Susquehanna Backs Prediction Markets With Balance Sheet Risk

Susquehanna以资产负债表风险支持预测市场 Susquehanna承担方向性风险库存，并借助超级预测者的价格信号向企业客户提供数百万美元的定制对冲服务。 本期内容探讨了Susquehanna如何作为主要做市商和流动性提供商，在预测市场中弥合散户活动与机构对冲之间的鸿沟。Maletz详细介绍了来自超级预测者的众包价格发现如何帮助Susquehanna进行模型构建，并针对从天气灾害到地缘政治事件等风险，提供数千万美元的定制化风险承接额度。讨论涵盖了做市机制、风险仓储、合规考量，以及二元结果合约如何提供比传统替代性资产组合更精准的对冲手段。最后，对话强调了预测平台在对冲特定细分且真实的经济风险时所具有的前所未有的快速响应能力。 预测市场的机构普及仍受限于浅薄的流动性和订单簿深度，形成了“先有鸡还是先有蛋”的困境：机构因交易量低而望而却步，而没有机构的参与交易量又持续低迷。 (0:00) Susquehanna International Group (SIG) 作为主要做市商和引导者，致力于弥合预测市场中买卖双方在时间和交易规模上的差距。 (4:47) 依托创始人Jeff Yass建立的概率与贝叶斯思维，Susquehanna利用其雄厚的资产负债表，为宏观事件合约承接并储存显著的单向方向性风险。 (7:35) 预测市场有效扮演了由超级预测者驱动的价格发现引擎，让做市商能够验证低流动性合约的公允价值，并向企业对冲客户提供数百万美元的定制化场外风险承接能力。 (9:27) 机构预测市场交易通常通过场内大宗交易或场外OTC掉期执行，并通过银行、经纪商和保险渠道等传统金融中介进行分销。 (11:51)

Susquehanna International Group is leveraging its options market-making background and balance sheet to transform prediction markets from retail-dominated novelty venues into institutional hedging infrastructure. According to Jeremy Maletz, SIG's head of macro trading and prediction markets, the firm pairs crowd-driven price discovery with large-scale risk warehousing to offer corporate clients direct hedges on specific real-world outcomes. Warehousing Directional Risk to Bridge Liquidity (4:47) Prediction markets face a chicken-and-egg problem: institutional hedgers avoid them due to shallow order books, yet order books remain illiquid without institutional flow. Susquehanna addresses this by serving as a dedicated market maker across regulated platforms like Kalshi, matching trades across time and size. Rooted in the probabilistic, Bayesian culture established by founder Jeff Yass, SIG departs from traditional risk-neutral market making by willingly 'warehousing' unidirectional risk. When macroeconomic shocks prompt participants to seek protection on the same side of a contract, SIG relies on its deep balance sheet to absorb and hold the opposite exposure. Converting Superforecaster Signals into Institutional Block Trades (9:27) Corporate entities needing tens of millions of dollars in protection—such as airlines hedging winter weather disruptions or firms hedging shipping chokepoints like the Strait of Hormuz—cannot rely on public order books containing only small retail sums. Maletz explains that SIG uses active superforecasters primarily for price discovery. Because knowledgeable participants establish fair value even in low-volume contracts, SIG validates those price signals through internal quantitative modeling and then provides large bespoke risk capacity off-exchange. To satisfy corporate compliance mandates, these transactions are structured as on-exchange block trades or OTC swaps arranged through traditional banks and brokers. Pricing Mechanics and Guardrails Against Manipulation (20:00) SIG prices contracts through a mix of proxy translation from traditional benchmarks (such as Federal Reserve rate expectations) and ground-up computational models for novel assets like global compute capacity. Drawing on poker strategy, traders evaluate counterparty incentives and monitor order flow to update Bayesian probabilities. Maletz notes that insider trading is far more conspicuous in hyper-specific event markets than in equities, allowing firms to report anomalies to regulators. To protect its franchise, SIG adheres to strict institutional compliance—refusing illicit web scraping—and systematically avoids easily manipulable 'mention markets' where an individual can single-handedly trigger contract settlement. Operational Agility Versus Imperfect Financial Proxies (25:00) Susquehanna's prediction market operations expanded out of its macro desk around the 2016 U.S. presidential election. At the time, investment banks structured equity baskets to hedge political outcomes under the consensus that a Donald Trump victory would cause a 5% to 7% crash. When markets quickly rebounded, these proxy baskets failed to hedge underlying risks, whereas binary contracts settled cleanly on the actual outcome. In addition to precision, Maletz emphasizes speed: while traditional futures exchanges often take nearly a year to approve new hedging contracts, modern event platforms can deploy contracts for niche variables like tariffs, shipping rates, or DRAM prices within hours. Although sports contracts still represent more than 50% of prediction exchange volume, Maletz views retail activity as a necessary infrastructure subsidy that boosts overall platform liquidity. The broader institutional test depends on whether intermediaries and corporate risk managers continue migrating from indirect financial proxies toward direct, binary event hedges.

