Risk

Risk Management for Copy Trading on Polymarket

Copying a great wallet is not enough. The traders who keep gains over a year are the ones who size positions and cap losses correctly.

Last reviewed · Maria Ostrowski, Poly Syncer

Copy trading does not eliminate risk — it transfers stock-picking risk to the leader you follow and leaves sizing, drawdown, and category-concentration risk firmly with you. A practical risk framework for automated Polymarket trading through a copy-trading Polymarket bot caps single-position size at 3–5% of bankroll, sets a daily loss limit of 4–6%, requires a 30% stop-loss per mirror, diversifies across 2–3 low-correlation leaders, and whitelists categories you actually understand. Defaults on Poly Syncer are tuned to those numbers.

Why copy trading is not "set and forget"

The pitch for copy trading often implies that following a top wallet outsources risk management. It does not. The leader manages the directional risk of each trade. You manage:

None of those is automatic. They are the difference between a copy trader who is up 6–9% in a steady year and one whose account is at zero.

Stop-loss in prediction markets

A stop-loss on Polymarket is structurally different from a stop in equities. There is no continuous price; there is a binary outcome. A YES contract bought at $0.55 can drop to $0.30 on news, then resolve YES at $1.00 anyway. Over-tight stops therefore destroy edge.

The defensible compromise is a percentage stop on entry price, not a hard floor:

The right number is wallet-specific. If the leader you follow has historically held losers to resolution and won, a tight stop will cost you. Inspect the leader's exit behavior on the leaderboard profile before configuring a stop.

Daily loss cap

The daily loss cap is the single most important defense against bad days. A leader with a 30-day Sharpe of 3.0 still has 4–6% drawdowns; you do not want a single such day to compound into a 15% account drawdown because you were greedy on size.

Poly Syncer defaults to a 5% daily loss cap. When hit, the engine pauses your mirror for 24 hours and emails (well, dashboards — we don't ask for email) a notification. Recommended ranges:

Position-size range

Set a min and a max in USDC per single mirrored trade. The min keeps you from spending gas on dust trades; the max caps tail risk. From your dashboard → settings, recommended starting values for a $5,000 bankroll:

If your bankroll is $50,000 the max scales to $2,500 (still 5%), but only if the leader's average position is large enough that mirroring is feasible without slippage. The strategies page models this.

Diversification across leaders

Two leaders both trading politics on the same election will have ~0.7–0.9 return correlation. Following both is not diversification — it is leverage. Useful diversification looks like:

Leader A category Leader B category Typical 30d return correlation
PoliticsPolitics0.74
PoliticsSports0.11
CryptoEarnings0.22
Sports (NBA)Sports (Soccer)0.18

Two leaders in different categories typically run at <0.25 return correlation, which is meaningful diversification. Two leaders in the same category run at >0.7. Pick across the rows of the table, not down the same column.

Category gating

The default Poly Syncer install enables all 25 categories. We recommend tightening this aggressively. A common starter whitelist for a new copy trader:

Three uncorrelated categories is plenty. Add Crypto or Geopolitics later as you build comfort. The categories explained post covers the liquidity and resolution profile of each.

Drawdown psychology

The hardest part of copy trading is not the configuration. It is sitting through a 12% drawdown without unsubscribing. Some realities to internalize:

A leader with a 30-day Sharpe of 3.0 will still have, on average, one 8–12% drawdown per quarter. If you cannot tolerate that, your sizing is wrong.

Survivorship bias warning

Every leaderboard, including ours, only ranks wallets that are currently active. Wallets that traded aggressively, lost everything, and stopped trading are not visible. The cohort of "elite" wallets you see is partly a survivor population. Implications:

  1. The median top-100 Sharpe of 2.7 is real but it is the survivor median.
  2. Following one wallet exposes you to that wallet's full failure mode — including blowing up.
  3. Diversifying across 2–3 leaders is the cheapest protection against single-wallet survivorship.

We discuss this in detail on the methodology page.

Tax considerations

Poly Syncer does not provide tax advice and your jurisdiction's rules differ. A few neutral observations:

Speak to a qualified tax professional. We are not one.

Position-correlation: a worked example

Imagine you follow three leaders: a politics specialist (P), an earnings specialist (E), and an NBA churner (N). Their pairwise 30-day return correlations from our internal dataset:

P E N
P1.000.140.08
E0.141.000.11
N0.080.111.00

If each leader has a standalone 30-day return standard deviation of 12%, and you allocate equal capital across the three, the portfolio standard deviation is approximately 7.4% — substantially below the simple average. By contrast, three politics specialists with pairwise correlation 0.74 produce a portfolio standard deviation of about 11.0%, almost no improvement on a single leader. The math is the entire reason category diversification matters more than wallet count.

Concrete defaults by bankroll size

One-size-fits-all defaults break at the extremes. Recommended starting configuration by bankroll:

Bankroll Max position Daily cap Leaders Categories
$1k–$5k$50–$250 (5%)5%22
$5k–$25k$250–$1,250 (5%)5%33
$25k–$100k3% of bankroll4%44
>$100k2% of bankroll3%5–75+

Larger bankrolls move to lower percentage caps because Polymarket book depth is the binding constraint. A 5% mirror on a $200,000 bankroll is $10,000 — in many markets that is a meaningful share of top-of-book and will create slippage that erodes the leader's edge. The strategies page models depth-aware sizing in detail.

Pre-mortem checklist

Before you connect a wallet and follow a leader, run this 8-point check. If you can't tick all eight, do not start yet.

  1. Bankroll allocated is money you can lose without changing how you live.
  2. Max position is ≤ 5% of bankroll.
  3. Daily loss cap is ≤ 6% of bankroll.
  4. Stop-loss is configured (default 30%).
  5. You are following 2–3 leaders, not 10.
  6. The leaders are in different categories.
  7. You have read each leader's full trade history, not just the headline Sharpe.
  8. You have decided in writing what you will do at a 10% account drawdown.

Frequently asked questions

Does Poly Syncer enforce these limits or are they suggestions?

They are enforced. Position size, daily loss cap, stop-loss, and category whitelist are hard gates in the executor. The engine cannot place a trade that violates them, regardless of what the leader does.

What happens when the daily loss cap triggers?

The engine pauses all mirrors for 24 hours, marks the dashboard with a clear banner, and refuses to fill any leader trade in that window. After 24 hours mirrors resume automatically unless you've extended the pause manually.

Can I lose more than my deposit?

No. Polymarket positions are pre-funded in USDC; you cannot borrow against them and there is no margin. Your maximum loss on any trade is the USDC you put into it.

Should I use the Kelly criterion to size?

Sized properly, yes — but most copy traders should size at fractional Kelly (typically 0.25× to 0.5× full Kelly) because the leader's true edge is uncertain. We cover the math in our Kelly post.

Behavioral rules that beat clever ones

Most retail blowups in copy trading do not come from a bad leader pick. They come from behavioral mistakes on top of a reasonable leader pick. The rules below are deliberately simple because complexity invites discretion, and discretion invites the exact behaviors you are trying to automate around.

Boring rules, consistently applied, beat clever rules applied selectively. This is true in trading, in software, and in copy trading specifically.

Configure conservatively, then loosen

The pattern that works for most users is to start with Poly Syncer's defaults — 5% max position, 5% daily cap, 30% stop, 2–3 leaders — run for 30 days, and only then loosen any gate based on what you actually saw. Pricing is at /dashboard/billing; the leaderboard is at /leaderboard; the full risk schema is on /methodology.