The most active trader wins the dynasty title 34% more often than chance
In 4,235 public dynasty leagues, the team that made the most trades won the championship 12.4% of the time, against 9.3% by chance. Much of that edge tracks how active a manager is overall: waiver and free-agent moves predict titles even more strongly.
By Elliot Pollaro · Published 2026-10-06 · Pre-registered (DOI 10.17605/OSF.IO/F8K59)
Key findings
- The league's most active trader won the title 12.4% of the time (95% CI 11.4%–13.4%), against 9.3% expected by chance: 1.34 times the chance rate (p < 0.001).
- The pre-registered primary model compares teams only within their own league and holds waiver and free-agent activity equal. In it, one standard deviation more trading went with 1.07 times the odds of a title (95% CI 1.02–1.12). The same step up in waiver and free-agent moves went with 1.50 times the odds (95% CI 1.44–1.56).
- Exploratory: teams in the most active third of their league by trades won 11.1% of the time, the middle third 8.9% and the least active third 7.5%.
- Replicated in an independent sample of 2,082 leagues from 2024: the most active trader won 11.2% of the time against 9.0% by chance (1.24 times), and the primary model's trading odds ratio was 1.10 (95% CI 1.03–1.18).
What we found
Across 4,235 leagues and 46,632 teams in the 2025 season, the median league completed 11 trades. The team that made the most trades in its league won the championship 12.4% of the time (95% CI 11.4%–13.4%). If trading had nothing to do with winning, that figure would be 9.3%, the average of one divided by the number of teams. In the 3,001 leagues with a single most active trader, that team won 13.0% of the time (95% CI 11.8%–14.2%), against 9.1% by chance.
The pre-registered primary test compares each team only with the other teams in its own league, so league size and league culture drop out. It also holds waiver and free-agent activity equal, which separates trading from general engagement. On that test the trading effect is real but small: 1.07 times the odds of a title per standard deviation more trading (95% CI 1.02–1.12). Waiver and free-agent activity carries the bigger signal, at 1.50 times the odds (95% CI 1.44–1.56).
| Group | Title rate | 95% CI | Leagues |
|---|---|---|---|
| Most active trader | 12.4% | 11.4%–13.4% | 4,235 |
| Chance (1 ÷ teams) | 9.3% |
Title rates by trade activity (exploratory)
Within each league, teams were split into the most, middle and least active third by number of trades. The title rate falls steadily from the most active third (11.1%) to the least (7.5%). This breakdown was frozen into the analysis before the results were unblinded. It is not part of the registered plan, so it is labeled exploratory.
| Trade activity in its league | Title rate | 95% CI | Teams | Titles |
|---|---|---|---|---|
| Most active third | 11.1% | 10.6%–11.5% | 13,672 | 1,513 |
| Middle third | 8.9% | 8.6%–9.2% | 17,499 | 1,555 |
| Least active third | 7.5% | 7.2%–7.9% | 15,461 | 1,167 |
A twist: heavy traders don't win more regular-season games
The registered exploratory outcomes hold waiver activity equal. On them, heavier traders did not have better regular seasons. Each standard deviation more trading went with -0.30 regular-season wins (95% CI -0.35 to -0.25) and 0.88 times the odds of making the playoffs (95% CI 0.85–0.90). Yet the same teams won more titles.
One reading is that heavy traders are two groups: rebuilders, who sell for the future and lose now, and contenders, who buy to win now. These outcomes are exploratory, and none of this shows that trading causes titles.
| Outcome (exploratory) | Per SD more trading, waiver activity held equal | 95% CI |
|---|---|---|
| Made the playoffs (odds ratio) | 0.88 | 0.85–0.90 |
| Regular-season wins | -0.30 | -0.35 to -0.25 |
| Points-for rank (higher = worse) | +0.19 | +0.15 to +0.24 |
Robustness checks
The headline held in every pre-registered subgroup. The adjusted trading effect was smaller, and not statistically significant, in 8–10-team leagues (95% CI 0.94–1.08) and in 1QB leagues (95% CI 0.96–1.13).
