Thanks to all who showed up for the March Madness tourney. To recap, the format was dual staged. First a round robin stage where all players played each other once. Second an eight person bracket where a best of three match determines who advances on to the next round of the bracket.
The objectives of the tournament were:
* Allow for plenty of play for everyone. The round robin stage was put in to make sure that everyone got to play a minimum of roughly two hours of poker. Regardless of heads up performance, no one was eliminated early.
* Correlate payouts with the best performing players. Players who did the best would not only collect game money from the round robin stage but would advance furthest in the brackets to collect the placement payouts.
* Balance the duration of the tournament with deep enough stacks to avoid crapshoot endings.
Overall, I thought the tourney worked out better than I thought. I was concerned about the format since I knew it was complex and unconventional. Elevating the blinds once per orbit is something that Sun and I had come up with for playing during poker breaks at the office and I thought it was appropriate for keeping all the games moving though I knew it certainly had the potential to be confusing. Marsh felt that the blinds were too oppressive at the end of the championship matchup and his opinion is good enough for me. After reviewing the blinds I would agree that the capped blinds could have been maybe 1600/3200 instead of 2000/4000 which would have been more inline with the first round. A rule of thumb for estimating the end of a tourney is when the BB equals 10% of the chips in play. Even though that estimating method is for full table tourneys it was the most applicable for heads up. I also thought that all the extra hands that were available in the later rounds would have gotten all the money in before the blinds maxed out but that wasn't the case.
I also don't think that everyone understood the impact of the game money. That was borrowed from Jeh's fantasy football league to prevent a player who might be mathematically eliminated from getting into the bracket from tanking their final games. The side effect of that was that there was less money for the placement payouts. For a $40 buy in, first place paid out $144 on top of any game money. Second place paid $72 and 3rd/4th places paid $36. I think part of the issue was that I was allowing for a few more players which would have boosted the placement payouts. As is, the first place winner won roughly 3.5x their buy-in which is less than a typical first place payout in a conventional format tourney. I think for the field we ended up with we could have gone with paying out just 1st and 2nd.
So NEXT time, I think what I would change would be:
* Adjust the blind caps down to avoid people feeling pot committed at the end out the blind levels.
* Adjust payouts so that the highest placements receive a proportion of the prize pool which are more similar to a conventional tourney. Maybe even have a stepped payout system depending on how many participants enter.
* Double check deck setups beforehand. How embarrassing! Clearly there was a breach in standard protocol for retiring setups since any setups which are short should have been pulled out of circulation. My bad.
Any other feedback for adjustments to make are welcome. Thanks.