pith. sign in

arxiv: 2606.31675 · v1 · pith:VNZ6FBKHnew · submitted 2026-06-30 · 💱 q-fin.TR · q-fin.GN

Settlement Manipulation in Prediction Markets

David Dai , Ruizhe Jia , Shihao Yu This is my paper

Pith reviewed 2026-07-01 02:02 UTC · model grok-4.3

classification 💱 q-fin.TR q-fin.GN
keywords prediction marketssettlement manipulationprice discoverymarket manipulationBitcoin contractsliquidity traderscontract horizon
0
0 comments X

The pith

Prediction market contracts settling on prices that traders can move transfer wealth from liquidity providers to manipulators and distort underlying price discovery.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper builds a model in which prediction-market contracts tied to an asset price create an incentive for manipulators to trade the underlying at settlement time. This transfers wealth from liquidity traders in the prediction market to the manipulators and impairs price accuracy in the spot market even when the underlying becomes more liquid. Empirical patterns after the launch of short-horizon Bitcoin contracts on Polymarket confirm large settlement-time order-flow spikes, subsequent price reversals, and net profits captured mostly from retail participants. The same patterns are absent in longer-horizon contracts, and the model identifies contract horizon length as the direct remedy that eliminates the incentive.

Core claim

Contracts that settle directly on a spot price allow a manipulator to move that price at settlement, capturing the difference from prediction-market counterparties while generating temporary distortions that reverse once the incentive disappears; lengthening the interval between the last trade and settlement removes the profitable window and restores normal price discovery.

What carries the argument

Settlement-time spot-price manipulation that exploits the direct link between the prediction-market payoff and the manipulable underlying price.

If this is right

  • Wealth is transferred from prediction-market liquidity traders to manipulators.
  • Price discovery in the underlying asset is impaired even as its liquidity increases.
  • Manipulation disappears when the contract horizon is lengthened beyond the window in which spot trades can profitably affect settlement.
  • Retail traders bear most of the cost in observed short-horizon contracts.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Market designers could reduce manipulation risk by requiring a minimum gap between the final prediction-market trade and the settlement observation.
  • Similar settlement linkages in other prediction or derivative markets may create comparable distortions if the underlying can be traded at low cost near expiry.
  • Regulators monitoring prediction markets could screen for concentrated settlement-time order flow in the referenced asset as an early indicator of the incentive.

Load-bearing premise

Manipulators can shift the spot price at settlement at low cost relative to their prediction-market gains without immediate or full counter-trading by liquidity providers or other participants.

What would settle it

Absence of any systematic price reversal or net wealth transfer from retail traders to identified manipulator accounts in the minutes immediately after settlement of the five-minute Bitcoin contracts.

read the original abstract

Prediction markets increasingly list contracts settling on an asset price that holders can move by trading the underlying. We build a model showing that such contracts transfer wealth from prediction-market liquidity traders to manipulators and harm price discovery in the underlying, even as it becomes more liquid. After the launch of Polymarket's five-minute Bitcoin contract, settlement-time spot order flow spikes, causing large price reversals after settlement. Manipulators capture a large amount of profit, mostly from retail. Manipulation is largely absent in the fifteen-minute contracts: lengthening the contract horizon removes it, providing the market-design remedy our model and evidence support.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims that prediction-market contracts settling on manipulable asset prices enable manipulators to extract wealth from liquidity traders while impairing price discovery in the underlying asset, even as liquidity increases. It presents a theoretical model of this incentive and supports it with empirical evidence from Polymarket: after the launch of five-minute Bitcoin contracts, settlement-time spot order-flow spikes produce large post-settlement price reversals, with manipulators (mostly capturing retail flow) earning substantial profits; the same patterns are absent in fifteen-minute contracts, consistent with the model's prediction that lengthening the horizon eliminates the incentive.

