A good tournament ROI for a winning poker player is typically somewhere in the range of 5–50%, depending heavily on field size and game difficulty, but that number is nearly meaningless until you have several hundred tournaments in your sample. Here is what the benchmarks actually look like and why variance makes your current ROI almost certainly a mirage.
What Tournament ROI Actually Measures
Return on investment in MTTs is straightforward: take your total winnings, subtract your total buy-ins (including rebuys and add-ons), then divide by those buy-ins. Multiply by 100 to express it as a percentage.
Formula: ROI = (Total Winnings − Total Buy-ins) ÷ Total Buy-ins × 100
A worked example: you play 200 tournaments with a $100 average buy-in, spending $20,000 in total. Your cashes add up to $23,000. Your ROI is ($23,000 − $20,000) ÷ $20,000 × 100 = 15%.
That 15% means for every dollar risked, you returned $1.15 over those 200 events. Whether 200 tournaments is enough to trust that figure is a separate, more important question (spoiler: it usually is not).
For a deeper look at what metrics to track alongside ROI, including in-the-money percentage and average buy-in, the how to track tournament results guide covers the full picture.
ROI Benchmarks by Field Size
There is no universal "good ROI" because the number trades off against variance in a direct relationship: softer, smaller fields have less variance but lower realistic ROI ceilings; large-field majors offer massive payouts but so much variance that even strong players can post negative ROI over thousands of games. The table below reflects consensus ranges from high-volume players and poker coaches across multiple formats.
| Format | Typical Field Size | Realistic ROI for a Solid Winner | Variance Level |
|---|---|---|---|
| Sit-and-Gos (9-max) | 9 players | 3–8% | Low |
| Small MTTs (online regs) | 100–300 players | 8–20% | Moderate |
| Mid-size MTTs (rec-heavy) | 300–1,000 players | 15–40% | High |
| Large-field MTTs / Sunday Majors | 1,000–5,000+ players | 20–60%+ (over enormous samples) | Extreme |
| Live casino MTTs (mixed fields) | 100–500 players | 20–50% | High |
A few things are worth spelling out. SNGs have tighter ROI ceilings because the flat payout structure (typically top 3 of 9) means there are no monster scores to skew the mean upward. A 5% ROI over 2,000 SNGs is genuinely strong. Contrast that with large-field MTTs where the top 3 finishers might collect 30–40% of the prize pool: a single deep run can inflate ROI dramatically, and a long stretch without one will crater it equally hard.
These are ranges over large samples, not guarantees. A player running 50% ROI over 150 large-field MTTs is almost certainly running hot rather than demonstrating a true edge that large.
Why MTT Variance Dwarfs Cash Game Variance
In cash games, your results converge toward your true win rate at a comparatively rapid pace. Most of your expected value is earned in small, recurring edges across thousands of hands. MTTs work differently: the payout structure is top-heavy by design. A typical 100-player MTT might pay out 15 spots, with the top three places accounting for 40–50% of the entire prize pool. That means the vast majority of your long-run EV is concentrated in a small number of very good outcomes.
The practical consequence: you can play fundamentally correct poker and lose money over 200, 300, or even 500 tournaments simply because those big finishes have not landed yet. This is not a fluke or a sign that your strategy is wrong. It is arithmetic. The standard deviation on a single tournament result is enormous relative to its expected value, and a bad run can string together long enough to discourage even an experienced player.
Late-stage dynamics make this even more complex. ICM reshapes the value of every chip as you approach a final table, meaning the skill required to convert a chip lead into a proportionally good finish is real and learnable, but it still does not eliminate the luck component of being the one who wins the all-in with two players left. Understanding bubble strategy is part of capturing more of your EV, but even perfect play does not remove variance.
How Large a Sample Do You Actually Need
This is where most players dramatically underestimate the problem. Because cash game players often talk about needing 50,000–100,000 hands to have a statistically meaningful win rate, tournament players sometimes assume a few hundred MTTs is equivalent. It is not. Hands and tournaments are not comparable units of variance.
The rough consensus among high-volume tournament players and coaches:
- SNGs: 500–1,000 games before your ROI starts to settle meaningfully. The lower variance format helps, but outlier scores still exist.
- Small to mid-size MTTs: 500–1,500 tournaments before you can begin drawing confident conclusions, and even then the confidence intervals are wide.
- Large-field MTTs: 1,000–3,000+ tournaments. A handful of final table runs can make or break ROI over any sample shorter than this.
If you want a concrete sense of how much your current ROI might be misleading you, the win rate confidence calculator lets you enter your results and see the actual confidence interval around your observed edge. The width of that interval over small samples is usually sobering.
One way to partially compress the sample-size problem is to weight your results by average buy-in (ABI). Playing a mix of $10 and $500 tournaments means a single $500 score distorts a raw-count ROI calculation. An ABI-weighted approach gives a more accurate picture faster. The tournament results tracking guide covers how to apply this in practice.
Why ITM Percentage Is Not ROI (and Can Mislead You)
In-the-money percentage measures how often you finish in a paying spot. It feels like a proxy for ROI, but the relationship is weak and sometimes inverted in ways that matter.
Consider two hypothetical players across 500 identical tournaments:
- Player A: cashes 18% of the time, mostly minimum-cashes, no final tables. Net result: slight loser (rake erodes the tiny min-cash edges).
- Player B: cashes 11% of the time but has three final table appearances including one win. Net result: strong winner with 35% ROI.
Player A has a higher ITM% and is losing money. Player B cashes less often and is winning significantly. This happens because min-cashes often pay 1.5–2× the buy-in, while a tournament win might pay 100–200×. A strategy optimized to squeak into the money at every opportunity is often actively harmful to ROI because it sacrifices chip accumulation during stages when taking calculated risks would lead to deeper runs.
A high ITM% combined with flat or negative ROI is a diagnostic signal worth investigating, not a comfort. It usually means either overly tight bubble play, poor chip-lead aggression near the money, or simply running well at the min-cash level while missing on deeper runs. Tracking both numbers together gives you information that neither gives alone.
Practical Takeaways
- Do not draw conclusions from small samples. Under 200 tournaments, your ROI is dominated by variance. Under 500, it is still heavily influenced by it.
- Calibrate expectations to format. A 6% ROI in tough online SNGs is a real win. A 6% ROI in live casino MTTs with soft fields probably means you are underperforming.
- Track both ROI and ITM%. Divergence between them often points to a strategic leak worth investigating.
- Account for buy-in weighting. Mixing stakes without normalizing by ABI produces a ROI number that is hard to interpret or compare across time periods.
- Use confidence intervals, not point estimates. A single percentage is less useful than knowing the range your true ROI plausibly falls within given your current sample.
How PokerCharts Helps
The tournament tracker on PokerCharts automatically calculates your ROI, ITM%, and ABI across every session you log, and breaks those numbers down by format, buy-in level, and date range. Rather than maintaining a spreadsheet or trying to remember your last 300 results, you get an always-current view of how your tournament game is actually performing, with enough granularity to spot when your live results are diverging from your online numbers or when a particular stake is consistently dragging down your overall edge.
PokerCharts is free for your first 10 sessions, then $1.99 per month billed annually ($23.95 per year). If you are playing enough tournaments to care about what your ROI actually means, that level of tracking pays for itself quickly in the strategic clarity it provides.