## Background 该内容由彭博社（Bloomberg）于 2026 年 6 月 6 日在 RSS / Web 播客平台发布。节目简介指出，预测市场正逐步普及，但能否容纳对冲基金等大型机构投资者仍存疑问；针对盘口浅与流动性不足的痛点，本期节目在纽约 City Winery 现场录制，邀请老牌做市巨头萨斯奎汉纳（Susquehanna International Group, SIG）宏观交易与预测市场主管 Jeremy Maletz，深度探讨其与 Kalshi 展开的做市业务、机构对冲用例、风险仓储及做市盈利模式。 ## TL;DR 老牌做市商萨斯奎汉纳（SIG）以其深厚的贝叶斯概率文化为底座，正式切入预测市场并充当向华尔街机构示范的“领路人”角色。针对预测市场因流动性不足而阻碍机构对冲真实经济风险的死结，SIG 凭借庞大的资产负债表承担起跨时间与规模的撮合中介，主动进行单边“风险仓储”，并通过场内大宗交易和场外掉期（Swap）为企业提供千万美元级的承接能力。访谈还详细拆解了做市机制的差异化定价、利用贝叶斯推断识别内幕交易并坚守合规底线的原则，以及相较于耗时漫长的传统期货，预测市场以极速推新模式在算力（GPU）租赁和非标大宗商品对冲中展现出的颠覆性优势。 ## Atmosphere & style 现场录制于纽约 City Winery，气氛轻松热烈且充满机智幽默的现场互动，主持人与极少公开露面的顶级交易员在笑声与调侃中碰撞出高密度的专业思考。对话既有华尔街资深操盘手的严谨、克制与对合规底线的恪守，又透露出用概率解构世间万物的德州扑克博弈哲思。整场探讨从通俗生动的日常比喻层层递进至深奥的金融工程逻辑，真实展现了传统顶级做市机构在面对前沿金融创新时的敏锐洞察与务实审慎。 ## Bulletin - `00:23` 老牌做市巨头萨斯奎汉纳（SIG）入局合规预测市场（如 Kalshi），不仅提供双边报价资金，更利用其雄厚资产负债表承担“风险仓储”职能，破除机构因盘口浅而无法大额对冲的先有鸡后有蛋困境。 - `30:00` 预测市场推新速度远超传统交易所，将非标对冲工具的挂牌周期从一年压缩至数小时，正加速向 GPU 算力租赁、DRAM 现货价格及海运运费等缺乏标准化期货的领域渗透。 - `12:15` 在机构落地架构上，大额真实风险对冲不必直接在交易所盘口打印，而是主要通过场内大宗交易（Block Trade）及场外金融互换（Swap）形式完成，预测市场平台仅收取数据授权许可费。 - `32:30` 相比传统金融通过一揽子股票（Basket Trades）对冲政治或宏观事件容易遭遇基差风险和逻辑落空，预测市场的二元纯粹结算机制彻底规避了资产间相关性脱节的缺陷。 ## Key takeaways - `00:23` 预测市场面临结构性流动性瓶颈：虽然体育博彩天然具备散户流动性，但宏观经济与商业运营合约盘口较浅，导致实体企业与机构投资者此前无法进行规模化对冲。 - `03:52` SIG 的预测市场做市根植于联合创始人 Jeff Yass 奠定的贝叶斯概率文化，坚持世间万事皆可拆解为胜率与赔率的离散概率模型。 - `05:44` 做市商的本质价值在于跨越“时间”与“规模”撮合交易，中介零散交易意愿并承担持仓波动成本，纯 P2P 模式在现实中难以承载大体量需求。 - `06:40` 与受制于资本而追求绝对风险中性（Risk Neutral）的中小做市商不同，SIG 拥有雄厚资产负债表，能够在单边恐慌下为市场扮演“风险仓储者”角色。 - `10:00` 预测市场的核心价值在于信息本身与超级预测者（Superforecasters）的价格发现机制，哪怕合约成交量较小，公允价格确立后做市商即可提供千万美元级承接力。 - `12:15` 针对机构难以直接穿透合规防火墙的问题，SIG 依托投行、主经纪商及保险中介网络，利用场外互换（Swap）与场内大宗交易实现机构流动性交付。 - `16:35` 专业做市商不涉足无经济价值的娱乐八卦（如名人婚期），而是聚焦于算力供需、地缘政治与宏观指标等具备实质实体对冲需求的事件。 - `20:00` 非体育类的真实世界事件市场正呈现出高于体育博彩的增速，高频体育投注所带来的流量与认知度正有效带动严肃宏观品类的繁荣。 - `20:00` 预测市场做市采取分层定价逻辑：对利率等宏观标的映射传统金融代理，对特异事件则借鉴德州扑克思维研判超级预测者下注并搭建独立量化模型。 - `23:00` 合规监管平台与无 KYC 的 DeFi 平台存在本质风控差异，SIG 仅在受监管平台展业，并通过异常资金流的动机偏差进行贝叶斯更新以侦查并举报内幕交易。 - `26:15` 对于极易被现实串谋直接操纵的投机品（如播客是否提及特定词汇的“提及市场”），专业做市机构会严格从准入名单中剔除。 - `28:45` 大型持牌机构恪守网站服务条款与合规红线，绝不采取游资或独立交易员常用的违规爬虫与嗅探未发布 URL 等灰色套利手段。 - `30:00` 预测市场核心竞争力在于敏捷推新速度，彻底解决了传统交易所长达数月乃至一年的上市审批无法响应突发供应链或关税风险的痛点。 - `32:30` 2016年大选华尔街一揽子概念股对冲全面失效的历史，验证了传统宏观代理组合脆弱的基差风险，凸显出直接事件合约的点对点纯粹对冲价值。 ## Detailed walkthrough ### `00:23` 交易品种大爆炸与预测市场的机构化难题 在纽约 City Winery 举办的《Odd Lots》现场演出中，主持人 Joe Weisenthal 与 Tracy Alloway 围绕“市场与交易的未来”拉开帷幕。除了近年来席卷华尔街的“人工智能交易（AI Trade）”，金融市场当前最显著的演变莫过于可交易金融工具呈现指数级爆炸式增长。过去数十年间，投资者的工具箱主要局限于股票、债券与传统期权，随后期权逐渐走向奇异化（如备受关注的零日到期期权 0DTE）。然而如今，市场正在演变为“凡是有确定结果的事件皆可下注”——无论是纽约市的降雪量、特斯拉的季度交付数据，还是超级碗中场秀的持续时长，几乎所有事件都能被包装成合约进行交易。 随着预测市场平台（如 Kalshi）的兴起，行业面临的核心考验在于这类市场能否突破小众圈层，吸引真正的机构投资者与对冲基金入场。体育赛事下注天然拥有海量流动性，但对于宏观经济风险或商业运营风险相关的合约而言，流动性匮乏成为横亘在机构参与者面前的“先有鸡还是先有蛋”的结构性死结。没有足够的市场深度，机构无法用其对冲真实经济风险；而没有机构入场，深度又无从谈起。为此，老牌做市巨头萨斯奎汉纳国际集团（Susquehanna International Group，简称 SIG）选择躬身入局。本期对话的嘉宾正是 SIG 宏观交易与预测市场主管 Jeremy Maletz，一位极少在媒体公开露面、却站在推动预测市场机构化最前沿的交易员。 ```mermaid flowchart TD accTitle: 预测市场的流动性困境与做市商角色 accDescr: 描绘预测市场在没有机构做市商时面临的流动性死结，以及做市商介入后如何打通机构对冲需求。 subgraph Problem["先有鸡还是先有蛋的流动性困境"] A["浅层交易量与缺乏流动性"] --> B["机构无法进行规模化对冲"] B --> C["机构资金与对冲基金离场观望"] C --> A end subgraph Solution["做市商 SIG 的破局角色"] D["SIG 提供资产负债表与双边报价"] --> E["跨时间与规模撮合交易"] E --> F["仓储单边风险，提供机构级承载力"] F --> G["吸引企业与机构入场对冲"] end Problem -.->|"做市商介入打破循环"| Solution ``` --- ### `03:52` 预测市场的“引路人”与 SIG 的贝叶斯概率文化 Jeremy Maletz 介绍了 SIG 预测市场台（Prediction Markets Desk）的核心职责。作为一个横跨体育、宏观经济与政治等多个维度的主要流动性提供者，该交易台的首要任务是做市（Market Making）。但更深层次的使命在于扮演行业“牧羊犬（Shepherd）”的角色。作为第一家正式涉足预测市场的传统金融机构，SIG 不仅向生态系统注入做市资金，更致力于向其他华尔街机构示范如何安全合规地融入这一新型市场生态，打消大型机构的顾虑。 针对 Tracy 戏称许多人将 SIG 视为“金融界的 Salesforce”——名气极大却让人看不透具体在做什么——Jeremy 幽默回应称自己的妻子当天过生日，平日也总问他同样的问题，引来全场笑声。他阐明，SIG 的立足之本始终是期权做市业务，触角覆盖几乎所有股票期权及各门类资产。更关键的是，SIG 拥有源自联合创始人杰夫·亚斯（Jeff Yass）的独特交易文化：崇尚预测市场，坚持以“贝叶斯概率（Bayesian probabilities）”拆解世间万物，通过胜率与赔率来衡量一切不确定性。无论公司内部拓展出多少看似天马行空的随机业务分支，其核心始终围绕着用坚实的概率模型为新兴资产类别提供流动性。 - **业务定位**：SIG 预测市场交易台身兼做市商与机构领路人双重角色，为政治、宏观、体育等多类合约提供流动性。 - **底层文化**：以创始人 Jeff Yass 奠定的贝叶斯概率思维为导向，强调任何资产或事件均可拆解为离散概率模型。 - **业务基石**：以股票期权及跨资产做市（Market Making）作为立身之本，将做市技术迁移至新型预测合约。 --- ### `05:44` 为何纯点对点撮合走不通：跨越时间与规模的做市逻辑 面对“为何预测市场不能完全依靠点对点（P2P）交易、非要引入做市商”的疑问，Jeremy 指出，在理论上，如果一个市场汇集了无穷无尽、随时随地都渴望交易的人群，确实可能实现去中介化的 P2P 撮合。然而在现实世界中，交易者的需求存在天然的错配。做市商的核心价值，正是充当跨越“时间（Time）”与“规模（Size）”的过桥中介。 