I have seen talk about this a lot lately, i have also been there going for a leaderboard back in the day. spending a lot of time playing one or two sngs at a time working on getting the best average roi is nice and all but say you are getting double the roi of the average grinder at your particular game.......
say you get 10 %roi in $10 games when you play two at a time and they get
5% roi playing 12 games. they are making 3x the amount of money that you are in that particular game even though they have a worse roi.
if you are playing poker for money you should always be looking at $/HR and not your roi percentage.
I'd really like to do some analysis on this sometime.
I think plots of ROI and total profit against number of games played across a selection of poker players would be very useful. It would have to be controlled by the type of table - it's easier to multi DoNs and MTTs than ST turbos and HUs.
Anyone's max ROI is gonna be at 1 table but what's the optimum number of tables to maximise profit per hour? I can keep an ROI of 28% 3-tabling 6-seater turbos. To move to four I would need 21% just break even. My attempts at playing four haven't been very successful but to really find out with confidence you'd have to play thousands of tournies at each number. My gut feeling is I would probably max my earnings at about 4-5 tables. Cardinal23 plays 6-9 with an ROI of 20% which is just awesome. But these are regular 9-seaters so they take twice as long as the 6-seater turbos I play.
Also it's important that poker is fun as I'm a recreational player. With 3-tables I feel I can keep track of what my opponents are up to. 5/6-tabling I would simply become a human pokerbot - which just doesn't appeal to me. However, I would 5-table if I were a pro or simply playing for the cash. Which I'm not.
The only way to answer this one definitively is to pool data from lots of players. Also unfortunately it's very hard data to collect. There's no way of searching for how many tables played at once on sharkscope.
^Also, it would be player dependent. You could pool data for various players but if you have a player who starts to leak hugely at 4 tables+ then that'll skew the data and wont be representative for another player who can get up to 6 tables comfortably.
But i'd agree with landsbanki on the recreational thing. At the end of the day i'm still a recreational player and 3-tabling the turbos allows me to relax as well as focus on how i'm playing.
Another thing to consider is variance. If you maintain a higher ROI at a low $/hour then downswings will affect you less - you'll suffer them less often and will lose less with a higher ROI. However, increasing table ratio allows you to get in more volume which actually lowers the overall variance of the experience - even though short term swings would be bigger. Lol, i think that's right anyway. =)
"Another thing to consider is variance. If you maintain a higher ROI at a low $/hour then downswings will affect you less - you'll suffer them less often and will lose less with a higher ROI. However, increasing table ratio allows you to get in more volume which actually lowers the overall variance of the experience - even though short term swings would be bigger. Lol, i think that's right anyway. =)"
Very good point Alex.
Lower ROI = slightly lower variance BUT longer and deeper downswings. Multitabling has a much bigger effect on ROI than variance and it's the ROI effect than drives the longer, deeper downswings not variance which is often incorrectly blamed.
If you do the maths it is often better for a pro who has a minimum monthly amount to win i.e. to cover essential bills/keep roof over their head etc to choose to sacrifice some $/hr to lower the chances of not hitting their minimum.
Here's an example:
Let say you can get an ROI of 8.5% 6-tabling $50 9-seaters and an ROI of 16% 3-tabling them.
Profit per hour says 6-table.
Let consider someone who has time in a month to either play 1000 6-tabling or 500 3-tabling. Let's also assume they must make $1000 minimum to pay the rent etc.
The 95% confidence range for 6-tabling 1000 tournies is $17600 to -$200
The 95% confidence range for 3-tabling 500 tournies is $13460 to $1100
3-tabling should get the rent paid but 6-tabling could make you homeless.
However, if you've no minimum and are only interested in profit then maximising profit/hour is the way ahead.
^EPIC. The little statistics nerd inside me just gave you major props, lol.
Hi all,
I certainly agree that some players can turn a much bigge rprofit while multi tabling, but how much extra profit will definately differ between players. I personally haven't really got the hang of multi tabling, i find myself panicking when lots of decisions start appearing at once and not reading the individual situations carefully enough to really find a decent edge as soon as I multi table. my ROI is 27% I think but im sure if i was multi tabling it would be below 10%. Some can multi table loads though, so I guess its just getting used to it but i dont like the thought of losing money while im still getting the hang of multi tabling lol.
And as landsbanki mentioned, it's important for most people that they enjoy playing the game rather than just make optimum moves one after another like a computer programme.
Landsbanki wrote:
"
Lower ROI = slightly lower variance BUT longer and deeper downswings. Multitabling has a much bigger effect on ROI than variance and it's the ROI effect than drives the longer, deeper downswings not variance which is often incorrectly blamed.
"
I´m not really sure what that means. First u say variance is lower but downswings are longer and deeper which is obv a contradiction. the bigger ur swings the higher is the variance.
Second, variance by definition has 0 (zero) effect on ur roi in the long term. Variance is not to blame for ur downswings? Again by definition it is, or u can say swings are to blame for ur variance since variance is a measure for the size of the swings u have to expect.
I dont understand what roi effect means. I agree that multitabling has an effect on ur roi, but it´s still the variance which "drives ur downswings". I guess multitabling influences variance too, but a large number of games will size the swings down in relation to ur overall volume and so reduce ur variance (not 100% sure about that last point but i think alex said the same).
