Learn the 7-step evaluation framework you can apply to any contract on Kalshi, Polymarket, FanDuel Predicts, or any other platform in under five minutes.
Last updated
Most prediction market losses start before the trade, not after it. The contract looked mispriced. The probability seemed off. You bought in, and then discovered the resolution criteria referenced a data source that didn’t measure what you thought it measured. Or the spread was so wide that your “edge” evaporated the moment you entered.
Evaluating a prediction market contract takes less time than placing the trade itself, yet most traders skip it entirely. They check price, form an opinion, and click buy. Professional traders in any market, from equities to sports betting, run a pre-trade checklist. Prediction markets deserve the same discipline.
This article gives you a 7-step evaluation framework you can apply to any contract on Kalshi, Polymarket, FanDuel Predicts, or any other platform in under five minutes. Each step targets a specific failure mode that costs real money. If a contract fails any step, you move on. No exceptions.
Sports bettors spend minutes researching a game before placing a wager: checking injury reports, comparing lines across books, reviewing recent performance data. Traders who understand prediction market basics often still skip the one step that separates consistent profitability from expensive lessons. They see a price, disagree with it, and trade.
The cost of that shortcut is specific and avoidable, ranking among the most expensive mistakes new prediction market traders make. Contract disputes on prediction markets arise almost exclusively from resolution ambiguity, not from market manipulation or platform failure.
A contract asking “Will Company X exceed $10B in revenue?” sounds straightforward until you realize the resolution source uses GAAP revenue while the headline number in the earnings release uses non-GAAP. The event happened. The payout didn’t.
A structured evaluation process eliminates these mistakes. The framework below works on any platform, any contract category, and takes under five minutes once you’ve internalized the steps. Treat it like a pre-flight checklist: every item gets checked, every time, regardless of how obvious the trade looks.
Before committing capital to any prediction market contract, run through these seven checks in order. If a contract fails on any step, stop. Move to the next opportunity.
The rest of this article develops each step with worked examples and platform-specific context so you can apply the checklist immediately.
Resolution criteria determine whether you get paid, and they vary significantly across platforms. On Kalshi, contracts resolve against named third-party data sources: the Associated Press for election results, ESPN for sports outcomes, government data agencies for economic indicators, and CF Benchmarks for crypto price contracts.1Kalshi, “Platform Features & Resolution,” kalshi.com, March 2026 On Polymarket’s global platform, the UMA Optimistic Oracle handles resolution mechanics through a decentralized dispute mechanism where token holders vote on outcomes if the initial resolution is challenged.2Polymarket, “Platform Features & UX,” docs.polymarket.com, March 2026 FanDuel Predicts lists contracts through CME Group’s exchange infrastructure, with resolution tied to CME-defined contract specifications.3CME Group, “Prediction Markets,” cmegroup.com, March 2026
These differences matter practically. A Kalshi contract resolving against an AP call produces a definitive, rapid result with no ambiguity. A Polymarket contract resolved through the UMA Oracle introduces a dispute window where the outcome can be challenged, potentially delaying settlement. Neither approach is inherently better, but each carries different risk profiles you should factor into your evaluation.
What to look for in resolution criteria:
| Quality Indicator | Good Sign | Red Flag |
|---|---|---|
| Data source | Named, specific (“BLS CPI report”) | Vague (“general consensus”) |
| Measurement definition | Precise (“GAAP revenue as filed with SEC”) | Ambiguous (“revenue exceeds $X”) |
| Timing | Clear date or trigger (“by 11:59 PM ET, June 30”) | Open-ended (“by end of year”) |
| Dispute mechanism | Defined process (UMA Oracle, exchange arbitration) | Silent on disputes |
Expert Tip
From an operator perspective, platforms invest heavily in resolution criteria design because disputes destroy user trust. Contracts citing specific, verifiable data sources (“Will the BLS report CPI above 3.0% for March 2026?”) are well-designed. Contracts with subjective criteria (“Will inflation be high?”) invite disputes. If you can’t identify the exact data source and the exact threshold, the contract fails Step 1.
A contract with vague resolution criteria isn’t a trade. It’s a dispute waiting to happen.
A mispriced contract with no liquidity isn’t an opportunity. It’s a trap. Before evaluating whether the price is wrong, confirm you can actually enter and exit the position at prices close to what you see on screen.
Volume signals whether enough informed participants have traded to establish a meaningful price. A contract with $50 in total volume and a price of $0.45 is not “the market pricing in a 45% probability.” It’s one person’s opinion with no counterparty pressure.
On high-volume political and sports markets, Polymarket produces tight spreads of 1 to 3 cents.4Polymarket, “Liquidity Assessment,” docs.polymarket.com, March 2026 Lower-volume markets on either platform can show spreads of 5 to 8 cents or wider. The spread functions similarly to the vig built into sportsbook pricing, except here you can see it in real time.
Spread determines your real entry cost. If a YES contract shows an ask of $0.52 and a bid of $0.44, your 8-cent spread means you’re paying $0.52 to enter a position you could only immediately exit at $0.44. Your “edge” needs to exceed that 8-cent round-trip cost to be profitable.
| Spread Range | Assessment | Action |
|---|---|---|
| 1 to 3 cents | Tight. Normal for high-volume contracts. | Proceed to fee analysis. |
| 4 to 6 cents | Acceptable if your edge is significant. | Only trade if your probability gap exceeds 10 cents. |
| 7+ cents | Wide. Likely low liquidity. | Walk away unless you have exceptional conviction and can use limit orders. |
Pro Tip
Use limit orders instead of market orders on contracts with spreads above 3 cents. You set your price rather than accepting the ask, which preserves your edge. Both Kalshi and Polymarket support limit orders through their order book interfaces.5Kalshi, “Trading Interface (CLOB),” docs.kalshi.com, March 2026
Liquidity isn’t static. Check volume trends, not just the snapshot. A contract that traded $500 yesterday and $50 today is losing attention, which means the spread will likely widen further as you hold.