Jeremy 用主持人的现场互动举例生动拆解了这一机制：假如 Joe 此时此刻想要建立仓位，而 Tracy 却要等一个小时后才打算交易，若没有中介，两人的交易意愿便无法达成；做市商的作用就在于此时充当 Joe 的对手方接下合约，随后承担价格波动的等待成本，直到一小时后 Tracy 进场时再将头寸转移给她。除了时间维度的对齐，做市商还能将零散零售买家的微小订单聚合成对手盘，满足大体量资金的单笔大单进出需求，从而抹平供需两端的摩擦。 --- ### `06:40` 雄厚资本支撑下的风险仓储与万亿对冲疑问 谈及资产负债表与风险偏好时，Jeremy 强调了 SIG 作为头部自营做市商的非对称优势。市场上虽有许多小型独立团队或个人做市商，但往往因资本极度受限而必须严格追求每一笔交易的绝对风险中性（Risk Neutral）。相比之下，SIG 拥有几乎不受资本约束的资产负债表，这使其有能力报出巨大的合约头寸，承担起“引导市场启动（Bootstrap a market）”并吸纳机构级规模的重任。在风险管理上，SIG 并不教条式地坚守绝对风险中性，而是沿袭其在传统期权市场的作风——愿意承担单边风险敞口，为全市场进行“风险仓储（Warehousing risk）”。 这种风险仓储能力对于预测市场至关重要。例如在面临某些波及全社会的全局性事件时，市场上可能所有参与者都出于恐慌或对冲目的单向买入某一结果，若没有做市商敢于在另一侧承担敞口，市场将彻底瘫痪。敢于承担这一风险的前提，是团队对事件概率定价拥有极其深厚的信心。然而，主持人 Joe 也尖锐地指出了当下的现实鸿沟：尽管预测市场理论上可以成为实体企业的对冲工具（例如航空公司利用降雪量合约对冲航班取消损失），但放眼当下的预测平台，像“纽约降雪量 6 至 8 英寸”这类合约的成交量往往只有区区 15 万美元，这种水准的流动性对于动辄产生数千万美元实际经济损失的大型实体企业而言，无异于杯水车薪。 - **资本优势**：与受制于资金规模的中小型做市商不同，SIG 充裕的资本使其能够直接吞吐机构级巨额订单。 - **做市机制**：不机械追求瞬时的 Delta 中性或风险中性，主动扮演街面“风险仓储者”角色，吸收单边极端避险头寸。 - **现实鸿沟**：当前非体育类合约的流动性体量（如降雪合约仅约 15 万美元成交量）距离大型企业真实的宏观或经营对冲需求仍有巨大差距。 ### `10:00` 机构流动性的冷启动与预测市场的价格发现 当谈及预测市场何时能真正成为机构投资者的实用对冲工具时，杰里米·马莱茨（Jeremy Maletz）回顾了萨斯奎汉纳（Susquehanna International Group, SIG）的入局路径。在参与初期，SIG 的主要任务是为散户市场“冷启动”（bootstrapping）提供大额流动性；而眼下他们面临的全新挑战，则是如何为机构级别的流动性破局。面对外界普遍质疑许多合约盘口浅、成交量无法支撑机构对冲数千万美元风险的现状，马莱茨直言，这正是他公开做客播客的原因——向市场明确传递一个信号：SIG 随时准备为机构背书并承担这类大额风险。 马莱茨指出，预测市场最核心的价值并非当下的成交量，而是“信息本身”与“价格发现机制”（price discovery mechanism）。平台上聚集着一批极其敏锐的“超级预测者”（superforecasters），哪怕某份合约目前的累计成交量并不大，只要有一群聪明的大脑反复推敲博弈，就能迅速收敛出一个公允价格。在此基础上，SIG 会叠加自身的内部量化评估。一旦确认价格合理，哪怕该合约在交易所里只成交了很少的金额，SIG 也敢于直接为需要对冲监管变动或霍尔木兹海峡（Strait of Hormuz）地缘冲突风险的实体企业，提供数千万美元级别的风险承接能力。 ```mermaid flowchart TD accTitle: 预测市场机构对冲落地流程 accDescr: 展示从预测市场超级预测者价格发现，到做市商量化评估，再到以不同交易形式向机构提供流动性的链路。 subgraph Discovery["价格发现层"] A["超级预测者社群"] --> C["公允市场价格"] B["小额散户交易"] --> C end subgraph Internal["做市商内部评估"] C --> D["SIG 内部模型复核与风控"] end subgraph Execution["机构流动性交付"] D --> E["场内大宗交易 (Block Trade)"] D --> F["场外掉期互换 (Swap 合约)"] end subgraph Channels["触达终端机构"] E --> G["终端客户 (如航司/基金)"] F --> H["银行 / 经纪商 / 保险中介"] H --> G end ``` ### `12:15` 机构落地的阻碍与交易架构：场外互换、大宗交易与中介网络 在与机构投资者的实际沟通中，马莱茨发现最大的痛点首先是“认知偏差”。大型机构往往看一眼屏幕就下结论说“这里流动性不足，没法做大规模对冲”，而 SIG 的回答非常直接：“我们就是流动性本身（We will be the liquidity）”。另一个关键瓶颈在于合规与基础设施：预测市场对机构而言过于前沿，法律监管边界尚不清晰，风控与合规部门根本不知道该如何让这类头寸穿透内部防火墙。相比雷厉风行的散户，机构天生行动迟缓，因此 SIG 正在扮演“引路人”角色，设计多样化的落地通道来化解阻力。 针对交易机制，马莱茨透露了具体的操作模式： - **场内大宗交易（Block Trade）**：机构可直接接入预测市场平台，与做市商在交易所内撮合成交。 - **场外掉期合约（Swap）**：许多大体量的对冲头寸实际上不会直接在交易所前端盘面上打印出来。交易双方可以在场外以金融互换（Swap）的形式完成交易，仅向预测平台购买底层数据的授权许可（market data licensing）。这种模式已经在逐步搭建中。 针对具体对接链条（例如航空公司对冲降雪量风险），马莱茨表示，SIG 极具自知之明——他们精于定价与扛风险，但并不擅长去逐一敲门拜访每一家航空公司的客户关系。因此，SIG 倾向于借助传统金融中介网络，与投行、主经纪商（Prime Broker）及保险公司紧密合作。由这些长期服务实体企业的顾问机构负责前端揽客，再由 SIG 在后端承接所有风险敞口。 ### `16:35` 挑选做市领域：量化模型 vs 泰勒·斯威夫特婚期 面对平台上海量、甚至略显荒诞的预测议题，主持人提出疑问：SIG 是否承诺为所有合约都做市？例如“耶稣是否会重临”、“外星人是否会入侵”，甚至是“超级碗中场秀时长”这类充满不确定性的边缘合约，做市商该如何下注？ 马莱茨坦率承认，SIG 不会、也不可能为每一个合约都提供报价： - **术业有专攻**：对于流行文化或名人八卦（比如泰勒·斯威夫特何时结婚），SIG 并没有特别的情报或建模优势，庞大的超级预测者社群在这类事件上完全可以自给自足，平台并不需要大型做市商介入。 - **聚焦经济价值**：SIG 的核心价值在于将高额资本、强大的计算科技、量化模型以及资深宏观交易员的敏锐洞察相结合。因此，他们选择将精力倾注在真正具备“重大经济价值”与实体对冲需求的事件上。 马莱茨强调，预测市场的生态系统本身具备高度互补性：广泛的预测爱好者社群擅长挖掘五花八门的信息与文化事件，而像 SIG 这样的专业机构则在宏观与金融风险层面托底，二者各司其职。 ### `18:45` 市场构成的演化：从体育博彩基本盘向宏观经济对冲延伸 针对外界常常批评预测市场“本质上就是体育博彩、偶发的政治大选投机以及加密货币衍生品翻版”的观点，主持人提问，当下真实的资金流向中，是否已经能观察到“穿西装的体面人”（即传统金融机构）所关注的严肃事务正在抬头。 马莱茨对此给出了肯定的回答。他并不避讳当前的现实情况： - **体育仍是基本盘**：目前超过一半的交易量依然集中在体育赛事上。原因在于美国长期以来存在碎片化、按州割裂且监管繁琐的体育博彩历史，预测市场以更优越的交易形态整合了这一庞大存量需求，开局即拥有巨额流量是完全自然的。 - **溢出效应带动全盘**：庞大的体育交易群体极大提升了预测市场的整体认知度。正是借由这些高频交易带来的流动性与平台关注度，才进一步“拉动”（dragging up）了诸如经济指标、监管政策和宏观地缘等严肃品类的成长。市场正在经历从娱乐投机向实体宏观风险管理的深刻跃迁。 ### `20:00` 预测市场做市机制与传统金融工具的异同 杰里米·马莱茨（Jeremy Maletz）指出，在当前的预测市场生态中，除了体育博彩类别的自然增长外，非体育类的“真实世界事件市场”（Real-world event markets）虽然起步基数较低，但其增速实际上已经超过了体育类。萨斯奎汉纳（Susquehanna International Group, SIG）眼下的核心目标，正是推动套期保值（Hedging）用例的落地。尽管用于对冲风险的交易目前占整个市场的比重仍然不大，但在马莱茨看来，从长期的行业生态构建角度而言，让套期保值交易成长为市场中流砥柱是必不可少的关键环节。 随后主持人提问：如果为“美联储下一步是加息、降息还是按兵不动”这类事件合约做市，表面上看与传统金融市场中的利率衍生品极为相似，那么从底层的做市机制与运作逻辑来看，预测市场与传统期权交易到底存在怎样的差异？ 马莱茨解释称，做市逻辑并没有放之四海而皆准的固定模板，而是取决于标的性质： 1. **存在传统金融代理变量的宏观事件**：例如美联储的利率决议或是标普500指数（S&P 500）在当天收盘时的具体点位。在传统金融市场中，已经存在能够较好反映这些预期的现存工具。事件合约的作用，是将这些预期提炼并浓缩成一种更直接、纯粹且能精准表达交易者想法的合约形态，做市商可以将传统金融市场的定价直接映射转换到预测市场中。 2. **缺乏传统金融对冲工具的单发独立事件**：例如“霍尔木兹海峡（Strait of Hormuz）是否会在8月底前保持通航”或“凯尔·斯塔默（Keir Starmer）是否会下台失去英国首相职位”。这类高度特异的事件无法在传统市场中找到成熟的对冲标的，必须依赖深度的独立投研。 3. **基于预测市场内生信息的定价与德州扑克哲学**：马莱茨强调，预测市场自身蕴含着海量信息。SIG的核心做市哲学源自其在德州扑克培训中贯穿的思维方式——“理解他人掌握了什么信息，以及他人正在采取什么行动”（"understand what other people know and what other people are doing"）。做市团队会观察市场中超级预测者（Superforecasters）的仓位和下注水位，将其隐性智慧与机构内部的独立研究相结合，最终给出一个基准概率。