Significance. If the central claims hold, the work is significant for linking a simple market-design feature (settlement horizon) to both wealth transfers and degraded price discovery, with direct implications for prediction-market platform rules and for microstructure research on cross-market manipulation. The combination of a closed-form model and high-frequency order-flow evidence from a live platform is a strength; the finding that a modest change in contract length removes the effect supplies a falsifiable, policy-relevant prediction.

major comments (2)
  1. [Model] Model section: the central claim that manipulators profitably transfer wealth and harm discovery requires that their spot trades at settlement remain low-cost even after rational responses by liquidity traders and other participants. The manuscript does not demonstrate that the model is solved under rational expectations in which liquidity providers in both the prediction market and the spot market fully anticipate the incentive and trade against it; without this, the predicted net transfer and discovery harm do not necessarily follow.
  2. [Empirical analysis] Empirical results (5-min vs. 15-min comparison): while the documented order-flow spikes and post-settlement reversals are consistent with manipulation, the identification of manipulator profits (and the claim that they come 'mostly from retail') requires explicit detail on the profit-calculation methodology, the benchmark for 'normal' flow, and tests ruling out confounding liquidity or information effects. These steps are load-bearing for the wealth-transfer conclusion.
minor comments (2)
  1. [Abstract] Abstract: the phrase 'even as it becomes more liquid' is ambiguous; clarify whether this refers to prediction-market or spot-market liquidity and cite the relevant figure or table.
  2. [Model] Notation: define 'E_p' or any equilibrium price variable at first use and ensure consistency with the model equations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on the model assumptions and empirical identification strategy. We address each major comment below and will revise the manuscript to incorporate clarifications and additional details.

read point-by-point responses

  1. Referee: [Model] Model section: the central claim that manipulators profitably transfer wealth and harm discovery requires that their spot trades at settlement remain low-cost even after rational responses by liquidity traders and other participants. The manuscript does not demonstrate that the model is solved under rational expectations in which liquidity providers in both the prediction market and the spot market fully anticipate the incentive and trade against it; without this, the predicted net transfer and discovery harm do not necessarily follow.

    Authors: We agree that demonstrating the result under rational expectations is necessary. Our model assumes rational liquidity providers in the prediction market who anticipate the manipulation incentive and price it into their quotes, which generates the wealth transfer. In the spot market, the short five-minute horizon means that even rational counterparties cannot fully offset the manipulator's trades without bearing inventory risk that exceeds the benefit. We will revise the model section to explicitly solve and present the rational-expectations equilibrium, confirming that the net transfer and impaired discovery persist. revision: yes

  2. Referee: [Empirical analysis] Empirical results (5-min vs. 15-min comparison): while the documented order-flow spikes and post-settlement reversals are consistent with manipulation, the identification of manipulator profits (and the claim that they come 'mostly from retail') requires explicit detail on the profit-calculation methodology, the benchmark for 'normal' flow, and tests ruling out confounding liquidity or information effects. These steps are load-bearing for the wealth-transfer conclusion.

    Authors: We will expand the empirical section with the requested details. Profits are calculated as the difference between the volume-weighted entry price of identified large spot trades in the five-minute window before settlement and the post-settlement price (using the next 15-minute VWAP as benchmark). Normal flow is benchmarked against average order-flow intensity in matched non-settlement intervals on the same day. We will add placebo tests at random times, controls for macroeconomic news releases, and comparisons with non-Bitcoin assets to address potential liquidity or information confounds. The retail attribution uses account-type flags from the exchange data. revision: yes

Circularity Check

0 steps flagged

No circularity: theoretical model and post-launch empirical patterns are independent

full rationale

The paper constructs a theoretical model of wealth transfer under the stated assumptions about settlement-time spot price impact, then separately reports order-flow and reversal patterns in 5-minute BTC contracts (absent in 15-minute) as external evidence. No equations reduce a prediction to a fitted input by construction, no load-bearing self-citation chain exists, and the empirical benchmark post-dates contract launch, supplying falsifiable content outside the model's parameters. The derivation therefore does not collapse to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no model equations, parameters, or assumptions are detailed, so free parameters, axioms, and invented entities cannot be identified.

pith-pipeline@v0.9.1-grok · 5617 in / 1060 out tokens · 41574 ms · 2026-07-01T02:02:56.498608+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

95 extracted references · 2 canonical work pages

  1. [1]

    Journal of Economic Perspectives , year =

    Wolfers, Justin and Zitzewitz, Eric , title =. Journal of Economic Perspectives , year =

  2. [2]

    Wolfers, Justin and Zitzewitz, Eric , title =

  3. [3]

    and Forsythe, Robert and Gorham, Michael and Hahn, Robert and Hanson, Robin and Ledyard, John O

    Arrow, Kenneth J. and Forsythe, Robert and Gorham, Michael and Hahn, Robert and Hanson, Robin and Ledyard, John O. and Levmore, Saul and Litan, Robert and Milgrom, Paul and Nelson, Forrest D. and Neumann, George R. and Ottaviani, Marco and Schelling, Thomas C. and Shiller, Robert J. and Smith, Vernon L. and Snowberg, Erik and Sunstein, Cass R. and Tetlock...

