zachvac
Legend
Silver Level
Correct, but there are people who lose ALL the time. But they're great players, they're just the unluckiest players in the world. Just ask them, they'll tell you (note this isn't a slight to the OP, just people in general who talk of how good they are and the only reason they're not rolling in dough is they're the unluckiest person on the planet, met more than a few of those in real life).
K_Kahne_Fan
Legend
Silver Level
I'm not talking about me with this question. For the 1st year or so I was a losing player, but that's because I was learning. As of about the last 6 months I've been holding steady at a dollar amount. Hopefully, with the help of CC, I will start to be a winning player soon.
I'm talking about in general.
dj11
Legend
Silver Level
For some uncontrollable reason I feel I must add to this.
-The most unlikely event in any random situation will be the statistically perfect event.
-At any 9 handed poker table, over time, you will get dealt the best hand 1/9th of the time (11.111111111111111~%). It is too easy to think that you will get cards once every 9 hands. Statistically statistics can't really even take effect until the sample numbers are much larger than 9 hands, or 90, or really even 900 samples. Though at 900 some trends can begin to appear.
-Against one opponent poker hand odds will stand up over time.
-Everyone tends to ignores or forgets the 'schooling' effect. If against a single opponent you are favored, or against any of many opponents you are favored, if you put all of those opponents together against you, you fall below a 50% expectancy. Think total outs against you.
This being said, OP's original thought about a consistent 51-49 'luck factor' edge are actually within statistical norms until you get close to an infinite number of comparisons.
I am reminded about a California man who has won the California lottery twice.
I have to consider that the OP has a correct thought. I also have to believe that all samples before now have been me being on the underside of the 51-49 split, and that any day now, statistics and nature will force me to the other side!
-The most unlikely event in any random situation will be the statistically perfect event.
-At any 9 handed poker table, over time, you will get dealt the best hand 1/9th of the time (11.111111111111111~%). It is too easy to think that you will get cards once every 9 hands. Statistically statistics can't really even take effect until the sample numbers are much larger than 9 hands, or 90, or really even 900 samples. Though at 900 some trends can begin to appear.
-Against one opponent poker hand odds will stand up over time.
-Everyone tends to ignores or forgets the 'schooling' effect. If against a single opponent you are favored, or against any of many opponents you are favored, if you put all of those opponents together against you, you fall below a 50% expectancy. Think total outs against you.
This being said, OP's original thought about a consistent 51-49 'luck factor' edge are actually within statistical norms until you get close to an infinite number of comparisons.
I am reminded about a California man who has won the California lottery twice.
I have to consider that the OP has a correct thought. I also have to believe that all samples before now have been me being on the underside of the 51-49 split, and that any day now, statistics and nature will force me to the other side!
zachvac
Legend
Silver Level
Good point dj until the end. While it's possible that you could be 49% in coin flips so far, your chances of winning the next coin flip is still 50%. No individual person can have the odds against them in coin flips (real ones, poker-related ones usually aren't 50-50, I'm talking about just a 50-50 random event). So even if up until now you're at 49%, the odds are still 50-50 for the next flip.
zachvac
Legend
Silver Level
all samples before now have been me being on the underside of the 51-49 split, and that any day now, statistics and nature will force me to the other side!
The
makes me think you understand this, but I thought I'd point this out to people who don't. If you are at 49% over the last hundred thousand hands, the odds are after an infinite amount of time your percentage will approach but never reach 50%. Because the future isn't impacted by running badly in the past, you would have to run well to cross to the other side. It's called regression to the mean.
dj11
Legend
Silver Level
Absolutely ZACH, however, I put it all in the 'Nature abhores a vacuum' category, and believe that in the end, things do even out. If I didn't believe that, I would have given up the game after my first losing session.
As it works out, I avoid coinflips when they seem apparent to me, and so when I do get in I think (can't prove it) that I am ahead in coinflips.
As it works out, I avoid coinflips when they seem apparent to me, and so when I do get in I think (can't prove it) that I am ahead in coinflips.
Effexor
Cardschat Elite
Silver Level
I have 55,211 hands in my database. Maybe the sample size isn't large enough, but it's got to be a good representation of expected results I'd think.
I've been dealt AA, 212 times. The expected value is 250.959 times.
Do your math and tell me if this is within a normal range statistically speaking.
I had a statistics class in Pharmacy school, but that was so long ago that I only have a basic understanding left.
J
joeydy123
Rising Star
Bronze Level
To get back to the original question, you are correct in that ALL players will either win more or less than their actual statistical expectation.