对于某些完全崭新的未知领域，SIG甚至会从底层构建全新的量化模型（比如对算力供需的建模能力），这早已深深植根于其工程与交易的基因之中。 ```mermaid flowchart TD accTitle: 预测市场做市定价路径 accDescr: 展示萨斯奎汉纳根据不同事件类型采取的做市与定价机制流程图。 subgraph Input["事件类型分类"] A["传统宏观/金融事件<br/>(如美联储利率决议、标普500收盘)"] B["地缘政治/突发事件<br/>(如霍尔木兹海峡通航、政要去留)"] C["新型复杂科技/工业领域<br/>(如算力供需、新型算力基准)"] end subgraph Process["定价与风控机制"] D["映射传统金融衍生品代理标的"] E["结合市场超级预测者下注流 + 内部独立研究<br/>(德州扑克博弈思维)"] F["从零搭建底层物理/数学量化模型"] end subgraph Output["最终报价生成"] G["确定事件概率并挂单双边做市"] end A --> D --> G B --> E --> G C --> F --> G ``` --- ### `23:00` 内幕交易甄别与合规市场的风控底线 针对预测市场最常为人诟病的“内幕交易”（Insider Trading）问题，主持人提出了犀利的疑问：对于像 SIG 这样每天承接海量订单流的顶级做市商而言，通常可以依靠统计法则让噪音流相互抵消；但如果市场中混入了并非凭借推理预测、而是已经提前获悉真实结果的“绝对知情流”（Deeply informed flow），做市商是否能够察觉？这是否会颠覆机构的风险管理模型？ 马莱茨对此给出了清晰的界定，首先必须区分两种截然不同的预测市场环境： - **受监管的合规预测市场**：如 Kalshi、CME（芝商所）的事件合约平台等，这些平台均严格执行客户身份审查（KYC）制度； - **基于加密货币的去中心化金融（DeFi）平台**：如以太坊/Polygon 上的去中心化平台，这些平台完全缺乏 KYC 审核机制。 马莱茨指出，公众在媒体报道中所看到的大多数内幕交易丑闻，几乎全部集中在无需实名认证的加密去中心化平台上，而这正是 SIG 坚定选择仅在受监管平台参与做市的根本原因。在合规体系下，做市商享有更高的规则保护。 更深层次的防线来自于对交易对手动机的贝叶斯推断。在传统股票市场中，一个人买入苹果公司（Apple）的股票可能有成千上万种合理解释；但在预测市场上，标的意图极度单一而直白——如果某一天盘面上突然出现大量毫无由头、不计成本猛砸“委内瑞拉总统马杜罗（Nicolás Maduro）即将下台”的激进买盘，其动机异动的异常信号就会被即刻捕捉。马莱茨直言：“当发生内幕交易时，其动机不合理性要明显得多，随后事件如果发生，我们很容易做出可靠的贝叶斯更新，断定这大概率是内幕交易并向监管机构举报。” 此外，做市商对美国司法部（DOJ）近期开始全面介入调查加密预测市场违规行为感到欣慰。过去一些投机者天真地以为区块链是匿名的法外之地，但现实恰恰相反，链上所有流水和哈希记录都永久公开可查，“人们原本以为在加密世界能逍遥法外，现在终于意识到‘我迟早会被抓获’”，这种监管确定性正是遏制违规套利的终极刹车阀。 --- ### `26:15` 市场巨鲸操纵与做市标的选择机制 除了内幕交易，主持人还提出了另一大行业痛点：拥有巨额资金的“巨鲸”（Whales）是否可能通过下重注人为扭曲市场隐含概率，甚至试图反过来影响现实世界事件的最终走向？ 马莱茨对这种担忧的实际威胁程度表示怀疑。他指出，到目前为止，这种操纵企图并没有真正成为困扰 SIG 的现实问题。一方面，预测市场中并非只有单一巨鲸，类似 SIG 这样资金实力雄厚的深度流动性提供者同样在场，且极其擅长研判交易对手的意图动机；如果某个市场本身很难被现实干预，仅仅是有人试图在盘口砸钱造势，整个市场的其他理性交易者与做市商很快就会察觉偏差，并果断吃掉其订单以赚取套利空间。 不过，如果某个合约的底层结算结果极易受到现实层面的直接操纵，SIG 则会从源头上将其从做市名单中剔除。马莱茨特别提到了所谓的“提及市场”（Mention Markets）： - **提及市场的荒谬性**：现场主持人和马莱茨笑谈这类滑稽的投机品——“大家知道什么是提及市场吗？比如押注‘乔（Joe）、特雷西（Tracy）或者杰里米今天在播客里会不会说出 Dogecoin（狗狗币）这个词’”。如果有投机客砸下一大笔资金押注某个词会出现，随后私下塞钱给播客主持人让其念出该词，这类合约就彻底沦为了极度容易被定点操纵的赌局。 - **做市标的准入原则**：SIG 绝不涉足这类滑稽的提及市场，马莱茨建议普通交易者如果担心被操纵也应当远离此类投机。对于严肃的市场，若结算机制存在被现实绑架的风险，机构同样坚决不予参与。 关于 SIG 的做市范围是遵循固定的预设清单，还是类似“伦敦劳合社”（Lloyd's of London）那种接受定制化承保询价的模式，马莱茨表示两者兼具： - **核心基准合约**：SIG 签订了严格的做市义务协议，提供 24/7 全天候不间断的双边流动性挂单； - **临时特色事件**：当某些重大或高关注度突发事件涌现、且 SIG 认为自身具备定价优势并能发挥实质性市场作用时，团队会采取灵活的特设（ad-hoc）机制介入提供流动性。 --- ### `28:45` 数据抓取合规底线与机构声誉的权衡 访谈切入了一个颇具火药味的话题：主持人提及《纽约时报》（The New York Times）当周发表了一篇关于预测市场职业高手（Sharps）狂揽暴利的专访报道。报道中有一位独立交易员透露，自己曾去萨斯奎汉纳面试，但发现 SIG 内部存在严格的数据限制，公司禁止抓取某些特定网站的数据；这名独立交易员认为这种合规束缚削弱了自己的套利空间，因此最终拒绝了加盟 SIG 的机会。 主持人直接发问：“你们的数据限制到底具体指什么？” 马莱茨坦诚且严肃地阐明了顶级机构与散户交易者之间的本质鸿沟： - **品牌声誉与法律风险的绝对优先权**：SIG 拥有庞大的金融专营业务特许权（Franchise）和几十年来树立的行业声誉，必须全力保护这些核心资产。“我们绝不可能去违反任何网站的服务条款与使用协议（Terms and Conditions）”。 - **个人散户与持牌机构的双重标准**：一个散户或独立游资去钻空子写爬虫抓取禁止爬取的数据，甚至通过猜测未正式上线的隐蔽 URL（未发布网页的前置测试链接）来提前获取新闻，可能短期内侥幸不会惹上官司或招致诉讼。但对一家受全球严密监管的华尔街巨头而言，任何违背网站使用条款的数据抓取行为都是合规红线，绝无妥协余地。 - 报道背景：《纽约时报》当周深度报道预测市场高阶独立交易者，涉及求职者因数据合规门槛而放弃入职 SIG 的轶事。 - 核心分歧：独立交易者依赖灰色手段（如越权爬虫、探测未发布 URL）抢先下注；正规做市机构则将遵守网站使用条款视为不可逾越的底线。 ### `30:00` 机构合规的自我约束与算力（GPU）金融化的前景 在谈及量化机构与数据抓取的关系时，杰里米·马莱茨（Jeremy Maletz）坦言，尽管外部有许多人随意爬取网页数据，但萨斯奎汉纳（Susquehanna International Group, SIG）绝不会为了获取信息而违反任何网站的服务条款与使用协议。作为老牌顶级做市商，公司在声誉与合规风险上设立了极高的保守底线。加入这类大型机构固然意味着个人操作上会受到某些流程和制度的“束缚”，但机构级平台的雄厚资本与基础设施所带来的交易优势也是显而易见的。 随后，主持人将话题引向了前半场节目探讨的核心议题：当前备受瞩目的算力（GPU）市场是否具备高度金融化的基本雏形？马莱茨给出了坚定的肯定回答，并透露预测市场平台当下正在重点布局这一领域。在他看来，预测市场真正颠覆传统衍生品生态的核心变革力量，正是其无与伦比的“上市推新速度”（speed to market）。 为了说明传统金融在面对新兴风险时的迟钝，马莱茨分享了一个亲身经历的“起源故事”：在第一次中美贸易摩擦爆发期间，他的一位好友在一家乐器公司担任首席财务官（CFO），由于该公司的乐器全部依赖从中国进口，关税剧增的阴影让其时刻面临破产威胁。作为一名宏观交易员，马莱茨本能地想：“我是个做宏观的，这种风险我天天都在对冲，完全能帮他设计一个对冲工具。”然而，当他们尝试在传统交易所挂牌上市一款相关期货合约时，整个流程耗费了整整一年时间，节奏缓慢到对实际企业运营毫无用处。 相比之下，新型预测市场将新合约的创建与上线周期从数月甚至一年，压缩到了短短一天、甚至数小时之内。这种极速推新的机制，完美契合了算力租赁、DRAM 内存芯片价格以及海运运费等领域的风险管理需求——这些资产具有大宗商品的经济属性，却缺乏传统标准化的衍生品工具，预测市场恰好填补了这块巨大的空白。 ```mermaid flowchart TD accTitle: 对冲工具上市流程对比：传统交易所 vs 预测市场 accDescr: 对比传统期货产品上架（耗时数月至一年）与预测市场快速上线（一天以内）的路径差异。 subgraph Traditional["传统金融交易所 (如期货合约)"] direction TB A1["实体产业面临突发宏观风险<br/>(例如：乐器公司遭遇关税冲击)"] --> A2["设计定制化期货合约"] A2 --> A3["漫长的监管审批与上市申报流程<br/>(通常耗时一年以上)"] A3 --> A4["对冲工具落地过慢，错失避险窗口"] end subgraph Prediction["新型事件/预测市场 (如 Kalshi)"] direction TB B1["新兴非标风险暴露<br/>(如：GPU 算力波动、DRAM 现货价格)"] --> B2["快速构建事件结算规则"] B2 --> B3["轻量化极速推新机制<br/>(耗时一天甚至数小时内)"] B3 --> B4["敏捷完成建仓与宏观风险对冲"] end ``` - **合规红线**：大型做市机构严格恪守网页使用条款（Terms and Conditions），绝不采取非法爬虫或违规手段收集另类数据。 - **推新周期差距**：传统交易所上市期货合约往往需要耗费约一年时间，而现代预测市场推出对应算力或突发事件的合约可在一天之内完成。 - **潜在金融化标的**：GPU 算力定价、DRAM 内存价格、国际海运运费等具备大宗商品特征但缺乏标准化期货的产品，正成为预测市场的核心拓展前沿。 --- ### `32:30` 宏观交易与预测市场的结合点：从2016年大选套期保值谈起 主持人乔·威森塔尔（Joe Weisenthal）打趣地指出，马莱茨在萨斯奎汉纳同时兼任“宏观交易主管”与“预测市场主管”，这两个职务组合在华尔街极为罕见，并询问这两个市场之间究竟存在怎样的业务协同效应。马莱茨笑着坦言，这并非是高层一开始在画板上凭空规划出来的产物，其真正的起点全部源于“美国大选交易”。 早期，马莱茨在掌管宏观交易时专精于大选事件的定价，当时美国本土尚未开放合规的政治预测市场，而萨斯奎汉纳通过其欧洲业务实体参与了英国博彩交易平台 Betfair 上的政治合约做市，这让他切身体会到了直接事件合约的巨大对冲价值。最具说服力的案例发生在2016年美国总统大选：当时机构客户面临极大的政治不确定性，各大投行纷纷推出由数十只股票打包而成的“一揽子概念股代理组合”（Basket Trades），华尔街普遍形成的压倒性共识是——“如果唐纳德·特朗普胜选，美股市场将暴跌 5% 到 7%”。 然而大选夜的实际走势戏剧性地打破了这一模型：美股仅仅暴跌了大约五分钟，次日收盘时整个大盘却全线大幅飙升。由于市场对选举结果的实际反应与金融代理组合的预期完全脱节，那些耗费重金买入一揽子股票进行对冲的机构全线亏损，所谓的“完美对冲”彻底失效。但与此相对，如果在预测市场上直接针对大选胜率进行下注或对冲，其兑付结算则严格遵循选票结果，不存在任何资产间相关性断裂的基差风险（Basis Risk）。正是目睹了这种传统“概念篮子交易”（比如如今依然常见的 DRAM 赢家与输家百万元一揽子股票）与事件现实脱节的顽疾，萨斯奎汉纳坚信预测市场具备不可替代的结构性价值，马莱茨的宏观部门也由此顺理成章地将预测市场纳入核心版图。 访谈在轻松的氛围中结束，主持人特雷西·阿洛韦（Tracy Alloway）与乔·威森塔尔感谢了杰里米·马莱茨在纽约 City Winery 现场的深度分享，并调侃如果未来有人在预测市场上开出“外星人入侵”合约，他们一定会再次邀请萨斯奎汉纳探讨该合约的做市与定价逻辑。 - **2016 年华尔街大选共识破灭**：华尔街曾押注特朗普胜选会导致美股崩跌 5% 至 7%，但恐慌抛售仅持续了约 5 分钟，次日市场全面收涨，暴露出股票概念篮子对冲政治风险的严重基差缺陷。 - **直观工具价值**：与基于复杂相关性假设的股票篮子不同，预测市场的二元结算机制能够提供无杂音、点对点的纯粹宏观风险对冲。 - **业务演变轨迹**：从欧洲 Betfair 交易经验，到将宏观对冲融入美国本土合规预测平台，SIG 跨界布局预测市场源于实战风险管理的自然演进，而非人为预设的战略拼接。 ## Notable quotes > `00:23` 凡是有确定结果的事件皆可下注 — Joe Weisenthal > `12:15` 我们就是流动性本身 — Jeremy Maletz > `20:00` understand what other people know and what other people are doing — Jeremy Maletz > `23:00` 当发生内幕交易时，其动机不合理性要明显得多，随后事件如果发生，我们很容易做出可靠的贝叶斯更新，断定这大概率是内幕交易并向监管机构举报。 — Jeremy Maletz > `23:00` 人们原本以为在加密世界能逍遥法外，现在终于意识到‘我迟早会被抓获’ — Jeremy Maletz > `26:15` 大家知道什么是提及市场吗？比如押注‘乔、特雷西或者杰里米今天在播客里会不会说出 Dogecoin 这个词’ — Jeremy Maletz > `28:45` 我们绝不可能去违反任何网站的服务条款与使用协议 — Jeremy Maletz > `30:00` 我是个做宏观的，这种风险我天天都在对冲，完全能帮他设计一个对冲工具。 — Jeremy Maletz > `32:30` 如果唐纳德·特朗普胜选，美股市场将暴跌 5% 到 7% — Jeremy Maletz ## Mentioned Susquehanna International Group (SIG), Jeremy Maletz, Joe Weisenthal, Tracy Alloway, Odd Lots, City Winery, Kalshi, Jeff Yass, Salesforce, CME (芝商所), Betfair, The New York Times, Department of Justice (DOJ), Donald Trump, Keir Starmer, Nicolás Maduro, Taylor Swift, S&P 500, Apple, Dogecoin ## Video description <p>Prediction markets that enable you to bet on pretty much everything are everywhere nowadays. But there's still a big question over whether they can expand to include larger institutional investors like hedge funds. Part of the problem is that a lot of prediction market contracts are illiquid and trading volumes can sometimes be shallow. That's where trading firm Susquehanna International Group comes in. In this episode, recorded live at New York's City Winery, we talk to Jeremy Maletz, Susquehanna's head of macro trading and prediction markets, about the firm's market-making business with Kalshi. We talk about how big investors could use prediction markets, what Susquehanna is seeing in terms of flows, how a market-maker hedges risk on these contracts, and how it makes money from them.</p><p>See <a href="https://omnystudio.com/listener">omnystudio.com/listener</a> for privacy information.</p> [Source](https://podtrac.com/pts/redirect.mp3/tracking.swap.fm/track/UVBrz8bN8aM2Xe47PEPu/traffic.omny.fm/d/clips/e73c998e-6e60-432f-8610-ae210140c5b1/8a94442e-5a74-4fa2-8b8d-ae27003a8d6b/ba11bcd6-2009-4d79-98f8-b46001259850/audio.mp3?in_playlist=982f5071-765c-403d-969d-ae27003a8d83)

[0:23] Joe Weisenthal: 凡是有确定结果的事件皆可下注 [12:15] Jeremy Maletz: 我们就是流动性本身 [20:00] Jeremy Maletz: understand what other people know and what other people are doing [23:00] Jeremy Maletz: 当发生内幕交易时，其动机不合理性要明显得多，随后事件如果发生，我们很容易做出可靠的贝叶斯更新，断定这大概率是内幕交易并向监管机构举报。 [23:00] Jeremy Maletz: 人们原本以为在加密世界能逍遥法外，现在终于意识到‘我迟早会被抓获’ [26:15] Jeremy Maletz: 大家知道什么是提及市场吗？比如押注‘乔、特雷西或者杰里米今天在播客里会不会说出 Dogecoin 这个词’ [28:45] Jeremy Maletz: 我们绝不可能去违反任何网站的服务条款与使用协议 [30:00] Jeremy Maletz: 我是个做宏观的，这种风险我天天都在对冲，完全能帮他设计一个对冲工具。 [32:30] Jeremy Maletz: 如果唐纳德·特朗普胜选，美股市场将暴跌 5% 到 7%

[4:47] Jeremy Maletz: Joe, you might want to trade now and Tracy might want to trade in an hour, right? But that doesn't help you if she's not there now. So we're basically saying, 'Hey, we're going to be there when you want to trade, and then we're going to wait. And when Tracy wants to trade, we'll take the other side of it.' [20:00] Jeremy Maletz: There's a million reasons that someone can buy Apple stock, but like there's not that many reasons that someone can buy like 'Is Maduro going to be out?' [25:00] Jeremy Maletz: Someone could put a bunch of money on something and then, you know, on this podcast perhaps say... 'Will Joe, Tracy, or Jeremy say the word like Dogecoin?'... We're not going to participate in those.