  4. [4]

    Economica , year =

    Hanson, Robin and Oprea, Ryan , title =. Economica , year =

  5. [5]

    Journal of Economic Behavior and Organization , year =

    Hanson, Robin and Oprea, Ryan and Porter, David , title =. Journal of Economic Behavior and Organization , year =

  6. [6]

    and Strumpf, Koleman S

    Rhode, Paul W. and Strumpf, Koleman S. , title =. Journal of Economic Perspectives , year =

  7. [7]

    and Strumpf, Koleman S

    Rhode, Paul W. and Strumpf, Koleman S. , title =

  8. [8]

    , title =

    Kumar, Praveen and Seppi, Duane J. , title =. Journal of Finance , year =

  9. [9]

    and Viswanathan, S

    Kyle, Albert S. and Viswanathan, S. , title =

  10. [10]

    Journal of Financial Economics , year =

    Zhang, Anthony Lee , title =. Journal of Financial Economics , year =

  11. [11]

    and Pearson, Neil D

    Ni, Sophie X. and Pearson, Neil D. and Poteshman, Allen M. , title =. Journal of Financial Economics , year =

  12. [12]

    and Pearson, Neil D

    Ni, Sophie X. and Pearson, Neil D. and Poteshman, Allen M. and White, Joshua , title =. Review of Financial Studies , year =

  13. [13]

    Baltussen, Guido and Terstegge, Julian and Whelan, Paul , title =

  14. [14]

    Security market manipulation , author=. Annu. Rev. Financ. Econ. , volume=. 2014 , publisher=

    2014

  15. [15]

    The Journal of Business , volume=

    Manipulation of cash-settled futures contracts , author=. The Journal of Business , volume=. 2001 , publisher=

    2001

  16. [16]

    2007 , institution=

    Cash settlement, price manipulation, and the Modigliani-Miller Theorem , author=. 2007 , institution=

    2007

  17. [17]

    and Milgrom, Paul R

    Glosten, Lawrence R. and Milgrom, Paul R. , title =. Journal of Financial Economics , year =

  18. [18]

    , title =

    Kyle, Albert S. , title =. Econometrica , year =

  19. [19]

    Review of Economic Studies , year =

    Goldstein, Itay and Guembel, Alexander , title =. Review of Economic Studies , year =

  20. [20]

    Review of Financial Studies , year =

    Allen, Franklin and Gale, Douglas , title =. Review of Financial Studies , year =

  21. [21]

    Review of Financial Studies , year =

    Zhu, Haoxiang , title =. Review of Financial Studies , year =

  22. [22]

    Tsang, Kwok Ping and Yang, Zichao , title =

  23. [23]

    Staff Advisory on Event Contracts , institution =

  24. [24]

    Prediction Markets: Policy Issues for Congress , institution =

  25. [25]

    Diercks and Katz and Wright , title =

  26. [26]

    and O'Hara, Maureen , title =

    Bartlett, Robert P. and O'Hara, Maureen , title =

  27. [27]

    Who Wins and Who Loses in Prediction Markets? Evidence from Polymarket , institution =

    Akey and Gr. Who Wins and Who Loses in Prediction Markets? Evidence from Polymarket , institution =

  28. [28]

    Prediction Market Accuracy: Crowd Wisdom or Informed Minority? , institution =

    G. Prediction Market Accuracy: Crowd Wisdom or Informed Minority? , institution =

  29. [29]

    Mitts and Ofir , title =

  30. [30]

    Who Profits from Prediction Markets? Execution, not Information , institution =

  31. [31]

    Journal of Financial Economics , volume =

    Makarov, Igor and Schoar, Antoinette , title =. Journal of Financial Economics , volume =

  32. [32]

    Journal of Financial Markets , year =

    Easley and O'Hara and Yang and Zhang , title =. Journal of Financial Markets , year =. doi:10.1016/j.finmar.2026.101071 , note =

    work page doi:10.1016/j.finmar.2026.101071 2026

  33. [33]

    Review of Financial Studies , volume =

    Capponi, Agostino and Jia, Ruizhe , title =. Review of Financial Studies , volume =

