Is Staking on Numerai Profitable? What 1,200 Rounds Show

Across ~1,200 Numerai rounds, stakers earned roughly 2.5M NMR against 1.1M burned — net profitable in aggregate, but the median round returns only 0.05% of stake.

Across 1,097 resolved Classic rounds, stakers have earned 2.44M NMR and lost 1.05M to burns — a net 1.39M NMR delivered to participants. Yet the median round returns just 0.054% of stake. The protocol pays out; the typical round barely moves the needle.

Numerai lets anyone stake NMR on a model's predictions. Stake well and you earn more NMR; stake poorly and your NMR is burned — destroyed on-chain, not redistributed. For a primer on the tournament mechanics, see How Numerai Works.

How Staking Works (Quick Refresher)

You can stake as little as 0.01 NMR on any model you control. Each round, your payout or burn is determined by your model's performance on two key metrics:

Payout = stake x clip(0.5 x CORJ60 + 2.0 x MMC, -0.25, 0.25) x payout_factor

Payouts and burns per round are capped at 25% of your stake. Stake 100 NMR on a strong round and you might earn up to 25 NMR; on a bad one, you can lose the same amount, and that NMR is gone for good.

The payout factor is a tournament-wide multiplier Numerai adjusts over time. For a fixed model score, it is one of the largest external drivers of realized NMR payout. See Round Economics for how it interacts with stake levels.

Cumulative Earned vs Burned

Tournament economics start with the aggregate NMR earned and burned across every participant.

Cumulative NMR earned, burned, and net across all Numerai tournament rounds, with earned plotted above zero, burns below zero, and net positive through the full history
Cumulative NMR earned, burned, and net across all Numerai tournament rounds, with earned plotted above zero, burns below zero, and net positive through the full history

Cumulative earned sits near 2.5M NMR against around 1.1M burned, leaving ~1.4M NMR of net payouts to stakers. The tournament pays out more than it destroys. The same series runs live on the dashboard home and round-by-round on the trends page.

The gap is lumpy. April 2023 alone — rounds 459, 464, 469 — burned 88K NMR against 6K earned, a net loss of 82K in three weeks. When model consensus breaks down, burn rates spike and the net curve flattens. A net-positive pool also does not guarantee a net-positive staker — aggregate numbers hide the individual distribution.

Payout by Stake Tier

Do bigger stakers earn more per round because they are better, or just because they have more at risk?

Median per-round NMR payout grouped by stake tier, with the 1K+ tier earning about 1.3 NMR median and smaller tiers near zero
Median per-round NMR payout grouped by stake tier, with the 1K+ tier earning about 1.3 NMR median and smaller tiers near zero

Median per-round payout scales almost linearly with stake size. The 1K+ tier earns a median of ~1.3 NMR per round (n=112,899 round-model observations), the 100–1K tier earns ~0.1 NMR, and the three smaller tiers (10–100, 1–10, and <1 NMR) sit at essentially zero. That pattern is what you see when percentage returns are roughly similar across tiers and only the base grows.

Sample sizes skew the picture too. Sub-1 NMR stakes account for 2.24M observations but produce almost no aggregate payout, while the 1K+ tier moves most of the tournament's NMR with far fewer observations. Big stakers dominate NMR flows, yet per unit staked they earn no dramatic premium. See payout ROI by stake tier for the rate-of-return view. Starting small does not push you into a structurally worse payout regime.

Payout Factor Trends

The payout factor is the biggest variable in staking profitability that you cannot control. It acts as a global dial on every payout and burn.

Payout factor over round number, starting at 1.0 in early rounds, dropping through rounds 250-450, and stabilizing around 0.10 from round 500 onward
Payout factor over round number, starting at 1.0 in early rounds, dropping through rounds 250-450, and stabilizing around 0.10 from round 500 onward

The factor started at 1.0 in the early rounds, dropped sharply through rounds 250-450, and has sat near 0.10 since round 500 -- roughly a 10x compression from the early era. A model earning 5 NMR per 100 staked in 2019 earns closer to 0.5 NMR on the same score in the compressed-factor era. The live curve sits next to earned and burned NMR on the round economics view.

When the factor drops, strong scores earn less and bad rounds burn less; when it rises, both sides amplify. Stakers who sized up during high-factor periods without strong models paid for it. A lower factor means you need better raw scores for the same NMR return.

Distribution of Per-Round Returns

Averages hide what the typical staker actually sees round to round.

Histogram of per-round return as payout divided by stake, tightly clustered around zero with a median of 0.05% and a thin positive tail extending further than the negative tail
Histogram of per-round return as payout divided by stake, tightly clustered around zero with a median of 0.05% and a thin positive tail extending further than the negative tail

Per-round returns pack tightly around zero. The median round returns 0.05% of stake — a rounding error on any one round. Most of the mass sits inside ±2%, with a thin positive tail stretching further out than the negative tail. That shape lines up with the aggregate net-positive number from earlier.

It also explains why staking can feel unrewarding in a net-positive system. A 0.05% median means many stakers go through long flat stretches before compounding shows up. Models that avoid deep burns tend to outperform over time even if they rarely top the leaderboard — a dynamic covered in Model Survival.

Practical Takeaways

Staking is net profitable in aggregate: roughly 2.5M NMR earned against 1.1M burned over tournament history. That headline hides a 0.05% median round return and wide variance between individual stakers.

What the data suggests:

  • Model quality is the main driver. Stake tier buys no percentage-return edge; larger tiers just move more absolute NMR because they stake more.
  • Watch the payout factor. It has compressed from 1.0 to around 0.10. A halved factor means you need double the raw model performance for the same NMR return.
  • Consistency beats volatility. The return distribution rewards models that avoid large burns. Slightly positive most rounds compounds better than big swings.
  • Size your stake to your conviction. The 25% per-round burn cap means a streak of bad rounds can erode stake quickly, and burned NMR is gone permanently.
  • Time horizon matters. With a median round return near 0.05%, short-term results are mostly noise. Expected value only shows up over dozens of rounds.

Staking on Numerai is not passive income. The aggregate pool has paid out more NMR than it burned, but individual profitability depends on sustained model edge, stake sizing, and the payout-factor regime.