[0:00] Whatever your goal, trade show giveaways, client gifts, or team gear, 4imprint has the promo products to match. With thousands of options, from apparel and drinkware to tech and totes, it's easy to find the right fit for your brand and budget, with standout choices at every price point. And with their 360-degree guarantee, you can be 4imprint certain your order will show up just right, right on time. Explore more at 4imprint.com. 4imprint. 4certain. So there's a lot of noise about AI. But time's too tight for more promises. So let's talk about results. At IBM, we work with our employees to integrate technology right into the systems they need. Now, a global workforce of 300,000 can use AI to fill their HR questions, resolving 94% of common questions. Not noise. Proof of how we can help companies get smarter by putting AI where it actually pays off, deep in the work that moves the business. Let's create smarter business. IBM. Wise is the smart way to manage the currencies you need around the globe. When you send money abroad using your bank, you could get hit with hidden fees and exchange rate markups. There's a better way. Try Wise. Wise uses the exchange rate you usually find on Google with no unwelcome surprises Plus most transfers happen in under 20 seconds which means your money arrives in less time than you been listening to me It simple and free to sign up when you download the WISE app. Be smart, get wise. T's and C's apply. Bloomberg Audio Studios. Podcasts, radio, news. Hello and welcome to another episode of the Odd Lots podcast. I'm Jill Weisenthal. And I'm Tracy Alloway. So Tracy, we're still rolling out shows from our live show on May 28th in New York City at City Winery. As we discussed in our last episode from this show, it had a sort of future of markets, future of trading theme to the night's conversation. Right. Right. And if you're talking about future of markets and trading, we have to talk prediction markets, right? Yeah, that's right. So in addition to the fact that there is the quote, AI trade, unquote, the other big thing going on in markets is the sheer explosion of instruments with which people can trade right So it like you used to have stocks and bonds and options and then the options started getting more exotic like zero options which you love and so forth But now it like if you could think of something that would resolve in some way whether it is snowfall in New York City Tesla deliveries or how long a halftime show is at the Super Bowl there probably is a way to bet on it Right. And so one of the big questions is whether or not these markets are going to take off from an institutional perspective, whether or not you're going to see more professional investors get into the business of betting on not just snowfall in New York, but maybe halftime shows and that sort of thing. But that said, we do have some institutional participation in the market already because you have market makers that are starting to come in to try to make these markets more liquid. Yeah but you really said the key thing here which is we know there like a ton of liquidity for like the sports betting etc Like that not the problem that needs to be solved It these other things where in theory these might be useful instruments for hedging or some sort of economic risk Maybe not the Taylor Swift one or the Halftime Show ones but some of these other ones But like in theory some of these contracts could be useful for hedging And so the question is yeah but will anyone use them And so on this discussion we really had the pleasure of someone who has had very little media in general and who is right in the heart of trying to essentially solve this chicken and egg problem. Listen to our conversation with Jeremy Mallitz, the head of prediction markets at Susquehanna International Group. All right. Why don't you tell us, just let's start really simply, what does the prediction markets desk at Susquehanna do? So essentially, the main function that we do is market making. So we're the main liquidity provider, one of the main liquidity providers on quite a lot of platforms. And that means we're providing liquidity in everything from sports to economics to politics. but I'd say also a big part of what we do is we were the first institution that got involved in prediction markets in the first place. So we're really trying to have a role of kind of being a shepherd for other institutions as they get involved. So I'd say it's a two-pronged goal really of providing the liquidity for the ecosystem to help it to grow and then bringing others into the ecosystem, which I think has been one of possibly even the more important role of what we've done to this point in time. Wait, can I ask an even simpler question, which is I kind of feel like Susquehanna is like the sales force of the finance world where I have this like [5:00] big idea of what you do, but also not really. What does Susquehanna do? Well, my wife asked me the same thing. It's her birthday, by the way, so happy birthday to her. It's incredible that you're here. I know. I owe her big time. So we're primarily a market-making firm, and our bread and butter has always been options. So we make markets in pretty much every option, equity options. But obviously, we trade a lot of instruments across a lot of things. And we have a culture overall that's very much looking for new opportunities and thinking about things and probabilities. And it all starts from Jeff Yoss, our founder. He loves things like prediction markets. So we have a lot of random businesses at Susquehanna all over the place. You wouldn't believe the number of things that we do randomly. And it all comes back to the same culture we have of thinking in Bayesian probabilities and trying to think of everything in terms of break it down into what the odds are. But the bread and butter remains market making. And we try to use that to kind of make a market or bring a market to any asset class we looking at Let talk about okay there some contract out there who is gonna win you know the texas primary like let just say why you know and so there an exchange and there an instrument and it's either gonna uh end at 100 or zero etc why does this market need market makers why can't it be entirely peer-to-peer such that if all of us in the room just wanted to trade we make a price amongst each other on an exchange. Why is a market maker an important part of the infrastructure for this to work? So if you have a market where you have an insane amount of people that are all trying to trade all the time, you might be able to make it work without a market maker. But what a market maker is really doing is it's helping to bridge the gap between the different people who are trying to trade. So Joe, you might want to trade now and Tracy might want to trade in an hour, right? But that doesn't help you if she's not there now. So we're basically saying, hey, we're going to be there when you want to trade, and then we're going to wait. And when Tracy wants to trade, we'll take the other side of it. So we're basically a function that matches the buyers to the sellers across time and also across size So how big does your balance sheet have to be to be dedicated to this particular business how sizable are you with an entity like CalShea So you know it really depends on the way that you do it There a lot of small independent groups or just individual people who are able to be out there and they make markets and their goal is to try to balance it and they have to think a lot about capital. We are very fortunate that capital isn't generally a constraint for us, right? So that's one of the things that it's an advantage for us, but it's also something we can bring to the market. We can provide much, much larger size on things. It's why we're well suited to bootstrap a market and to do institutional type size because we're not constrained by capital in that way. And do you generally try to stay risk neutral? No. I think that is pretty deeply in our culture that obviously all things equal, we would like to be risk neutral, but we're willing to put ourselves out there and wear a position. And in general, on our options trading, we tend to warehouse a lot of risk for the street. Sometimes there's just something where everyone needs to hedge a risk in one direction and someone needs to be on the other side of that. And that's an important function in prediction markets, especially when you think about use cases such as hedging where you got some global risk to the world Everybody needs to hedge that risk on one side Someone needs to be on the other side And we willing to hold ourselves out there to do that And there a lot of pieces that go into being able to do that Obviously when you not neutral on a risk basis you have to be more confident that you're right. But that's a big part of what our team has striven to build. So this sort of leads into the next question. And it's maybe the multi-billion dollar question of prediction markets, which is, okay, we know that there is a huge amount of the prediction markets business, which is just sports betting under a slightly different form. But you mentioned hedging, and this gets to the core question, which is, in theory, there are a lot of instruments on these prediction market platforms that could be useful hedging instruments for corporations, say, a market on snowfall in New York City, which might affect an airline or something like that. Maybe they want to hedge that. When I look at the platforms currently, I see a price for like snowfall, but I don't see anywhere near the volume level that say would really justify like an airline could, you know, you see $150,000 in volume on whether there'll be between six and eight inches of snow. That's obviously nowhere near deep enough for a serious economic impact. [10:00] actor to participate in. So where are we on that in terms of the promise of actually useful instruments for hedging? Right. So when we first got involved in prediction markets, our real role was to bootstrap the liquidity. There were no institutions yet, and it was largely going to be a retail product. We were bootstrapping for large volume retail liquidity. Now our next challenge is we want to bootstrap institutional liquidity. So yes, you might look at a market that doesn't seem like it has enough volume for an institution to hedge tens of millions of dollars of risk. But that's part of the reason that I'm out here doing a podcast today is we're trying to make sure that people start to understand this is viable and we're putting ourselves out there that we will be willing to put that kind of risk. And we can put out that kind of risk on a contract where far less volume is traded. And that's because what the prediction markets really provide is information. It's a price discovery mechanism. So you have this phenomenal community of super forecasters that exist on a prediction market. And it doesn't take as much volume as you would think to get to a fair price And that allows us to say hey okay we got a reasonably fair price on this prediction market Maybe it only traded But we know that there been a lot of smart people that have looked at this We can do our own internal vetting at the same time also. And now we're comfortable going out there and saying we're confident enough in this price because of the price discovery mechanism that we'll make tens of millions of dollars of risk to a company that needs to hedge its risk of some regulation or straight-up Hormuz or whatever it is that's happening in the world. So, you know, someone needs to go out and do that. And I think we're kind of uniquely positioned because of our culture of saying, yes, we are willing to take that risk. And yes, we want prediction markets to grow. What have your conversations been like with institutional players so far? Like, what do they say is their main either constraint or, I guess, reluctance to get on some of these markets? So the first piece is the exact thing that you just brought up, awareness. They look at the markets and they might say, well, I don't see how we could actually hedge some of these things because there's not enough volume and liquidity, to which our response is we will be the liquidity. Then the other question is, OK, well, we sort of need our compliance to get comfortable with this. This stuff is so new. What's the legal landscape? How do we get this stuff even over our firewall I think institutions were always going to be the slower