However, the larger the sample size, the more their actual winnings/losses will trend towards their actual statistical expectation.
For example, if we play a simple coin tossing game where we each bet $1 on heads or tails, after 100,000 tosses it is incredibly unlikely that we are going to be even – one of us will have earned more and the other will have lost more that we would have expected. In this case, if I win 49.5% of the time instead of the expected 50%, I will be down $500.
What is key though, as Zach has said, is that on the next toss I still have exactly a 50% change of winning.
However, if I find a fish that is willing to give me 1.5:1 odds on a coin toss and I play a sufficient amount of times I am essentially guaranteed to be ahead (but still unlikely to ever be exactly the amount ahead as I would expect)
Example 2:
I bet $10 into a pot of $10 with a pair of aces
Villain calls with a flush draw (he is all in, so this ends the betting on the hand)
My expectation for the bet is $14 (win $20 80% of the time and lose $10 20% of the time)
His expectation for the bet is -$4 (win $20 20% of the time and lose $10 80% of the time)
However, everytime we do this one of us will do better than expected and the other will do worse than expected. The more we play, the closer the results will get to the expected 80% / 20% but they are unlikely to ever be exactly right.
Sorry, got a bit carried away with this. To answer the question, yes some people do lose more than they should, but it will be insignificant over a sufficient number of hands with responsible bankroll management.
However, the larger the sample size, the more their actual winnings/losses will trend towards their actual statistical expectation.
For example, if we play a simple coin tossing game where we each bet $1 on heads or tails, after 100,000 tosses it is incredibly unlikely that we are going to be even – one of us will have earned more and the other will have lost more that we would have expected. In this case, if I win 49.5% of the time instead of the expected 50%, I will be down $500.
What is key though, as Zach has said, is that on the next toss I still have exactly a 50% change of winning.
However, if I find a fish that is willing to give me 1.5:1 odds on a coin toss and I play a sufficient amount of times I am essentially guaranteed to be ahead (but still unlikely to ever be exactly the amount ahead as I would expect)
Example 2:
I bet $10 into a pot of $10 with a pair of aces
Villain calls with a flush draw (he is all in, so this ends the betting on the hand)
My expectation for the bet is $14 (win $20 80% of the time and lose $10 20% of the time)
His expectation for the bet is -$4 (win $20 20% of the time and lose $10 80% of the time)
However, everytime we do this one of us will do better than expected and the other will do worse than expected. The more we play, the closer the results will get to the expected 80% / 20% but they are unlikely to ever be exactly right.
Sorry, got a bit carried away with this. To answer the question, yes some people do lose more than they should, but it will be insignificant over a sufficient number of hands with responsible bankroll management.
elfmanvt07
Enthusiast
Silver Level
Can I just say this? The original post essentially only really asks one thing. Does luck exist? Not you know, "oh you got lucky that time, Jimbo," kinda luck. I'm talking about, that is a LUCKY DUDE. So I guess there's really only one thing you gotta ask yourself.
Do you feel lucky, punk?
Well do ya?
Do you feel lucky, punk?
Well do ya?
K_Kahne_Fan
Legend
Silver Level
So the 49.5% players HAS to steal blinds, find the fish, and be a little more agressive than the 50.5 player (?).
As for what cards are next, I understand the logic that what is dealt next will have the same odds every time, I'm talking about a player's... I guess luck as y'all keep saying, over the long run. I know 15 blacks on roulette doesn't mean a red HAS to come next, I'm talking about the two players who were playing and 1 bet black those 15 times and the other bet red. Will the red hit, sure, will they ever recover that 15 black streak... maybe not. Will that roulette table keep hitting the percentage of black/red/blank sure... but that player won't be there to see the variance because they just lost in that session... and they may lose in the next... and the next... and... While the person betting black may win in the next... and the next... and....