Your probability estimate says the contract is mispriced. The spread is acceptable. Now run the fee math, because a 5-cent edge can shrink to break-even depending on the platform and contract price.
Fee structures differ meaningfully across the three major regulated platforms. Consider a $100 position (200 contracts at $0.50) where you believe the true probability is 60%, giving you a 10-cent expected edge per contract:
| Platform | Fee Calculation | Fee on $100 Position | Net Edge/Contract |
|---|---|---|---|
| Kalshi (general) | round_up(0.07 x 200 x 0.50 x 0.50) = $3.50 | $3.50 | ~$0.082 |
| Polymarket (global, politics 1.00%) | Dynamic taker fee at $0.50 price | ~$2.50 | ~$0.087 |
| Polymarket (global, crypto 1.80%) | Dynamic taker fee at $0.50 price | ~$4.50 | ~$0.077 |
| Polymarket US DCM | 0.30% x (200 x $0.50) = $0.30 | $0.30 | ~$0.10 |
| FanDuel Predicts | CME: $0.01 x 200 = $2.00 + FanDuel: 2% x $200 = $4.00 | $6.00 | ~$0.07 |
6Kalshi, “Fee Schedule,” kalshi.com/docs/kalshi-fee-schedule.pdf, February 20267Polymarket, “Trading Fees,” docs.polymarket.com/trading/fees, April 2026. Note: taker fees expanded to most categories effective March 30, 2026.8CME Group, “Event Contracts Fee Schedule,” cmegroup.com, February 2026; PokerNews, “FanDuel Predicts Review,” pokernews.com, 2026
The same trade produces meaningfully different returns depending on where you execute it. On Polymarket’s US exchange, the 0.30% taker fee barely dents a 10-cent edge. On the global platform, category-dependent taker fees now range from 0.75% (sports) to 1.80% (crypto), making the cost picture more nuanced than the platform’s former zero-fee reputation suggests. On FanDuel Predicts, dual-layer fees consume 30% of your expected profit on this position size.
Three rules for fee-adjusted evaluation:
First, calculate your expected edge in cents per contract. Subtract the per-contract fee. If the result is negative or near zero, the trade isn’t worth the execution risk.
Second, recognize that fee impact is price-dependent. Kalshi’s formula charges maximum fees at $0.50 (where P x (1-P) is maximized) per the platform’s published fee schedule and minimum fees near $0.00 or $1.00. If you’re trading contracts priced near the extremes, Kalshi’s fee impact drops substantially.
Third, factor in round-trip costs. If you plan to sell before resolution rather than hold to expiry, double the entry fee for your cost estimate. Settlement is free on all three platforms, but early exits cost the entry fee again.
Fee comparison alone doesn’t determine where to trade. A contract with tighter spreads on a higher-fee platform may net better execution than a zero-fee platform with wide spreads. Total cost of trading is fees plus spread, not either alone.
Time-to-expiration changes how a contract behaves. If you’ve traded options, the parallel is direct: prediction market contracts exhibit accelerating price convergence as resolution approaches, similar to theta decay. A contract trading at $0.55 with three weeks until resolution moves differently than the same contract at $0.55 with 48 hours left.
With weeks remaining, new information can move the price in either direction. Your edge has time to play out, but so does the risk of an information event reversing your thesis. Within 48 hours, most contracts gravitate toward $0.00 or $1.00 as the outcome becomes increasingly clear. The remaining uncertainty compresses into a narrow window where prices can swing sharply on final news.
Warning
Contracts priced between $0.40 and $0.60 within 48 hours of resolution carry the highest volatility relative to their remaining time. The outcome is genuinely uncertain, and a single data release or news event can move the price 15 to 20 cents. This is where both the most edge and the most destruction live for active traders.
Your go/no-go decision matrix:
| Scenario | Action |
|---|---|
| All 6 checks pass | Size the position per your bankroll rules and trade. |
| Resolution criteria unclear | Pass. No amount of edge compensates for payout uncertainty. |
| Spread exceeds your edge | Watchlist. Set a limit order below the ask and wait. |
| Fee-adjusted edge near zero | Pass, or look for the same contract on a lower-fee platform. |
| Time-to-expiry mismatch | Reassess position size. Shorter time = smaller position. |
Applying this checklist takes repetition before it becomes instinctive. The first five times feel slow. By the twentieth, it takes under five minutes and saves you from the trades that looked obvious but weren’t.
Every contract on every platform passes or fails the same seven tests. Resolution source, contract wording, volume, spread, fee-adjusted edge, time to expiration, and the final go/no-go. The checklist works because prediction market losses are predictable: they come from contracts with vague resolution criteria, illiquid order books, or edges too thin to survive fees.
The traders who consistently profit in prediction markets aren’t the ones with better information. They’re the ones who avoid the trades that look good but aren’t. Discipline compounds. Every bad trade you skip protects capital for the ones that actually pay.
If you’re ready to apply this framework, Kalshi, Polymarket, and FanDuel Predicts all offer contracts across politics, economics, sports, and more. Start with one contract, run the checklist, and build the habit before sizing up.