moving player relative to retail And so that why we really want to sort of hold some of their hands to go through this and figure out a lot of different ways that they can use prediction markets. Some of it might be an institution connects to a prediction market platform, does a block trade on an exchange with us. It could be that a trade goes up off an exchange as a swap, but it's licensing the exchange market data. And we're trying to create as many different setups as possible so that people can do these trades. And we're going to basically be the facilitator to say, okay, you need a hedge. We're going to figure out how to let that happen. So this is really important. So if I look at a contract and I see a number there, like $150,000, it's possible that there was a trade, an off-platform trade of much bigger size that was not printed on that, but that was more by swap form. And is that currently happening? It is possible. I'm not going to go and say that that's a big thing that's happening yet. We're trying to build out the infrastructure to have as many options as possible because we want institutions to start moving. And the more they do it, the more they're going to... If an institution, again let go to the airline hedging snowfall Would they have a relationship with you Would they go through a prime broker who then has a relationship with you What is the actual chain of phone calls or whatever that happens? It can be both, but ultimately, when I think of what we're really good at and what we're not really good at, we're not necessarily the best in the know-every-single airline customer business. We want to work with a lot of those other intermediaries. That could be a broker, it could be a bank, it could be an insurance company, the people who have those relationships and are in the business of constantly advising, this is what you should do. We understand we want to work with them and they could have a very important role in helping to be a part of the infrastructure that gets it from the customer with the risk to us being the ones that have the other side of this. Thank you. [15:00] You can customize it all, your logo, your message, your look, and many items come with no setup charge to help you save. And if you're really watching the bottom line, you'll find standout choices at every price point so you can make a real impact while staying on budget. Plus, you'll get expert help, fast turnaround times, and their 360-degree guarantee, so you can be 4imprint certain your order will arrive on time and look exactly right. Whatever your goal, 4imprint makes it easy to find your perfect promo match. Explore the possibilities today at 4imprint.com. 4imprint. 4certain. Support for this show comes from public.com. If you're actively involved in your portfolio, you probably catch yourself repeating the same actions. Buying the dip. Manually sweeping idle cash. Putting on a hedge. On public, you can now create AI agents that handle all these tasks on your behalf. Just describe what you want to do in plain English. Like, if the VIX hits 25, buy a put option on the S&P 500. or if my cash balance goes above $20,000, move the excess into my direct index. You approve the workflow and your agent handles the rest. Monitoring the market, watching for your conditions and executing your strategies exactly as defined An investing platform driven by your intent not just your clicks You can also get full read and write access to your account via the public API Go to public.com slash market and fund your account in five minutes or less. That's public.com slash market. Paid for by Public Investing. Brokered services by Open to the Public Investing, Inc., member FINRA and SIPC. Advisory services by Public Advisors, LLC, SEC Registered Advisor. Complete disclosures available at public.com slash disclosures. So there's a lot of noise about AI, but time's too tight for more promises. So let's talk about results. At IBM, we work with our employees to integrate technology right into the systems they need. Now, a global workforce of 300,000 can use AI to fill their HR questions, resolving 94% of common questions. not noise proof of how we can help companies get smarter by putting ai where it actually pays off deep in the work that moves the business let's create smarter business ibm joe's chosen a pretty like reasonable contract i would say um for his example there but like are you committed to making markets in all the contracts available on these platforms because i thinking about you know like the return of Jesus Christ or will aliens invade How would you even go about like the length of a halftime show yeah right um yeah so we don make everything um and and honestly we're not the best to make absolutely everything some of you know some of the things that you might say that something perhaps in pop culture we might we're probably not going to be that good at it relative to others right there are certain things are you trying to build that capacity up. Well, we'll see. But there is a community of a lot of people that can make a lot of these types of markets. And what we want to do is we want to scale our capacity with our technology, our quantitative models, our very trader-driven insights that we have in our capital. That's where we're going to add the most value to an exchange. If something just requires a bunch of people to dig into it, the universe of all the different super forecasters, we're probably not as necessary. So you probably don't need us to know something about when Taylor Swift is going to get married. But the fact is that the things that have the most economic value tend to be the things that need us the most. And so that's really where we try to play. And I think it great that the ecosystem has these complementary forces of the community of all of the super forecasters and then the people like us who both kind of have different skill sets Do you see right now so again, the people say, yeah, these are just, it's sports betting, et cetera. And then maybe every two years or every four years, there's some election activity. Maybe there's some trades that are sort of like, look like crypto derivatives, et cetera. When you actually look at today's volume of activity, do you see a meaningful shift towards what polite people who wear suits would say like real things? Is that happening? It is. Yeah. And so when you look at the volumes, obviously there's a real percentage of the volume, more than half the volume certainly is sports. Yeah. And there's a reason for that, right? Sports has been a big market in the United States for a long time. And it's happened in kind of a fractured way across a lot of different states. And there's certain states where you can't do it and it's a different regulatory system everywhere. But there was a big market that existed. And then prediction markets basically came around and they provided a better way for sports to trade. So it was natural that there was going to be this massive base from the start. But what we're seeing is it's dragging up all of the other things. So it's brought awareness to prediction markets. [20:00] And the more we see that, the more we see everything else growing. And by the way, a lot of that other stuff, the real market, they're growing at a faster rate than sports just from a lower starting point. And our goal now is really to boost that with the hedging cases, which that's growing too, but it's not a huge part of the market as of now. And we understand it needs to be. From an ecosystem perspective, it's essential that those hedging trades become a much larger part of what the market is. So if you're a market maker for an event contract, that's something like the Fed is going to raise rates or hold or lower or whatever. That's pretty similar to something that you might see in traditional markets. But how different can you like explain it to us from a mechanical perspective, how different the market making process would be for a prediction market contract versus like a traditional option? Right. So honestly, there's not a one size fits all because there's so many different things out there. So if you're thinking about kind of what's the Fed going to do next, there are instruments that capture that reasonably well And same thing as you know what the price of the S 500 going to be at the end of today for example There are other instruments out there that capture that reasonably well And really this is just distilling it into a way that might even better capture the idea someone has. There's quirks to the other types of things that you might have that are the proxies. So we can translate what's happening in the traditional financial markets to prediction markets. For other products, that is sort of, is this random one-off event going to happen in the world? Is the Strait of Hormuz going to be open by the end of August? Is Keir Starmer going to be out as prime minister? That stuff requires a lot of independent research. But the good news is that there's so much information contained within the prediction market itself. So one of the core market-making principles we have is, it really honestly comes from kind of a thing we teach with poker of, understand what other people know and what other people are doing. We can learn from the markets, seeing where they are. This is what all these other super forecasters know. We can combine that with our internal research and then get to a number. So that's, you know, some things kind of have to be, we draw it up and figure it out like that. Some stuff is we can use the information that in other markets And some stuff is we going to build a brand new model to figure this thing out because that kind of in our DNA to say like all right here a random thing but we're going to have to figure out compute, right? Let's figure out how to model compute and we'll build it. So one of the concerns or one of the things that people talk about with prediction markets is the possibility of insider trading. And, you know, like obviously, like when I think of like a market making firm or some of these firms that do a lot of flow and volume, you just sort of assume there's a lot of noise and it all sort of washes out, right? But how does it change you? Like you have to think like if there are participants in the market that are like not just forecasters here, like not just people who are good at predicting things, but deeply informed flow where they maybe just know the answer already. Are you able to spot that? Are you able to sense that? And does that change how you think about risk management within a given market? Right. So I guess the insider trading is definitely something that comes up. and there's certainly been plenty of articles that are written about it. I think there's a couple things to point out there. The first one is that there actually really two types of prediction markets that it not always clear to people what the distinction is But there the regulated prediction markets such as Calci such as Rothera Exchange that just launched, such as CME has one and Polymarket has a regulated exchange. And then there's the crypto-based decentralized finance platforms. And they don't have KYC, right? And they're crypto-based, their DeFi. So I think most of what you've seen, the vast, vast majority where people find something that's insider trading based is on the DeFi platforms because there is KYC on the regulated space. And that's why we are participating in the regulated space. So that's the first important thing is if you're in the regulated space, you're much more protected. The other thing is insider trading is something that does exist everywhere. And there's a mechanism that works to suss it out. And that's reporting from the people that are in the market. Generally, one of the things that we're always paying attention to is what are the incentives of the person on the other side of the trade? And when someone's insider trading, it's a lot more obvious that, okay, well, what are the incentives? They don't make sense. The thing then happens and, okay, we can make a pretty decent Bayesian update that this was probably insider trading and report it. And it's actually easier in prediction markets because this. [25:00] is more obvious. There's a million reasons that someone can buy Apple stock, but like there's not that many reasons that someone can buy like is Maduro going to be out. If you see a lot of people slamming that, okay. Right, so it's actually easier to spot this stuff. And