Last edited:
zachvac
Legend
Silver Level
Nope, because there's no such thing as a 49.5 player. If you're running at 49.5 for the last 2 million hands (which is actually pretty much statistically impossible I think) you are still a 50% player. What if you sit down and then get up and someone else sits there. Is that new player 49.5% because you are 49.5% in the past? What if the 49.5 and the 50.5 player switch seats? Does that change anything? The answer is no, there is no way to know if you'll run at 49.5 or 50.5 until after a significant time. Stealing blinds and finding fish is before you get the results, and before you get the results you will win exactly 50% of coin flips. Even though as mentioned getting exact results are very unlikely (it's a mean, not a magic prediction), before a coin is flipped it is 50% to be heads and 50% to be tails. You can never play differently because for the hand you're adjusting for you are 50-50 to win if it's a coin flip. For the session you assume you'll have average luck. And after the session when you found out you ran like God or ran like crap, you can't go back and change how you played.
zachvac
Legend
Silver Level
This last example is a very good one and one people often ignore. When you have AA vs. 77 and you win, you got lucky. You were only supposed to win 80% of the pot but you won 100% of it. So if all your monsters hold up and you lose the majority of your coin flips, your luck could still be positive for the day because you have to factor in the times where your best hand that wasn't 100% to win won 100% of the pot.
What? A main principle of poker is that luck is short-term. If you think a "dude" can be inherently lucky or feel lucky than perhaps poker isn't the game for you. It's possible to have a lucky session, but if you're a losing player you won't be up after you've played a significant amount of hands.
zachvac
Legend
Silver Level
Here's the thing, you're dealing with an ex post facto calculation. I'll do it in a sec but let me explain the dangers of this. There are so many ways in which you can be unlucky in poker. You can miss sets more than usual, you can get dealt AA less than usual, your AA can hold up less than usual, etc. When you pick one thing to examine it's usually the one that's the furthest from the mean statistically meaning that you are selecting a small portion of the data because it is abnormal and treating it as if it is all the data.
And example would be you flip a coin 100 times. There are 5 heads in a row. You say "wow, the odds of 5 heads in a row coming are 0.5^5 = 3.1%. Never mind that it's very likely for that to happen in the entire sample, but you just picked the most out of the ordinary data and isolated it. If after flip #43 you say "let's see how many heads we get in the next 5 flips" and then they're all heads, that's something somewhat unlikely, but when there is much more data and you pick one stat to use, and the reason you pick the stat is because it's probably low, you're not using sound scientific or statistical reasoning. But say going in you had said you were going to examine the number of AA vs. expected. Using a 1-proportion Z test (google it, we learned how to do the calculation in HS stat and I could look it up but my calculator has a function for this so I'll let it do all the work), we get that getting your data or worse (further away from the mean, as in less AA's) is 1.65%, pretty unlikely (just for reference the political polls use 95% confidence interval, so anything below 2.5% is outside their range including the + or -.
Yes you've been unlucky with AA. It's not unbelievably unlucky but you are in the bottom 2 percentiles of the poker population in terms of how often you get AA. Also along with the ex post facto comment is selection of people in poker as well. If something is 0.001% = 0.00001 to happen, we would usually dismiss it as close enough to impossible. However, when we have hundreds of thousands of people playing poker, and they are playing thousands and thousands of hands, all of a sudden the odds of one person having that happen to them is actually pretty likely.
Sort of like the lottery. Bad enough odds that I can say that statistically it's pretty much impossible for me to win, but some people win, right? When you have that many people who play, the odds of ONE of them goes up. Your personal odds still stay the same, and each individual person can more or less say that they won't ever win the lottery, but one of them will be shocked and will be one of the luckiest people in the world and hit their number. By the way, runner runner to 2 outs (as in they have to hit exactly 2 cards in the last 2 to win) is 1 in 1081. The odds of winning the lottery are 1 in 80,089,128 for the powerball (http://members.cox.net/mathmistakes/rawdata.htm). Catching back to back runner runner draws (to exactly 2 cards, example is opponent has nut flush and you have a straight flush draw on the turn to exactly 2 cards. example opponent has AJs you have 34s flop comes 789 all the same suit as both your hands. The odds of winning this back to back is
1 in 1081^2 = 1 in 1,168,561. So you are 80 times worse to win the lottery than you are to win back to back runner runner to exactly 2 cards draws.If this person got this kind of luck in poker on an opponent, they would certainly cry how poker is rigged, how they're nut flush got runner runner'd to a straight flush twice from the same guy and then the very next hand rivered a 1-outter on him (that's about the 1 in 80, lottery is still less likely than this because 1-outter is better than 1 in 80). But when someone wins the lottery it's always "well someone has to win it". So next time you take a horrible beat, think of it that way. The odds are such that it will happen, and someone has to take that beat. Unfortunately it was you, but that's the way it goes.
K_Kahne_Fan
Legend
Silver Level
What if you bet 15 reds, they all came black and you leave, could the next person betting red hit the next 15 in a row... sure. Will they, not sure, but they could. You could be dealt strong hands that keep getting cracked, get up, next players is dealt strong hands that hold up.