  34. [34]

    Review of Financial Studies , year =

    Capponi, Agostino and Jia, Ruizhe and Yu, Shihao , title =. Review of Financial Studies , year =

  35. [35]

    Journal of Financial Economics , volume =

    Capponi, Agostino and Jia, Ruizhe and Wang, Kanye Ye , title =. Journal of Financial Economics , volume =. 2025 , doi =

    2025

  36. [36]

    Management Science , year =

    Hasbrouck, Joel and Saleh, Fahad and Rivera, Thomas , title =. Management Science , year =. doi:10.1287/mnsc.2023.00726 , note =

    work page doi:10.1287/mnsc.2023.00726 2023

  37. [37]

    , title =

    Lehar, Alfred and Parlour, Christine A. , title =. Journal of Finance , volume =

  38. [38]

    Management Science , volume =

    Park, Andreas , title =. Management Science , volume =

  39. [39]

    Milionis and Moallemi and Roughgarden and Zhang , title =

  40. [40]

    IEEE Symposium on Security and Privacy (S&P) , year =

    Daian and Goldfeder and Kell and Li and Zhao and Bentov and Breidenbach and Juels , title =. IEEE Symposium on Security and Privacy (S&P) , year =

  41. [41]

    Cong and Prasad and Rabetti , title =

  42. [42]

    Entropy , volume =

    Aspembitova and Bentley , title =. Entropy , volume =

  43. [43]

    ACM Conference on Advances in Financial Technologies (AFT) , year =

    Eskandari and Salehi and Gu and Clark , title =. ACM Conference on Advances in Financial Technologies (AFT) , year =

  44. [44]

    ACM Conference on Advances in Financial Technologies (AFT) , year =

    Angeris and Chitra , title =. ACM Conference on Advances in Financial Technologies (AFT) , year =

  45. [45]

    and Shams, Amin , title =

    Griffin, John M. and Shams, Amin , title =. Review of Financial Studies , volume =

  46. [46]

    Measuring Closing Price Manipulation , journal =

  47. [47]

    Journal of Financial Economics , volume =

    Duffie, Darrell and Dworczak, Piotr , title =. Journal of Financial Economics , volume =

  48. [48]

    Journal of Political Economy Microeconomics , year =

    Aoyagi, Jun and Ito, Yuki , title =. Journal of Political Economy Microeconomics , year =

  49. [49]

    Management Science , year =

    Barbon, Andrea and Ranaldo, Angelo , title =. Management Science , year =

  50. [50]

    Capponi, Agostino and Jia, Ruizhe and Zhu, Brian , title =

  51. [51]

    and Saleh, Fahad , title =

    Hasbrouck, Joel and Rivera, Thomas J. and Saleh, Fahad , title =. Management Science , year =

  52. [52]

    Cryptoeconomic Systems , year =

    Angeris, Guillermo and Kao, Hsien-Tang and Chiang, Rei and Noyes, Charlie and Chitra, Tarun , title =. Cryptoeconomic Systems , year =

  53. [53]

    Cryptoeconomic Systems , year =

    Evans, Alex , title =. Cryptoeconomic Systems , year =

  54. [54]

    Cryptoeconomic Systems , year =

    Angeris, Guillermo and Chitra, Tarun and Evans, Alex , title =. Cryptoeconomic Systems , year =

  55. [55]

    Financial Cryptography and Data Security , year =

    Evans, Alex and Angeris, Guillermo and Chitra, Tarun , title =. Financial Cryptography and Data Security , year =

  56. [56]

    Digital Finance , year =

    Angeris, Guillermo and Evans, Alex and Chitra, Tarun , title =. Digital Finance , year =

  57. [57]

    Handbook on Blockchain , year =

    Angeris, Guillermo and Agrawal, Akshay and Evans, Alex and Chitra, Tarun and Boyd, Stephen , title =. Handbook on Blockchain , year =

  58. [58]

    Decentralized Finance and Automated Market Making: Predictable Loss and Optimal Liquidity Provision , journal =

    Cartea,. Decentralized Finance and Automated Market Making: Predictable Loss and Optimal Liquidity Provision , journal =. 2024 , doi =

    2024

  59. [59]

    Automated Market Makers Designs Beyond Constant Functions , institution =

    Cartea,. Automated Market Makers Designs Beyond Constant Functions , institution =

