{"id":"7a932084-a45a-4be9-8252-70631183aaef","arxiv_id":"2607.06166","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.5,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Proper betting s=∇G(p)−∇G(q) is the essentially unique strategy that converts a proper-score edge over market prices into robust expected profit in general prediction markets.","lead":"The paper shows that a specific “proper” betting rule turns a forecast that beats the market under any proper scoring rule into positive expected profit in general prediction markets, not just automated market makers. It also proves this rule is essentially unique, and backs the theory with AI-forecast backtests and a live Kalshi deployment.","discovery_kind":"unification","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified beyond the reader's liquidity-offset caveat; Theorems 1–2 and the profit decomposition are internally consistent.","rationale":"The central claim is the profit decomposition (score gap + Bregman − liquidity loss) for s_G=∇G(p)−∇G(q) and the essential uniqueness of any robustly profitable strategy. Both rest on elementary convex analysis (Proposition 1 / McCarthy characterization and the definition of Bregman divergence) and hold pointwise for every realized outcome, so the empirical-score and sequential extensions follow immediately. The AMM recovery (Fact 1, Corollary 1) is the expected special case in which D_G exactly cancels L_ρ. The only place the guarantee can fail is precisely when liquidity loss exceeds the Bregman term—the condition the reader already isolates. Offline experiments deliberately set impact to zero; the live +80% ROI is consistent with a Conservative persona under Brier but is too short and filtered to stress-test the offset condition. No algebraic inconsistency or circularity appears. Therefore the reader's CONDITIONAL / MODERATE verdict already correctly balances acceptance of the theory against the practical liquidity caveat; no adjustment is warranted.","tokens_in":31339,"tokens_out":625,"duration_ms":9433,"concrete_test":"Recompute the offline ROI tables (Table 2 / Table 6) after injecting a simple linear price-impact model L_ρ(s;q)=(α/2)∥s∥² with α calibrated to typical Kalshi book depth (e.g., 1–5% of notional for the median position size used); if Proper (Brier) remains the only strategy with reliably non-negative ROI for models with ΔS>0 while heuristics stay negative, the practical claim survives the liquidity stress the theory already flags.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reader's weakest assumption (D_G(q,p) ≥ L_ρ(s_G;q) as an exogenous market-depth condition) is already the correct soft spot. Lemma 1 and Theorem 1 follow directly from the Bregman identity for any convex G; Theorem 2's uniqueness argument (normalization to the hyperplane orthogonal to 1, subsequence limits, and construction of p*_t = q+v with v·s* > c > 0 > v·ŝ) is standard and does not introduce a hidden algebraic gap. The offline tables set L_ρ=0 by design, and the live Kalshi run is short, filtered, and not a pure test of the robust-profitability claim. None of these points overturn the formal equivalence or the uniqueness result under the stated conditions.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper resolves the gap between classical AMM theory (accuracy equals profit) and modern CLOB prediction markets by defining, for any strictly proper scoring rule S with potential G, a proper betting strategy s_G(p,q)=∇G(p)−∇G(q). Theorem 1 decomposes expected profit into score gap + Bregman divergence − liquidity loss, so the strategy is robustly profitable whenever the forecast beats the market under S and D_G offsets L_ρ. Theorem 2 shows any robustly profitable strategy is essentially the same as s_G (up to rescaling and constant shift). Proposition 2 recharacterizes proper scoring rules via robust profitability; Corollary 1 recovers the classical AMM guarantee as the special case where Bregman cancels liquidity loss. Extensions cover empirical scores, bid–ask spreads, and sequential trading. Empirically, proper betting is the only strategy that reliably converts AI forecast accuracy into ROI on thousands of Kalshi markets; persona analysis links model behavior to preferred scoring rules; a 26-day live Kalshi deployment with Gemini 3 yields +80.33% ROI and Sharpe 3.35.","tokens_in":31632,"tokens_out":1292,"duration_ms":22137,"significance":"If the results hold, the paper supplies the missing formal bridge between proper scoring rules and profitability on general (including CLOB) prediction markets, strictly generalizing Hanson’s AMM theory. The profit decomposition cleanly explains both the accuracy–profit paradox and how uninformed strategies can still earn money via Bregman divergence. The uniqueness result and the new characterization of proper scores are of independent theoretical interest. Empirically, the work is strengthened by large-scale AI forecast evaluation, a falsifiable persona taxonomy, and a real-capital live deployment with documented trades—rare for theory papers in this area. The algebraic identities follow from standard convex-analysis representations (McCarthy/Gneiting–Raftery) without free constants fitted to ROI, which is a genuine strength.","major_comments":[{"comment":"Theorem 1’s robust-profitability guarantee is conditional on D_G(q,p) ≥ L_ρ(s_G;q). Offline experiments in §4.1 explicitly set zero price impact (L_ρ=0), so they test only the frictionless special case of the decomposition, not the full claim. The live deployment (§4.3) reports ROI and a Brier decomposition (ΔS=+0.7205, D=+0.0828) but does not measure or bound realized liquidity loss / slippage against D_G. Either report order-book depth / fill-cost statistics that verify the offset condition on traded markets, or state more sharply that