MTT Analysis

How to read your GGPoker tournament results in BluffLog: why some tournaments have no ROI, what Top 12% means instead of ITM, how re-entries enter the cost, and why hand results are in chips.

Tournament results in BluffLog come from two kinds of files: the hand histories and the tournament summaries, both exported from PokerCraft. The hands give you every decision; the summaries give you the buy-in, the finish and the prize. Most of what looks odd on these screens comes from a summary that is missing. Every screenshot is from version 0.18.1; click one to see it full size.

Sessions: tournaments per sitting

1. Tournaments, Profit and ROI% per session. A session groups the tournaments you played in one sitting: the same rule as in cash, hands with no gap longer than 15 minutes. A session with seven tournaments and a -45% ROI is one evening, not seven results.
2. Display: bb or Chips. The switch changes only the unit of the hand results: the big blinds of that hand, or tournament chips. Sessions and tournaments are always in money.
3. Prizes, bounties and knockouts for every tournament. The middle table lists the tournaments of the selected session: buy-in, place, the prize for the place, the money for the players you knocked out, and the knockout count. ”≈” marks an estimate: a GG summary states the payout as one sum, and BluffLog splits it by the tournament’s payout ladder. A click on the knockout count opens the list of knockouts; more in Payouts and knockouts.
4. Hand results are in bb. A hand at -8.7 bb is a share of a stack, not money; what the tournament cost is on the row above.

Profit: the curve and the tournaments outside it

1. “No summary” is the count of tournaments whose buy-in BluffLog does not know, because their summary file was not imported. They are in the hand count and in every play stat, but not in the profit line and not in the ROI. If this number is not zero, export the tournament summaries from PokerCraft into the same folder and import again; nothing is duplicated. The same line splits what you received into prizes, bounties and the knockout count.
2. Profit or ROI on the same curve. Profit shows the money; ROI shows the same run as a percentage of what it cost, which is the only way to compare a $2.50 evening with a $25 one.
3. A vertical jump is one deep run. Tournament results are a step function: dozens of small losses and a few large wins. Judging a sample of sixty tournaments by its curve is judging noise; the curve is for finding the tournaments to look at, not for measuring your game.

Analytics, Overview: ROI, ABI and the ITM question

1. ROI is profit divided by cost, and cost includes re-entries. A tournament with a $2.50 buy-in and two re-entries cost $7.50. ABI is the average buy-in of paid tournaments only; freerolls are counted separately so they do not drag the average to zero.
2. Top 12% stands in for ITM. GG does not export the data needed to say exactly which finishes were paid, so BluffLog shows how often you finished in the top 12% of the field, which is roughly where the money starts in most of their tournaments.
3. The footnote says the same thing. Top% is the average finish percentile: 58.7% means that on average you outlasted 58.7% of the field. It is a better measure of tournament play than ITM alone, because it counts a 15th place out of 900 for what it is.

In bounty tournaments the same chart gets a money EV line: all-ins priced by the ICM of the stage and the opponents’ bounties. It has its own page, Money EV with bounties.

Analytics, Results: the whole picture

1. In the money is counted by money received, prize or bounty, whichever came. “Busts before ITM” is the rest, so a bounty tournament where you cashed only through knockouts still counts as a cash. “Paid into bounty pool” is what you put into the knockout pools; what knockouts paid you is on Profit and in the tournament card.
2. Top 10%, 5% and 1% are shares of the tournaments with a known field size, and the line above them says how many that is. A tournament whose summary does not name the number of entrants is left out of these shares, not counted as a miss.
3. Re-entries are the count that went into the cost above. Forty-three re-entries in eighty-five tournaments is a fact about your bankroll before it is a fact about your play.
4. The finish distribution has a “50+” bucket and a “Place unknown” row. In large fields almost every finish lands in 50+, which is why the distribution says less than the percentile above. “Place unknown” is, again, tournaments without a summary.
5. Stack depth performance is in big blinds. It answers a question no money figure can: at which stack size do you lose the tournament. The bucket that goes negative is where the hands for the replayer are. The note under the table is a reminder that chips know nothing about bounties: in PKO a call made for a knockout can be right in money and losing in chips.
6. By format keeps regular tournaments, flipouts, Spin & Gold and satellites apart, each with its own ABI, profit and ROI. Free tournaments stay out of ABI, so a run of freerolls does not turn a $3 average buy-in into $1.

What to do with it

First make “no summary” zero by importing the summaries, or every number on these pages is about a subset. Then read the percentile rather than the ROI: with fewer than a few hundred tournaments the ROI is mostly the size of your last deep run. Then take the stack depth table to the replayer and look at the hands from the bucket that loses.

→ See also: Cash Game Analysis, where the same screens count hands instead of tournaments.