really impressive results. You've said you play mostly 4card with a bit of 5c sprinkled in, but how do you keep your variance so low? Seems like the biggest downswing I've seen in the graphs are 12-15 buyins. Seems extremely low, even with a great winrate for PLO. Do you play exclusively 6max? Very curious what your global poker screen name is. I feel like I know most of the midstakes+ player pool

I have savings from my career before poker so I don't have a set bankroll.
I do think about withdrawing once I have about 30 buyins of the highest stake I play on any site.

If you zoom into the 2022 graph, there is close to a 40 buyin downswing around the 250k hand mark, but it doesn't seem as large once you zoom out to the full year.
I play just 6-max but occasionally will start tables HU if it's not against a good reg. I do table select/bumhunt pretty aggressively, which helps me maintain a pretty high win rate (~10bb/100 for most of my poker pro career) and keep the variance low.

I don't feel like sharing my screen names but I play very low volume on Global. My favorite part of playing online is exploiting players with HUD and not having one on Global makes it less fun and also more challenging to multi-table effectively. I also time out preflop on Global way too often which tilts me a lot more than running bad.

I appreciate the detailed reply! I didn't notice the 40bi downswing on the 2022 graph until you pointed it out. It definitely wasn't noticeable without a closer look at the year graph with a large sample size. I briefly skimmed all the graphs and was mostly focused on the 2023 graphs when I wrote my initial post.

I have a similar approach as you in terms of where volume goes (6m vs HU) and the overall table selection aspect. Completely understand not wanting to share screen names. I also agree that a basic HUD is extremely helpful while multitabling across multiple stakes/sites where you don't have as in depth reads as if you were to play 1 site with the same players on a daily basis. Global's software is brutal for playing a lot of tables. Have also timed out many times pre or just overall made poor decisions due to the fast software with eventual 0 timebank.

After having a few sessions of running bad recently, I decided to look deeply into how unlucky I have been in my career.

In my entire database, I have just over 1 million hands played in PLO (including before my pro career) and I am running 240 buy-ins below EV.
This sounded insane to me and I would have guessed that this is easily in the bottom 1%, perhaps even bottom 0.1%, in terms of "luck".

With primedope, however, running 240 buy-ins below my "true" winrate over just over 1 million hands is only in the bottom 3% of running bad, with a standard deviation of 125bb/100.
I understand that the standard deviation of the difference between EV and money won is lower than that, which probably pushes my luck into the bottom 1%, but this exercise just illustrates the absurd variance in poker.

This did make me feel better that I am probably not the worst running poker player in the world and I just have to accept the nature of the game that I am playing.

Btw before someone says that running below EV is evidence of collusion, I don't believe in that.

EV is only calculated when someone is all-in before the river. The only scenarios I can think of colluding players making different decisions (from non-colluding players) in all-in spots are with marginal hands.

For example, if they have a weak 2p and face a jam, they can call if they know stronger made hands are blocked and that makes it more likely you are on a draw. That will actually make you run above your EV because your outs are not folded by the colluding players.

I can't really think of scenarios of the opposite and I doubt there will be any significant impact if at all.

I ended up doing the math and running 240 buy-ins below EV over just over 1 million hands is only in the 0.6th percentile in terms of luck, so it's probably not the worst run anyone has seen.
I did also think of some situations where colluders can run above EV, such as calling marginal draws against all-ins knowing that their outs were not folded, but I still believe that has a minor impact.

July 2023 Update:



The games were very good, so I was able to put in a lot of volume even after missing 10 days of play due to vacation.
Unfortunately, I was running extremely cold, which prompted my brief rant above.

I did do quite a bit of studying on 5c in the month, but will probably focus my study back on 4c for the rest of the year as that is still the majority of my volume.

What does your study process look like?

Mostly marking hands and going through them in trainer and monker to see how the spot should be played.