| Leagues | Count | Most active trader won | × chance | Odds ratio per SD of trading (95% CI) |
|---|---|---|---|---|
| Leagues with 5+ trades | 3,213 | 12.8% vs 9.1% | 1.41× | 1.09 (1.04–1.15) |
| 8–10 teams | 1,975 | 13.5% vs 10.5% | 1.29× | 1.01 (0.94–1.08) |
| 11–12 teams | 2,085 | 11.3% vs 8.3% | 1.35× | 1.10 (1.04–1.17) |
| 13–16 teams | 175 | 13.6% vs 6.8% | 1.99× | 1.27 (1.02–1.58) |
| Superflex | 2,596 | 12.1% vs 9.1% | 1.33× | 1.08 (1.02–1.15) |
| 1QB | 1,639 | 12.9% vs 9.5% | 1.36× | 1.04 (0.96–1.13) |
Does it hold in 2024? Yes
The same frozen code was run on an independent random sample from the 2024 season: 2,082 eligible public dynasty leagues. The most active trader won the title 11.2% of the time (95% CI 9.8%–12.6%), against 9.0% by chance: 1.24 times the chance rate (p < 0.001).
In the primary model, one standard deviation more trading went with 1.10 times the odds of a title (95% CI 1.03–1.18), and the same step up in waiver and free-agent activity went with 1.46 times the odds (95% CI 1.37–1.54). The finding replicated.
How the study works
Trading counts every completed trade a team was part of during the league-season, offseason trades included. A three-team trade counts once for each team in it. Engagement counts completed waiver claims and free-agent pickups. The outcome is winning the championship final.
- Primary analysis: conditional logistic regression with each league as its own stratum, since each league has exactly one champion. Trading and waiver/free-agent activity enter as standardized log counts.
- Headline statistic: how often the league's most active trader won the title. Teams tied for most trades share the credit. The exact 95% interval is Clopper–Pearson; the p-value comes from 20,000 simulated seasons with a random champion.
- Robustness checks: leagues with at least 5 trades, league sizes, superflex vs 1QB, and an independent sample from the 2024 season.
Who is in the sample
Sampled leagues, and how many each eligibility rule excluded.
| Sampled leagues | Count |
|---|---|
| No games played | 446 |
| More than one team without an owner | 137 |
| Not a completed dynasty season (or best ball) | 162 |
| No decided championship | 20 |
| Analyzed | 4,235 |
Limitations
- This is an association, not proof that trading causes titles. Strong teams may trade more because they can, contenders trade to win now, and abandoned teams stop trading and making waiver moves.
- Trade counts do not measure trade quality. A team can make many trades and still lose value.
- One platform's public dynasty leagues in one season. Private leagues, other platforms and other formats may differ.
Data, method and verification
Public dynasty leagues on a major fantasy football platform, read from the platform’s own public league records (settings, rosters, weekly lineups and scores, transactions, playoff brackets). None of Dynasty Dealer’s player values, trade database or users are used.
The plan was registered before any result was seen: Dynasty League Case Studies: Pre-Registered Analysis Plan (DOI 10.17605/OSF.IO/F8K59). Every per-league number was reproduced by a second, independently written implementation (another language, no shared code), and randomly chosen leagues were checked by hand against the raw league records.
How to cite
Elliot Pollaro (2026). The most active trader wins the dynasty title 34% more often than chance. Dynasty Dealer. https://www.dynastydealer.com/case-studies/does-trading-win-championships. Pre-registration: https://doi.org/10.17605/OSF.IO/F8K59
FAQ
Do dynasty teams that trade more win more championships?
Modestly, yes, in 4,235 public dynasty leagues: the most active trader won the title 12.4% of the time, against 9.3% by chance. Once overall activity is held equal the trading effect is small, and waiver and free-agent activity predicts titles more strongly.
Where does the data come from?
From the public league records of a major fantasy football platform: each league's settings, rosters, transactions and playoff bracket. None of Dynasty Dealer's player values, trade database or users are used.
Was the analysis planned before looking at the results?
Yes. The plan was registered on OSF before any result was seen (DOI 10.17605/OSF.IO/F8K59). The analysis code was written on shuffled outcomes, frozen, and then run once on the real ones.
Does this mean I should make more trades?
Not by itself. The study shows how trading and titles move together across real leagues. It cannot show that any one trade will help your team.