Since the person betting black hit 15 and you lost 15, now you switch, could the person betting black now hit 15 reds and you lose 15 again... sure.
I know probability and statistics and that what has happened has no effect on what will happen. That's why my thread title isn't, how will a cold player run in the future, it's "can some people just run cold (period)."
I guess it's like the saying "it's in the cards", maybe some people just won't get dealt winning cards in the long run and some will. Why can't this be the case?
Basically you're saying if we all played equal we would all be winning players... nope. It's the cards we're dealt. And some people may just be dealt worse cards (by that I mean board as well) than others.
zachvac
Legend
Silver Level
ok here's the thing. If you flip coins, some people will end up up and some will end up down in the long run. Not by a lot, but that will happen. If you have a 51% edge, you will not end up down in the long run, ever (make sure it's a long enough long run, but as number of hands approaches infinity, so do your winnings with a 51% edge
).I'm going to do an experiment actually. I'm going to write a quick simulation program that will flip a 50.5%-49.5% coin (numbers you suggested) 100,000 times. I will do this a couple thousand times and see how many out of those thousands the coin is losing over the run of 100,000 flips (think of them as hands). I'll let you know the results I get.
zachvac
Legend
Silver Level
Just ran the simulation, out of 10,000 people who play 100k hands each, 8 of them lost over that time. The 8 who did?
49914
49993
49983
49974
49992
49988
49966
49981
note that 50000 and over is winning long-term. So even with such a miniscule edge of 50.5%, over that many hands only 8 out of 10,000 lose, and most of those losses were minimal and close enough to breaking even.
49914
49993
49983
49974
49992
49988
49966
49981
note that 50000 and over is winning long-term. So even with such a miniscule edge of 50.5%, over that many hands only 8 out of 10,000 lose, and most of those losses were minimal and close enough to breaking even.
Effexor
Cardschat Elite
Silver Level
The only reason I picked the AA is that it's easy to just look in my database and add up the numbers. Trying to go through every hand to see how many flush draws I hit vs how many I missed would be far too laborious. There is no psychological reason that I picked that hand other than it's the one hand that has the highest EV, and can represent my "luck".
I guess my point was about the long run. Where I do believe that over the long run the statistics will hold up, if I am in the bottom 2% for catching pocket aces over 55,211 hands, I'm not sure that I'll be alive long enough for the statistics to catch up. Then if you compare my ROI against someone else of equal skill, but has actually had AA 250 times (the expected value) their ROI would be significantly higher. Play 39 straight hands at a full ring game where you get pocket aces every hand and see how much you are up.
If I am in a coinflip situation, and lose 32 out of 37 times (which happened btw), yes I know the next time I'm still 50% to win, but how long would it take to even out?
All things considered though with "luck" is that I firmly believe that skill can overcome this. I deposited $100 2 years ago, and am still up despite running worse than whats expected. There are so many terrible players out there that it is possible to win despite getting outdrawn and getting bad cards. I'm proof of that and I'm far from being great at poker.
I guess my point was about the long run. Where I do believe that over the long run the statistics will hold up, if I am in the bottom 2% for catching pocket aces over 55,211 hands, I'm not sure that I'll be alive long enough for the statistics to catch up. Then if you compare my ROI against someone else of equal skill, but has actually had AA 250 times (the expected value) their ROI would be significantly higher. Play 39 straight hands at a full ring game where you get pocket aces every hand and see how much you are up.
If I am in a coinflip situation, and lose 32 out of 37 times (which happened btw), yes I know the next time I'm still 50% to win, but how long would it take to even out?
All things considered though with "luck" is that I firmly believe that skill can overcome this. I deposited $100 2 years ago, and am still up despite running worse than whats expected. There are so many terrible players out there that it is possible to win despite getting outdrawn and getting bad cards. I'm proof of that and I'm far from being great at poker.
zachvac
Legend
Silver Level
First off the 32 out of 37 wouldn't even out. Expected is 18.5 and you're at losing 14 more than you should. Statistically after millions of coin flips, you should still be about 14 more losses than wins. The percentage gets closer to 50%, but overall the net should stay at -14.
