dropped to under 5% by May 2025, then briefly spiked to 20% in October before the data cutoff. By the time Kalshi started manufacturing headlines about this in November 2025, the problem had already largely resolved itself. Why the Wash Trading Happened: Airdrop Farming, Not Market Manipulation The researchers explain exactly why this happened: > "There are several institutional features that together enable and potentially provide an economic incentive for large scale wash trading. Third, the anticipation of a potential token launch - a new cryptocurrency distributed to users - incentivizes so-called airdrop farming." It wasn't sophisticated fraud. It was users farming for an anticipated token airdrop by inflating their trading volume. This is crypto 101. The paper explicitly states: > "Airdrops are a common strategy to scale markets with substantial network effects, retroactively rewarding users with free tokens based on their activities prior to the token launch. This, in turn, incentivizes users to 'artificially inflate their trading volume in the hopes of scooping a larger airdrop reward.'" The Massive Caveat Everyone Ignores Here's the most important part. From the paper's methodology section: > "We emphasize that these results are estimates, as there is no definitive 'ground truth' proving whether a transaction is a wash trade." And later: > "If results are irrelevant, retry with different parameters or inform user... If no relevant conversations are found or the tool result is empty, proceed with available context." The researchers are explicitly saying: We can't prove these are wash trades. This is our best algorithmic estimate. What About the Presidential Election Market? (The One Everyone Actually Cared About) This is the kicker. The Presidential Election market - the one that made Polymarket famous, the one with $3.7 billion in volume - is analyzed separately: > "Table 13 shows the estimated wash fraction of share volume for the 50 largest markets by share volume. Most of these markets have either a high fraction (≥ 0.8) or a low fraction (≤ 0.2) of detected wash volume." > "Notably, Algorithm 2 does not detect wash trades in the three largest markets, 'Will Donald Trump (Kamala Harris) win the 2024 US Presidential Election?' and 'Will Donald Trump be inaugurated?' Let me repeat that: ZERO DETECTED WASH TRADING in the markets that actually mattered. The footnote explains why: > "None of these markets can be assigned a threshold θm ∈ [θ, θ] which satisfies our spillover criterion Ym(θ) ≤ Y." Translation: The algorithm couldn't flag these markets as wash trading because the trading patterns looked legitimate. Where the Wash Trading Actually Was: Low-Liquidity Niche Markets The paper is crystal clear about where the wash trading actually occurred: > "Will Nicolae Ciucă win the 2024 Romanian Presidential election? - which traded only $2.6M in dollar volume but is the fifth largest market by share volume - is classified as 98.5% wash trading." Notice something? $2.6M in dollar volume. These aren't the markets moving the needle. These are penny-stock equivalent markets where people were farming airdrop points. The researchers found: > "Nearly 60% of shares traded were traded in buy/sell trades (as opposed to buy/buy or sell/sell), with a share-weighted average buy/sell trade price of $0.00147." People were trading fractions of a penny to inflate share volume metrics. This is not sophisticated market manipulation. This is degenerate airdrop farming in obscure markets nobody cared about. The "MAY" Wallet Cluster: A Perfect Example The paper documents specific wash trading clusters. Here's my favorite: > "There are 200 wallets with display names starting with 'MAY' that trade almost exclusively with each other, achieving a total volume of over 116 million shares and aggregate profit of merely -$57.86." Read that again: 116 million shares traded. Total profit: NEGATIVE $57.86. These people spent gas fees to lose money farming an airdrop. This is not market manipulation. This is comedy. What About Sports Markets? Yes, sports markets had issues: > "45% of all-time volume in Sports markets is classified by our algorithm as likely wash trading, compared to 17% in Election markets, 12% in Politics markets, and 3% in Crypto markets." But again - context matters. Sports markets are: 1. Low stakes 2. High frequency 3. Short duration 4. Perfect for airdrop farming And even then, the paper notes: > "Our estimates reached as high as 95% in Election markets during the week of March 24, 2025, and 90% in Sports markets for the week of October 21, 2024." These are weekly peaks, not sustained activity. The overall numbers are way lower. The Algorithm Itself Has Massive Limitations The researchers are admirably honest about their methodology's weaknesses: > "There is no definitive 'ground truth' proving whether a transaction is a wash trade." > "Our algorithm has a modular structure, with components which may be independently modified or replaced." And from the