the offline tables and live ROI are evidence for the score-gap+Bregman mechanism under low friction, not a direct test of robust profitability under nontrivial L_ρ.","section":null},{"comment":"Definition 3 and Theorem 2 define “essentially the same” via rescaling, constant shift by λ1, and non-vanishing difference along some sequence p_t→q. The proof construction (normalize to 1^⊥, subsequence limits ŝ,s*, pick v with v·s*>c>0>v·ŝ, set p*_t=q+v) is standard and appears correct for frictionless markets. However, the manuscript never checks whether natural practical modifications—e.g., clipping bets inside the bid–ask no-trade zone of §C.1, or discrete share/tick constraints in the live executor—remain “essentially the same” or can break robust profitability. A short remark or corollary clarifying which real-market truncations preserve the guarantee would make the uniqueness claim operationally usable.","section":null}],"minor_comments":[{"comment":"Examples 1–3 are helpful but use hand-picked p,q,p* triples; a one-line note that they are illustrative (not sampled from real books) would avoid over-reading.","section":null},{"comment":"Table 2 vs Table 6: the main text shows a 200-event subset while Appendix D.2 reports the full standardized set; cross-reference the subset construction more clearly in §4.1 so readers know which numbers are comparable.","section":null},{"comment":"Persona generation (§D.4) uses Beta modes, slopes, and a0 calibrated to a fixed ±0.05 Brier gap. These free parameters are fine for a controlled experiment but should be listed explicitly in the main text near Figure 1 so the taxonomy is reproducible without the appendix.","section":null},{"comment":"Live eligibility filters (2–14 day horizon, ≥10¢ move, MENTIONS exclusion, resolution-clarity drop) are material; a short sensitivity note (e.g., ROI if the 10¢ filter is removed) would strengthen §4.3.","section":null},{"comment":"Notation: ρ is introduced as price-impact and later as post-trade spot; a single sentence that ρ(s) is the marginal price path and L_ρ is its integral excess cost would help readers less familiar with AMM/CLOB abstractions.","section":null},{"comment":"Typos / polish: “them arket score” spacing in §2; arXiv id in the header is 2607.06166 while the live dates are April/May 2026—confirm consistency of the preprint versioning note.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The formal core (Lemma 1, Theorems 1–2, Proposition 2, Corollary 1) is clean and publishable; the liquidity-offset caveat is already the right soft spot and is fixable with clearer scope language and optional depth statistics rather than a redesign. The live +80% ROI is eye-catching but short and heavily filtered—editors may want authors to tone marketing language slightly so the theory, not the one-month Sharpe, carries the paper. Fit for a serious cs.AI / EconCS venue is good."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The real contribution is the proper bet s_G(p,q)=∇G(p)−∇G(q) for any strictly proper score, plus the three-term decomposition (score gap + Bregman − liquidity loss) that works for arbitrary price-impact ρ, not just AMM cost functions. Theorems 1–2 and Proposition 2 are the core: robust profitability whenever the score edge exists and Bregman covers slippage, and uniqueness up to rescaling and constant shift. Corollary 1 correctly recovers the classical Hanson/Chen–Pennock guarantee when Bregman cancels liquidity loss. That is new relative to the AMM literature and is cleanly done.\n\nLemma 1 is just the Bregman identity rewritten as idealized profit; the rest follows. The uniqueness argument (normalize onto the hyperplane orthogonal to 1, take subsequence limits, construct p*_t = q+v with opposite signs on the two limit directions) is technical but standard and does not hide a gap. Bid-ask and sequential extensions are sensible reductions, not afterthoughts.\n\nSoft spots are practical and already flagged correctly by the reader. Robust profitability is conditional on D_G ≥ L_ρ; that is an exogenous depth condition, not free from accuracy. Offline tables set L_ρ=0 by design, so they mainly show that proper bets dominate the listed heuristics under ideal fills. The live Kalshi run (+80% ROI, Sharpe 3.35) is short, filtered (horizon, 10¢ move, category exclusions), and uses a momentum rebalancer—useful existence proof, not a pure test of the theorem. Personas are a useful empirical taxonomy for choosing among Brier/Log/Spherical, not a formal result. Code/data package is not shipped.\n\nWho this is for: people who trade or design prediction markets, or who build AI agents that size positions from probabilistic forecasts. The math is solid enough that a serious editor should send it out. I would cite the decomposition and the uniqueness claim; I would treat the live number as anecdote until replication and liquidity stress tests exist. Engage with the theory; do not over-weight the deployment.","headline":"Clean generalization of the AMM accuracy–profit link to arbitrary price impact, with a usable uniqueness theorem; live ROI is illustrative, not the load-bearing claim.","tokens_in":32224,"tokens_out":535,"would_cite":true,"duration_ms":6258,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"A proper betting strategy tied to any proper scoring rule converts an accuracy edge over the market into guaranteed expected profit in liquid prediction markets, and is essentially the only strategy that does so.","keywords":["prediction markets","proper scoring rules","proper betting","Bregman divergence","central limit order books","automated market makers","liquidity loss","forecasting personas"],"falsifier":"On a liquid order-book market, take a forecast that beats the market under a chosen proper scoring rule yet still loses money under the