But poker is not about winning coin flips. If you rely on coin flips the rake kills you anyway. We want to be BETTER than our opponents, and that's where the difference comes in. If we have that 55-45 edge there is a great chance that over a good amount of hands we'll be over 50%. We may not be over 55%, which is our expected, but with even bad luck we can still be above 50% and be a winning player. Speaking of which, just ran a simulation of a 55-45 player who again plays 100,000 hands. I ran it 100,000 times (ate lunch while it was running, it did take a while) and exactly ZERO of these 100,000 players were down at the end of the 100,000 hands flipping 55-45 coins in their favor. I may do a similar one overnight where instead of 100,000 people I do 1 million or 10 million even, just to see if I can get at least one person who loses over 100k hands with a 55-45 edge.
elfmanvt07
Enthusiast
Silver Level
It was....
*drum roll*
A joke. I just wanted to type something that let me quote Clint Eastwood.:ridinghor
Q
quads
Rock Star
Silver Level
I have received an infraction from this forum for an insult to a fellow member in this thread.
I would like to at this time apologize for my behavior and for insulting a fellow member just because I don't agree with him. "I'm Sorry"
I would like to at this time apologize for my behavior and for insulting a fellow member just because I don't agree with him. "I'm Sorry"
Effexor
Cardschat Elite
Silver Level
In a cash game this is absolutely true. The flips I was speaking of were all in SNGs and MTTs. You know, your typical low M push with AQ or 66 type of thing.
The OP's question / statement was do some people running cold in general, and the real world answer to that I think is yes. Statistics and math do hold up over time and sample sizes but it is very possible to run bad and literally never ever see enough hands to "catch up" to where you should be statistically speaking.
I couldn't agree more with your above statement. What I want to add is that being better than your opponent is so much more than the luck of the cards and always making the right bets or calls with pot odds. (At least in NL, Limit may be a different story) In NL, making good reads, playing position etc etc. basically outplaying the villians can overcome cold cards to a certain degree.
K_Kahne_Fan
Legend
Silver Level
Quoted from This is why pokerstars is terrible...
So, when it comes to slot machines, no one is lucky because the machine knows no different? Look at poker/HE as a slot machine, 1 player sits for 4 hours and loses every hand, gets up, next player sits and hits the jackpot (royal flush) and cleans up the table because every1 else had boats, str8's, flushes, but they had the royal. It has been said "so if player 1 gets up and player 2 sits down, player 2 will get better cards?" Well, if I get up from a slot and someone else sits down... they could start winning, or heck just breaking even or losing less in that session on the same slot I just lost my paycheck on. BTW, I'm not talking about the skill aspect at all in this, strictly dealt cards. Some may argue slots are on a RNG and every milisecond makes a difference, but when you get up from a table more than likely the next person to sit down will not get the very next hand you were supposed to be dealt. Live, they probably have to get their chips in order, get their seat comfy, maybe order a drink... OL, the hands are dealt so fast that after you've lost your hand and are leaving the next hand is dealt before the second player sits. Both live & OL would be just like a RNG on a slot and completely change the dynamics of the hands being dealt before the next player gets their cards.
LittleSip
Enthusiast
Silver Level
Running Cold
Well for me I have run cold on this one site for a couple weeks. Some days I get no hands that win. Others one or two that win. I can say this is often thou I do have my good and great days.
I haven't been able to play many of the Cardchat freerolls, but thanks for hosting them. And thanks for a great poker forum.
Guess I'll hit a hot streak one month and get to play most of them.
Now when was that month vacation??? :icon_sant ??????
Well for me I have run cold on this one site for a couple weeks. Some days I get no hands that win. Others one or two that win. I can say this is often thou I do have my good and great days.
I haven't been able to play many of the Cardchat freerolls, but thanks for hosting them. And thanks for a great poker forum. Guess I'll hit a hot streak one month and get to play most of them.
Now when was that month vacation??? :icon_sant ??????
zachvac
Legend
Silver Level
Sorry I just don't get your point. In poker, slots, flipping coins, rolling dice, using a RNG on a calculator, or any other method of achieving something random, there's no "supposed to get". At any point before the generation of the event, the event has an exact probability distribution that doesn't change based on who's sitting there, how much time has passed, etc. That's the point, players getting up and moving make no difference whatsoever. You're making the mistake of being completely results-oriented and thinking of an RNG not as a random number generator but simply examining the short-term numbers it produces. Think of it as a black box and don't worry about how the numbers are produced. At any point in time, the probability of your aces holding up is ~80%. This has no bearing on who's sitting down, when you last ordered drinks, what astrological sign you are, whether you tipped the dealer last hand, etc. You just fail to grasp the notion of random, thinking of an RNG as the numbers it spits out rather than the probability distribution and generator of random numbers that it is. A random number has no memory and is COMPLETELY INDEPENDENT of any circumstances in the present or the past.