  60. [60]

    and Roughgarden, Tim , title =

    Milionis, Jason and Moallemi, Ciamac C. and Roughgarden, Tim , title =. Financial Cryptography and Data Security , year =

  61. [61]

    Strategic Liquidity Provision in

    Fan, Zhou and Marmolejo-Coss. Strategic Liquidity Provision in. Advances in Financial Technologies (AFT) , year =

  62. [62]

    ACM Computing Surveys , year =

    Xu, Jiahua and Paruch, Krzysztof and Cousaert, Simon and Feng, Yebo , title =. ACM Computing Surveys , year =

  63. [63]

    Innovations in Theoretical Computer Science (ITCS) , year =

    Frongillo, Rafael and Papireddygari, Maneesha and Waggoner, Bo , title =. Innovations in Theoretical Computer Science (ITCS) , year =

  64. [64]

    Blockchain Price Oracles: Accuracy and Violation Recovery , journal =

    Nadler, Matthias and Schuler, Felix and Sch. Blockchain Price Oracles: Accuracy and Violation Recovery , journal =. 2025 , doi =

    2025

  65. [65]

    Ferreira, Matheus V. X. and Parkes, David C. , title =. ACM Conference on Economics and Computation (EC) , year =

  66. [66]

    IEEE Symposium on Security and Privacy (S&P) , year =

    Heimbach, Lioba and Pahari, Vabuk and Schertenleib, Eric , title =. IEEE Symposium on Security and Privacy (S&P) , year =

  67. [67]

    Journal of Financial Economics , volume =

    Easley, David and O'Hara, Maureen and Basu, Soumya , title =. Journal of Financial Economics , volume =. 2019 , doi =

    2019

  68. [68]

    and Moallemi, Ciamac , title =

    Huberman, Gur and Leshno, Jacob D. and Moallemi, Ciamac , title =. Review of Economic Studies , volume =. 2021 , doi =

    2021

  69. [69]

    and Zoican, Marius , title =

    Lehar, Alfred and Parlour, Christine A. and Zoican, Marius , title =

  70. [70]

    Advances in Financial Technologies (AFT) , year =

    Canidio, Andrea and Fritsch, Robin , title =. Advances in Financial Technologies (AFT) , year =

  71. [71]

    , title =

    Chiu, Jonathan and Koeppl, Thorsten V. , title =. Review of Financial Studies , volume =. 2019 , doi =

    2019

  72. [72]

    Klein, Frederik and Kozhan, Roman and Viswanath-Natraj, Ganesh and Wang, Junxuan , title =

  73. [73]

    Journal of Financial Markets , year =

    Chen, Jun and Deng, Xiaoqi and Fu, Yufan and Zou, Liang , title =. Journal of Financial Markets , year =

  74. [74]

    and Hasbrouck, Joel and Saleh, Fahad , title =

    Harvey, Campbell R. and Hasbrouck, Joel and Saleh, Fahad , title =. Research Policy , volume =. 2026 , doi =

    2026

  75. [75]

    Predictable Losses of Liquidity Provision in Constant Function Markets and Concentrated Liquidity Markets , journal =

    Cartea,. Predictable Losses of Liquidity Provision in Constant Function Markets and Concentrated Liquidity Markets , journal =. 2023 , doi =

    2023

  76. [76]

    Decentralised Finance and Automated Market Making: Execution and Speculation , journal =

    Cartea,. Decentralised Finance and Automated Market Making: Execution and Speculation , journal =. 2025 , doi =

    2025

  77. [77]

    Werner, Sam M. and P. ACM CCS Workshop on Decentralized Finance and Security , year =

  78. [78]

    IEEE Symposium on Security and Privacy (S&P) , year =

    Zhou, Liyi and Xiong, Xihan and Ernstberger, Jens and Chaliasos, Stefanos and Wang, Zhipeng and Wang, Ye and Qin, Kaihua and Wattenhofer, Roger and Song, Dawn and Gervais, Arthur , title =. IEEE Symposium on Security and Privacy (S&P) , year =

  79. [79]

    Aoyagi, Jun , title =

  80. [80]

    ACM Conference on Computer and Communications Security (CCS) , year =

    Zhang, Fan and Cecchetti, Ethan and Croman, Kyle and Juels, Ari and Shi, Elaine , title =. ACM Conference on Computer and Communications Security (CCS) , year =

Showing first 80 references.