corresponding proper bet while a clearly different normalized strategy earns positive expected profit; that would refute the robust-profitability and uniqueness claims.","tokens_in":32211,"feed_emoji":"📈","tokens_out":1013,"duration_ms":16059,"temperature":0.7,"pith_summary":"Classical prediction-market theory equates better forecasts with trading profit, but only for automated market makers. On the central limit order books that dominate real exchanges, accurate forecasters often lose money and crude heuristics can still make money. This paper closes that gap: for every strictly proper scoring rule it defines a matching “proper” bet that depends only on the forecaster’s probability and the market price, and proves that the bet’s expected profit equals the score gap plus a Bregman divergence term minus liquidity loss. Whenever the forecast beats the market under that scoring rule and liquidity is deep enough for the divergence to cover slippage, expected profit is positive. The same strategy is essentially unique among strategies that enjoy this robust guarantee, and the equivalence also yields a new characterization of proper scoring rules themselves. Offline tests on thousands of AI forecasts and a live month-long deployment confirm that proper betting is the only allocation that reliably turns accuracy into returns.","feed_headline":"Only proper bets turn forecast edge into market profit","feed_subtitle":"A scoring-rule identity pins down the unique robust strategy; live trading returned +80% ROI","key_machinery":"Proper betting: the position s_G(p,q)=∇G(p)−∇G(q) induced by the potential of a proper scoring rule. Its profit identity (score gap + Bregman divergence − liquidity loss) generalizes the classical AMM guarantee and pins down both existence and uniqueness of robustly profitable strategies.","core_discovery":"For any strictly proper scoring rule with convex potential G, the proper position s_G(p,q)=∇G(p)−∇G(q) has expected profit exactly equal to the score gap between forecast p and market price q, plus the Bregman divergence D_G(q,p), minus liquidity loss. Hence the strategy is robustly profitable whenever p outperforms q under the scoring rule and divergence offsets slippage; moreover any robustly profitable strategy is essentially a rescaling or constant shift of this proper bet.","pith_inferences":["Exchanges could publish the proper-bet map for a chosen scoring rule as a default “informed order” type, making accuracy-to-profit conversion the default rather than an expert skill.","Persona-dependent rule choice suggests an adaptive meta-layer that estimates a forecaster’s margin–accuracy profile and switches among Brier, log, or spherical weights in real time.","The uniqueness result implies that any profitable black-box trading bot whose edge is purely informational must be approximately implementing some proper bet, which could be audited by recovering the implied potential G.","Market designers choosing liquidity subsidies or fee schedules are effectively choosing how large the Bregman term must be before informed traders can profit, shaping who participates."],"forward_implications":["Accuracy measured by a proper scoring rule can be turned into expected trading profit by a single explicit position that needs only p and q, without knowledge of the true probability.","Heuristic rules such as Kelly or max-margin can lose even with a score edge, and can win without one, because they ignore the Bregman term.","In automated market makers the proper bet recovers the classical “move price to your belief” trade and the score-gap profit formula as the special case where divergence exactly cancels liquidity loss.","Different scoring rules (Brier, log, spherical) weight disagreement differently, so the best proper strategy depends on how a forecaster’s errors concentrate across margins.","The same decomposition extends to empirical scores, bid–ask spreads, and multi-period rebalancing, giving operational rules for live books."],"fun_headline_variants":["Only proper bets convert forecast edge into robust market profit","Scoring-rule proper bets uniquely link accuracy to trading gains","Proper strategy equals score gap plus divergence, minus slippage","Any robust profit strategy is rescaling of the proper score bet","Live proper betting on Kalshi yielded +80% ROI Sharpe 3.35"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The market must be liquid enough that the Bregman divergence between forecast and price covers the slippage cost of the proper bet; accuracy alone does not guarantee that.","fun_headline_variants_meta":{"raw":{"variants":["Only proper bets convert forecast edge into robust market profit","Scoring-rule proper bets uniquely link accuracy to trading gains","Proper strategy equals score gap plus divergence, minus slippage","Any robust profit strategy is rescaling of the proper score bet","Live proper betting on Kalshi yielded +80% ROI Sharpe 3.35"]},"model":"grok-4.5","effort":"low","cost_usd":0.003052,"raw_usage":{"total_tokens":1118,"prompt_tokens":818,"num_sources_used":0,"completion_tokens":88,"cost_in_usd_ticks":30520000,"prompt_tokens_details":{"text_tokens":818,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":212,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":818,"tokens_out":88,"duration_ms":2679,"temperature":1.0,"reasoning_tokens":212,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T16:05:26.947207+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"On a liquid order-book market, take a forecast that beats the market under a chosen proper scoring rule yet still loses money under the corresponding proper bet while a clearly different normalized strategy earns positive expected profit; that would refute the robust-profitability and uniqueness claims.","supporting_citations":[],"review_version":2}