by the way, like one thing we're extremely happy about is the DOJ is now even going after the, you know, the crypto based platforms where, you know, people kind of, I think they kind of knew people probably thought they were safe. You're not safe in crypto and actually everything's on the blockchain, right? So you'll get caught. So there is an infrastructure, especially in the regulated space, but also now in the unregulated space for catching people that do it. And I think that's why you don't see more. That's what stops insider trading. People don't want to get caught. This was new. People probably thought like, oh, I can do it in crypto and we'll get caught. Now they're like, oh, I am going to get caught. OK, aside from insider trading, one of the other concerns or criticisms of prediction markets is the idea of you could have whales basically in the market, like people with a lot of money who are willing to spend it to perhaps like influence a particular probability with the hope of, I guess, influencing the ultimate outcome. You must have pretty good visibility into order flow Like how realistic is that concern from your perspective Honestly I don think it something that really come up for us to this point in time Like there are concerns we have around certain things in prediction markets but realistically like there's a lot of people in prediction markets with deep pockets, certainly us. And we're pretty good at, as I said, understanding the incentive of who's involved. And if someone's going out there and trying to move a market to like influence an outcome, you know, if it's, I guess it depends on what the outcome is. If it's an outcome that's easy to influence, we try to avoid those markets in the first place. So we don't trade like mention markets, right? Someone could put a bunch of money on something and then, you know, on this podcast perhaps and then say, Joe, here's a little... Does everyone know what a mention market is? Right. Does everyone know like a mention market is like, will Joe, Tracy, or Jeremy say the word like Dogecoin? Exactly, exactly. And someone bets like, it's crazy. Right. I mean, whatever. So we're not going to participate in those. if you're worried about something like that, you probably shouldn't trade those either. They're more on the side of just like, oh, if people want to punt around in them, fine, we're not in those. But for more serious markets, I don't have that much concern. If there's someone trying to do that if it not manipulable particularly it something that real and someone just trying to move the market the marketplace is probably going to figure it out and take the other side of it And if they can actually influence the outcome in a meaningful way then you probably need to think about what the settlement mechanism is And those are the types of markets we try not to get involved in. But I don't think you see a whole lot of that in the market. So as you mentioned, you don't make a market in every market and some things are more logical. do you have when you say okay you're market maker for calci do you have a list of like this is what we do and don't or is it like like i'm sort of in my mind the way i think about your role a little bit like a lloyds of london in the sense that it's like okay so someone has some risk and then they call you up and they like can you make us a price for this so is it a set list or is it like you take it as you see it. Yes. Okay. We can make a market. No, we, this is not something we're going to participate in on a sort of like per market base. Right. So it's a little of both. Um, there are some markets that we do all the time. We sign up for obligations. We're going to be out there all the time, 24 seven. And then there's some things that we kind of do on a more ad hoc basis. Like you know you it comes up it important We think we have can have a meaningful role in the market and we just figure out how to get involved in it So a little of both I don know if you read it but there was a piece in I think it was the New York Times this week about sharps in the prediction market. He definitely read it. Yeah. People who are making a lot of money. Oh, yeah, that's right. Okay, so you better have read it. I have read it. Okay, but there was someone quoted in there who was like an independent trader in prediction markets, And they said they interviewed at Susquehanna, but the firm said that Susquehanna is not allowed to scrape certain data or it has certain data restrictions. And I guess that made the job unappealing to this guy because he doesn't seem to have taken it. But what are your data restrictions exactly? So we have a large franchise and a large reputation that we need to take care of. So we're not going to violate the terms and conditions of a website, right? If you're a random person who's doing it, you probably can do that and you're probably not going to get in trouble more likely than not, although that's certainly not legal advice. This is like people putting in URLs that haven't gone live yet, right? And they have some feel and they test them out. It's not even necessarily – I mean that's a thing that can happen. It could just be like you have a website that has terms and conditions that says you're not permitted to scrape. [30:00] data, right? And something like, you know, plenty of people scrape the data, right? And we're not going to scrape that data. Like we're not going to violate the terms and conditions of a website. So we hold ourselves to a more conservative standard when it comes to those types of things. So I think that, I think that person was probably just referring to, hey, you get restrained in some ways if you want to work for us. Institutional, the usual like institutional things. Like we're going to be very careful when it comes to things around our reputation. But, you know, but I think that there's a lot of advantages to working here as well. I know this isn't a prediction markets question per se, but do GPU markets that we talked about on the first half of the show, do they have the sort of contours of what to you look like could be something a very actively financialized market? Absolutely. I mean, it's something that the prediction markets are looking And really, when I think about what's made prediction markets so different, like what actually changed, it's the speed to market. Say more about this. So, you know, it's funny, when we first started wanting to get involved with prediction markets, I have like a little bit of an origin story where I have a friend who is a CFO of a musical instrument company During the first China trade war he was worried that their company might go out of business because they imported instruments from China And I was like well I a macro trader I could hedge this It what I do all the time We actually tried to list a product and it took about a year and it was just too slow. And then, you know. Where did you try to list it? We tried to list it with my ex. And, you know, just that was the process for listing a future. And now prediction markets came around, and that process went to a day or even inside of a day. So I think really that's the most – that's the real valuable thing that happened with prediction markets. And I think that with compute, you see this ability of prediction markets to potentially move very quickly. So they can launch a product. There are products on prediction markets that have compute. And these other things that kind of look like compute, you might say DRAM prices. People worry about that all the time, right? The shipping costs, these things that are kind of like a commodity, but they don't really have a commodity. Prediction markets can be speed to market, and the system works in that way. So I absolutely think it's something we're thinking about, and I think they can totally have a role in that ecosystem. So just going back to the very beginning of this conversation you head of macro trading at Susquehanna Sorry I can say head of macro without cracking up You head of macro trading but also prediction markets at Susquehanna which is like kind of an unusual combined role. Like, what is the idea there? Is there some synergy between those two markets that you're trying to capture? You know, honestly, it all just comes down to the election. You know, as head of macro, I had sort of a niche for trading the election, and I kind of got exposed to prediction markets because they were trading in Europe on Betfair and we had a European entity and we were involved with it there. And I saw the value of it and I think 2016 election was like a place that it was really valuable because people had these massive risks they wanted to hedge. People would do it with these proxy products. Banks would put out baskets of like put together all of these names. and the broad market consensus in 2016 was that the market was going to be down five to seven percent if trump won and it turned out the market was like five minutes it was for like five minutes but yeah the next by the end of the next day it was it was up right and so the you know the hedge didn't work but we saw like the prediction market worked like if you if you actually just hedge this in a prediction market it would work really well and so honestly there not it doesn seem like there that much synergy It more just it evolved out of that of saying like hey we involved in this space and we see how valuable prediction markets can be and we really want to make it happen And so I think that it's really just kind of an evolution as opposed to this is how we would draw it up from scratch. Every once in a while I get one of those like basket trades where it's like trade this basket, we can do a hundred million dollars. Right. This, you know, DRAM winners, DRAM losers the basket. Anyway, Jeremy Mallitz, head of prediction markets at Susquehanna. Thank you so much for coming out. Thank you so much for having me. That was our conversation with Jeremy Mallitz of Susquehanna, recorded live at our New York show. I'm Tracy Alloway. You can follow me at Tracy Alloway. And I'm Joe Weisenthal. You can follow me at The Stalwart. Follow our producers, Carmen Rodriguez at Carmen Armin, Dash O'Bennett at Dashbot, Kale Brooks at Kale Brooks, and Kevin Lozano at Kevin Lloyd Lozano. And for more OddLots content, go to Bloomberg.com slash OddLots. We have a daily newsletter and all of our episodes. And you can chat about all these topics 24-7 in our Discord, Discord.gb. [35:00] And if you enjoy Odd Lots, if you like it when we do these live shows and ask Susquehanna if they can make markets in alien invasion contracts, then please leave us a positive review on your favorite podcast platform. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes absolutely ad free. All you need to do is find the Bloomberg channel on Apple Podcasts and follow the instructions there. Thanks for listening. Thank you. has the promo products to match. With thousands of options, from apparel and drinkware to tech and totes, it's easy to find the right fit for your brand and budget with standout choices at every price point. And with their 360 guarantee you can be 4imprint certain your order will show up just right right on time Explore more at 4imprint 4imprint 4certain When you running a business the best days are the ones where priorities stay on track. For midsize and large companies, risk can affect multiple parts of the organization at once, from property and liability to cyber and regulatory challenges. At that level, managing risk becomes an ongoing discipline. At the Hartford, the focus is on helping businesses manage risk before it turns into something more disruptive. And when losses do happen, that work is paired with insurance coverage shaped by years of underwriting, risk engineering, and claims experience. Learn more at thehartford.com slash risk mitigation. Policies provided by Hartford Fire Insurance Company and its property and casualty affiliates, Hartford, Connecticut. These days, it seems like AI agents are just about everywhere you turn, every field and every function. But without identity, you can't trust they'll serve your business instead of jeopardizing it. Fortunately, Okta helps you get identity right by securing your AI agent's identities, giving you a single layer of control, a single standard of trust. So whether an AI agent supports a single user or your entire enterprise, with Okta, you'll turn risk into opportunity. Secure every agent. Secure any agent. Okta secures AI.