TL;DR: I did it. 20 wins out of 61 Astral Pegasus 9-max games in a week for a perfect score on the “Best 20” leaderboard (Week 173). What’s my prize @SunPowerGuru ? 
And here is the long version:
Background: A couple of months ago, I analyzed all of my SNG results. I found that I had won about 20% of my 9-max SNGs. The sample size was 427 games. The vast majority of those were high-stakes games, but the sample also included a good number of medium-stakes and very few low-stakes and elite-stakes games. Thus, over 100 games, I could expect to win 20 games, but with a standard deviation of 4, so it could well take 120+ games if I ran poorly – a daunting task.
On the plus side, there are two reasons why I believe that my current win rate might be closer to 25% though:
- I improved my game a bit in the last couple of months.
- The Astral Pegasus games have become significantly softer over the past year.
However, even with a win rate of 25%, the expected number of games necessary would still be 80, and it could take 100+ games if I ran badly.
With these considerations in mind, I decided to give it a serious try this week if the first 20 games yielded a significantly higher win rate than 25%. I didn’t quantify how many wins I wanted to have, but I was probably looking for at least eight wins in the first 20 games, so that the expected number of games to reach 20 wins would drop to 20 + 12/.25 = 68, which seemed more manageable.
Sunday (22 games, 9 wins, 2 second places): I started playing on Sunday, June 27. I got off to a good start winning my first two games, but then didn’t win any of the next seven games, so I was almost back down to a 20% win rate. Then I won Games 10 through 15 for an incredible six game win streak and also won Game 19. In total, I won 9 games out of my first 20 games – an insanely good start. So with an assumed win rate of 25%, the expected number of games necessary to reach 20 wins dropped to 20 + 11/.25 = 64.
Monday (30 games, 8 wins, 5 second places): Unsurprisingly, my win rate dropped on Monday. I played 30 games on Monday and won 8, which is actually very close to the estimated 25% win rate.
Tuesday (9 games, 3 wins, 1 second place): Overall, the final few games went well with 3 wins out of 9 games. There was a real chance to reach the goal in 58 games if I had won the game in which I went into the heads-up with a significant chips advantage but lost.
Playing style: For the most part, I was playing my usual game. I played slightly more speculative hands in the early game and also in the mid game if I hadn’t built a stack yet. I was also taking slightly more, reasonable all-in risks early on; this would put me in a good position to play for the win if I doubled up, and not much time was wasted if I busted out early. I obviously didn’t care about the bubble. I was also playing up to three games simultaneously to keep the time commitment manageable.
Statistics: Here is the distribution of finishing positions over all 61 games:
| Position |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
| # |
20 |
8 |
6 |
7 |
5 |
4 |
3 |
3 |
5 |
| % |
33% |
13% |
10% |
11% |
8% |
7% |
5% |
5% |
8% |
Odds and ends:
-
Longest streak without a win: 7 games (Games 3-9)
-
I am currently leading the Astral Pegasus “First 20” leaderboard with about 259k points, one of the highest scores ever (perhaps even the highest?) on this leaderboard.
-
I am currently leading the June 2021 monthly SNG high leaderboard with 714k points, the highest score ever on this leaderboard.
-
I paid 6,100,000 chips in buy-ins and won 11,628,000 in prizes for a profit of 5,528,000 chips and an ROI of 91%.
-
Never have I gotten so frustrated by second place finishes. Every heads-up battle that I lost meant that I had to play another 4-5 games (on average) to reach the goal.
-
No more SNGs for a while.