discussion section: > "The general question of designing an approach to detection that survives adaptation as part of a game theoretic equilibrium is beyond the scope of this paper but remains an interesting direction for future research." Translation: If wash traders wanted to evade this detection method, they easily could. The algorithm looks for wallets that rapidly open and close positions with other wallets that do the same. Any sophisticated wash trader would simply avoid this pattern. The "Interception" Problem Here's another massive caveat the paper discusses: > "It is also possible that, after having legitimately acquired a non-zero net position in the market, the trader sells shares repeatedly through a sequence of wallets under common ownership and then closes out the position at the prevailing price." > "In both of the above cases, there is the possibility of an 'interception' in the following scenario: A trader who intends to execute a wash trade pings the Polymarket API to get the best bid and ask prices. Before the trader submits orders for two wallets under their control, a third, unaffiliated wallet places a limit order within the bid-ask spread." So even when the algorithm flags "wash trading," it might just be catching legitimate market makers who happened to trade with someone attempting a wash trade. The paper gives examples: > "Example 1 (Will the Republican candidate win Pennsylvania by 1.0%-1.5%?). As shown in Table 2, MAY175 first buys 7,291.07 shares with MAY20. MAY175 then trades its 'No' shares with MAY176 repeatedly, alternating as buyer and seller. After 90 such trades - over a 30-minute period during which there are only two non-MAY trades in the market - MAY176's buy order for the 'No' shares appears to be intercepted by 0x203...cd1." So even in their cleanest example of wash trading, a legitimate trader intercepted the wash trade and took their money. PART 3: The Smoking Gun - What Kalshi Doesn't Want You to Know The Timeline That Destroys Kalshi's Narrative Let me lay this out chronologically: July 2024: Wash trading begins trending upward on Polymarket December 2024: Wash trading peaks at ~60% of weekly volume April 2025: Wash trading drops to under 5% May-September 2025: Wash trading remains minimal October 2025: Brief spike to ~20% November 6, 2025: Columbia paper published November 2025: Kalshi starts pushing fake Bloomberg headlines about wash trading Notice the problem? By the time Kalshi started their smear campaign, the wash trading had already been resolved for 6+ months. What the Paper Says About Why It Stopped The researchers note: > "From June until late September 2025, detected wash trading accounted for less than 5% of weekly volume (this may be because Polymarket made efforts to curb wash trading, or because wash-trading wallets no longer close their open positions or trade exclusively with each other)." So either: 1. Polymarket fixed it 2. Airdrop farmers got smarter about not getting caught 3. The airdrop incentive diminished In any case, the problem largely resolved itself before Kalshi even started talking about it. The Markets That Actually Mattered Were Clean Let's go back to that table from the paper. The top 50 markets by volume: ZERO detected wash trading: - Will Donald Trump win the 2024 US Presidential Election? (1,568.7M shares, $1,184.0M dollars) - Will Kamala Harris win the 2024 US Presidential Election? (1,072.0M shares, $634.8M dollars) - Will Donald Trump be inaugurated? (400.4M shares, $324.2M dollars) - Will Zelenskyy wear a suit before July? (242.2M shares, $156.9M dollars) High wash trading detected: - Will Nicolae Ciucă win Romanian Presidential election? (326.5M shares, $2.6M dollars - 98.5% wash) - Will the Sacramento Kings win the 2025 NBA Finals? (378.0M shares, $34.6M dollars - 93.0% wash) See the pattern? The high-dollar-volume markets that actually drove Polymarket's growth were clean. The wash trading was concentrated in low-liquidity, high-share-count markets where people were farming airdrops. The "fengchu" Cluster: Follow the Money The paper documents one of the largest wash trading operations: > "In another instance, we discover a large network of 1,028 trading wallets which collectively traded 792M of share volume ($407M of dollar volume) almost exclusively in sports markets, starting October 23, 2024 and with a cumulative loss of only $511.31." > "Their capitalization can be traced to the wallet with display name 'fengchu', which transfers approximately 5,000 USDC to each of six children - named 'fdetdddw', 'duichong', 'DuiChong1', 'duic', 'miya', and 'DuiDui'." This is a single entity running 1,028 wallets to farm an airdrop. They traded $407M in dollar volume and lost $511. This is not market manipulation in any meaningful sense. This is one person (or group) running a bot farm to qualify for free tokens. And they're doing it so inefficiently they're barely breaking even. The Researchers' Own Caveat About Market Impact Here's what the researchers say about the impact of