Confidence at an all time low so playing badly and clueless.

[Honeybadger]

HONEYBADGER i know i said this early but after reading your last post must say it again, you REALLY do know how to get a point across. Without appearing to p1ss up your back and telling you its raining ,   iREALLY think, not just Jason but MOST people who read this forun should be glad you are posting on here , well done brilliant stuff.

Thanks pal, always nice to be appreciated

[robyong]

People make good points here but I feel the live long term is way shorter than the online stats players claim;

Example: if Keith Johnson played 2-5 nl every week in 2013 against a standard DTD Friday night line up, he will not lose over 52 sessions. Fact. Estimated hands played would be 25 per hour x 6 hours x 52 weeks = 7800 hands.

What is the theoretical long term for Keith in this game then? I wud actually say he would almost never lose over 3 months in reality.

What would be the long term online at 2-5 nl?

[JK]

I'm not sure if this is true, or the person I heard it from made it up (can't remember who or when I heard it from), but similar to hobeybadgers point.

A poker theorist ran a 4 handed holdem flip between 4 random computerised players. He ran millions of flips,expecting the players win/loss lines to rise and fall, crossing in the middle as variance would suggest. Supposedly, 2 players won the whole way and 2 lost, apparently disproving variance over x million sample size of completely random flips.

As I say, could be BS as I have spent a lot of time in gala, but its a good story nonetheless lol

[Honeybadger]

http://www.nsdpoker.com/2011/01/mtt-pros/

This is the best explanation I've read with regards to variance.

A really good article, and it is nice to see some sort of data/simul-based proof of what I have always intuitively believed.

In particular, this part of the conclusion should give everyone pause for thought:

"Anyone except a true tourney god is more likely to lose over 100 tourneys than to win, which also sucks, and even a great tourney player has a decent chance of losing over 1k+ tournaments"

[titaniumbean]

The definition of long term gets longer and longer as the winrates reduce. Bigger winrates lead to lower variance so ldo variance gets more severe.

http://www.evplusplus.com/poker_tools/variance_simulator/


Edit Rob, even with high winrates, 3 months of live poker results can be utterly inconsequential.

[Honeybadger]

I'm not sure if this is true, or the person I heard it from made it up (can't remember who or when I heard it from), but similar to hobeybadgers point.

A poker theorist ran a 4 handed holdem flip between 4 random computerised players. He ran millions of flips,expecting the players win/loss lines to rise and fall, crossing in the middle as variance would suggest. Supposedly, 2 players won the whole way and 2 lost, apparently disproving variance over x million sample size of completely random flips.

As I say, could be BS as I have spent a lot of time in gala, but its a good story nonetheless lol

I read a few years back about an experiment in which several identical poker bots were set to play each other. They all had identical skill levels and were running on identical processors. And they played something like 20m hands - i.e. more hands than any human could ever play lifetime. At the end of this period some of the bots were 'winning' millions and others were losing millions.

The experiment proved that a lifetime of poker play is NOWHERE NEAR long enough for variance to even out.

Or you could just point to Jamie Gold - no matter what happens to him in poker in the future, he will end his life having run above EV.

[Woodsey]

Think I'll just stick to enjoying playing poker, rather than thinking too hard about all this shit 

[Amatay]

Think I'll just stick to enjoying playing poker, rather than thinking too hard about all this shit 

Finally a decent post ITT

[AndrewT]

I'm not sure if this is true, or the person I heard it from made it up (can't remember who or when I heard it from), but similar to hobeybadgers point.

A poker theorist ran a 4 handed holdem flip between 4 random computerised players. He ran millions of flips,expecting the players win/loss lines to rise and fall, crossing in the middle as variance would suggest. Supposedly, 2 players won the whole way and 2 lost, apparently disproving variance over x million sample size of completely random flips.

As I say, could be BS as I have spent a lot of time in gala, but its a good story nonetheless lol

This is not what the maths says should happen - if two of the players go up early on, you would expect them to hold on to that lead more often than them cross over to become losers.

[CHIPPYMAN]

C Jason , every single person that read blonde poker love u to bits .
U r more famous than Prince William & Kate Middleton combine .

[skolsuper]

People make good points here but I feel the live long term is way shorter than the online stats players claim;

Example: if Keith Johnson played 2-5 nl every week in 2013 against a standard DTD Friday night line up, he will not lose over 52 sessions. Fact. Estimated hands played would be 25 per hour x 6 hours x 52 weeks = 7800 hands.

What is the theoretical long term for Keith in this game then? I wud actually say he would almost never lose over 3 months in reality.

What would be the long term online at 2-5 nl?

That's selection bias again. We're talking about going the other way, i.e. saying someone that won for 3-12 months in that game is probably as good as Keith Johnson. Hopefully you can see that that isn't true.

[Honeybadger]

People make good points here but I feel the live long term is way shorter than the online stats players claim;

Example: if Keith Johnson played 2-5 nl every week in 2013 against a standard DTD Friday night line up, he will not lose over 52 sessions. Fact. Estimated hands played would be 25 per hour x 6 hours x 52 weeks = 7800 hands.

What is the theoretical long term for Keith in this game then? I wud actually say he would almost never lose over 3 months in reality.

What would be the long term online at 2-5 nl?

Few points to make on this Rob:

First, you are talking about live cash games. Variance evens out in cash games much more quickly than in tournaments. The points I, and others, have been addressing re variance have mainly been in relation to tournament play. I very clearly said to Jason in one of my earlier posts that if he wanted to be a professional live poker player he HAD to do so by playing live cash games, not live tournaments.

Second, as Andrew pointed out - the bigger your edge the quicker variance will even out. Keith likely has a very big edge in the DTD cash games (he is a great player). Much bigger than even the very best tournament players can achieve on the live tournament circuit. This is another reason why Keith's 'long-term' will arrive much faster.

Third, I agree it is pretty unlikely that Keith will have a losing year in the DTD cash games. But it is just wrong to say "he will not lose over 52 sessions. Fact.". It is POSSIBLE that this will happen, fairly unlikely but entirely possible. If you ran Keith's next year of poker one million times (i.e. a million parallel universes) then in quite a few of those parallel universes Keith WILL have a losing year.  And tbh, we do not need anywhere remotely close to a million parallel universes for this to happen. It's only 52 six hour sessions after all.

Fourth, there is a MUCH greater chance of this happening if you are talking about only 3 months - that's only 12 to 14 six hour sessions. I have personally had some sick stretches of negative variance when playing in some insanely soft games (in which my edge would have been much greater than Keith's edge in a 2/5 DTD game, no matter how great a player he is). When I first started playing in the most profitable game I have ever played in, I was LOSING MONEY after my first 24 sessions in the game. It didn't bother me too much because I knew I was playing well, and I also knew that I was a 'big winner' in the game even though I had not yet won any money.

[stato_1]

Are we to accept that Rastafish, Wadey and Alex are on the right side of variance and Key, Keith and Mitch are on the other side.

Yes, thats the whole point.

[Honeybadger]

Are we to accept that Rastafish, Wadey and Alex are on the right side of variance and Key, Keith and Mitch are on the other side.

Yes, thats the whole point.

Yep. Furthermore one of those three 'right side of variance guys' realises this and, in fact, has stated this several times in his diary. And guess which one of those three is actually a really good player who understands poker...?

[Tal]

We had a lovely afternoon chatting about maths and then someone made it personal

Play nicely