this activity: > "When a wash trader places executable orders within the current prevailing bid-ask spread, this contributes neither liquidity nor information to the prediction market." But they also note: > "It is possible that wash traders no longer close their open positions or trade exclusively with each other." And most importantly: > "Until such time as the authenticity of trades can be quickly and reliably established, it may be better to rely on less manipulable measures of platform activity such as open interest, which cannot be inflated without limit by recycling capital across multiple trades." Open interest - the total value of outstanding positions - stayed healthy throughout this period (Figure 23 in the paper). Meaning real money was still in the markets, even when wash trading volume was high. PART 4: Why This Matters (And Why Kalshi Is Terrified) The Real Story: Polymarket Grew Despite Wash Trading, Not Because of It Here's what actually happened: 1. July-December 2024: Airdrop farmers inflate share volume in low-liquidity markets 2. November 2024: Presidential election drives MASSIVE legitimate volume to Polymarket 3. December 2024: Wash trading peaks, but in markets nobody cares about 4. April 2025: Wash trading drops precipitously 5. May-September 2025: Polymarket continues growing with clean volume 6. November 2025: Academic paper documents the wash trading (that already stopped) 7. November 2025: Kalshi manufactures fake headlines to weaponize the paper What Kalshi Is Really Scared Of The Columbia paper actually makes Polymarket look BETTER, not worse: 1. The wash trading was concentrated in irrelevant markets 2. The high-profile markets were clean 3. Polymarket detected and addressed it 4. The problem resolved itself before it became systemic 5. Real money and real users drove the platform's growth Compare this to Kalshi: - Lower volume - Less liquidity - Fewer users - Higher fees - And instead of building product, they're manufacturing fake Bloomberg headlines The Airdrop Farming Is Actually Proof of Demand Here's the irony Kalshi misses: If people are willing to run 1,028-wallet bot farms to farm a Polymarket airdrop, that's proof the Polymarket token will have value. Nobody farms airdrops for tokens they think will be worthless. The wash trading is actually a bullish signal about Polymarket's future tokenomics. The Data Kalshi Hopes You Don't See From the paper's Figure 7 and Figure 30: Overall estimated wash volume by week: - Peak (December 2024): ~60% - May 2025: <5% - June-September 2025: <5% - October 2025: ~20% Estimated wash volume by category (all-time): - Sports: 45% - Elections: 17% - Politics: 12% - Crypto: 3% But here's the key: Election and Politics markets - the ones that drove Polymarket's mainstream adoption - had the LOWEST wash trading rates. The Presidential Election specifically? 0% detected wash trading. The Columbia research paper actually vindicates Polymarket more than it indicts them: - Wash trading happened, but mostly in irrelevant markets - The markets people cared about were clean - The problem resolved itself quickly - Polymarket continued growing with legitimate volume Meanwhile, Kalshi is so desperate to slow Polymarket's growth that they're paying X accounts to manufacture fake Bloomberg headlines about a wash trading problem that already stopped six months ago. The market doesn't lie. Blockchain data doesn't lie. Academic research (when you actually read it) doesn't lie. The only people lying are Kalshi's marketing team. And unlike wash trading, you can't detect and remove a competitor's fake headlines with an algorithm. You have to do it the old-fashioned way: by calling them out publicly, with receipts. TL;DR: - Columbia researchers found wash trading on Polymarket peaked at 60% in Dec 2024, dropped to <5% by May 2025 - Presidential Election markets (the ones that mattered) had ZERO detected wash trading - Wash trading was concentrated in penny-stock equivalent markets for airdrop farming - The algorithm admits it can't definitively prove transactions are wash trades - Kalshi manufactured fake Bloomberg headlines about this in November 2025 - six months after the problem resolved - This is Kalshi's second smear campaign (after the FBI narrative in 2024) - Multiple sources confirm this is coordinated counter-marketing by Kalshi Kalshi's marketing budget: Millions Kalshi's product improvements: Unclear Polymarket's response: Continued growth and zero fucks given The data: Publicly available for anyone to verify

🚨 ALERT: A Columbia University study found that up to 60% of Polymarket’s trading volume may come from wash trading, raising concerns about artificial activity in prediction markets.

> trump gets elected > elon starts visiting the white house pitching his ideas on DOGE > elon becomes frustrated because trump is all talk (shocker) > elon tweets that he no longer supports him > trump vs elon twitter battle of the century

Not every day the Dunder Mifflin CFO stumbles onto Ramp’s campaign.