Landsbanki wrote:
"
Here's an example:
Let say you can get an ROI of 8.5% 6-tabling $50 9-seaters and an ROI of 16% 3-tabling them.
Profit per hour says 6-table.
Let consider someone who has time in a month to either play 1000 6-tabling or 500 3-tabling. Let's also assume they must make $1000 minimum to pay the rent etc.
The 95% confidence range for 6-tabling 1000 tournies is $17600 to -$200
The 95% confidence range for 3-tabling 500 tournies is $13460 to $1100
3-tabling should get the rent paid but 6-tabling could make you homeless.
"
(sorry have to figure out how to quote correctly)
I strongly disagree with Landsbanki here.
First i don´t think ur assuming even nearly correct numbers. someone who achieves a 16% roi at three tables wont lose nearly half of his winrate by sixtabling for some reasons:
1. If u really play for the money (like the pro in the example does) u easily can compensate ur lack of attention for a single table by using a pokertracker. The investion pays off for anyone who multitables.
2. If u start six sng´s at a time u´ll get to the bubble by average let´s say 3 times. Most of the decisions u have to make until then are no-brainers, u don´t have to pay attention to fold ur rags and raise ur goods until then. Even postflop ur decisions are pretty easy in that stage since u play premium hands or small pots with speculative hands.
Now if u threetable u have to cope with three simultaneous bubbles pretty often anyway, and that is where ur winrate is made. so i think there might be a slight difference but in no case it is a 50% decrease of ur winrate. Most push/fold decisions are obvious, u dont have to think everytime u have Q9s on the button with 8BB, do you?
3. If u lose say 15 sng´s in a row- will u tend to tilt as much sixtabling as if u were three-tabling?
i´m kind of a stats nerd also but no major prop from me. The point with the 95% confidence rate isolated is by no means a good one.
1. The pro in ur example will by average experience a 200$ loss sixtabling every 20 months. If he three tabled, he would have 1300$ more in the same bad month. But every 20 months he sixtables for 17.6k$ instead of 13.46k$. so he wins more than 4k$ compared to threetabling which is enough to compensate the bad month and keep some money for the next two losing months which will appear in the next 40 months by average.
2. The exact pro in ur example earns 8.7k per month by average sixtabling, and a little more than 7.2k threetabling. In the same period of 20 months he wins 28.4k$ more sixtabling! Enough to finance a house in the long term and never pay any rent anymore. Or to save enough money in the first twenty months to pay the rent for the next 2 years.
******************************
edit:
By that i assume that the average result per month is the middle of ur confidence range which it should be, isn´t that correct? Having said that i must add that in both cases this is not the case. If the 50$ for the pros tournaments include rake he earns 4k per month threetabling, 4.25k per month at six. about 10% more if rake is excluded. Both is far away from the average of ur confidence ranges so there is something wrong with ur numbers anyway. seems like it was 100$ tourneys the imaginary pro played, that would make more sense.
*******************************
Of course i assume that an aspiring pro knows enough about bankroll management to
handle the swings. Most people would never try to make a living with poker if they had to be concerned about the next say six months, u obv need some money BEFORE u decide to get professional.
Plus the pro will of course double his rakeback, and that is always the same amount of money if he keeps up his volume. This is a significant amount of money he can really plan with cos it depends on volume only without any variance.
The pro in ur example pays about 10K per month for rake sixtabling and 5k else. with a 27% rakeback sixtabling gives u 1350$ more every single month. both wont be homeless but the rakeback gives the sixtabler an extra apartment.
If you are a recreational player u obv play the way u like and enjoy the most and that of course is ok. but u dont need to justify that with statistics and obv not with a single number instead of looking at the situation as a hole. And that is the case with ur example.
gl at the tables and for the leaderboards,
frank
Frank,
I don't think you understand the concept of a hypothetical example. All I'm pointing out is if this is the case (and there will be times where it is and many where it isn't) that the increased variance from multitabling sometimes means you should adopt the risk averse less table strategy.
Criticising the model on the basis of the relavance of the assumptions is rather pointless as you're criticising it for no doing things it was never meant to.
Obviously if someone has a large bankroll or can multitable with affecting their ROI then the model isn't relevant to their situation.
The point is there will circumstances when it is relevant and that's when it should be used. But don't criticise something for not doing what it doesn't claim to do!
i guess i understand as many concepts as u do. a very common concept is to discredit someone who criticizes u. well applied.
i dont criticize ur assumptions or ur model, i criticize the conclusion: i point out that under ur assumtions, with ur model threetabling is not an optimal strategy even if u are prone to never lose a month. i dont mind threetabling until brm allows u sixtabling.
I just took ur model and watched it from all sides rather than just from that side u only see "sixtabling could make u homeless". i use ur assumptions to point out that "you can live in a better appartement sixtabling equally save in the same model" without ever discussing the relevance of ur assumptions. The model does not show what u said it shows.
so in short, i criticized something for not doing what it DOES claim right?
what i also criticize are some statements like "low variance = longer and deeper downswings". What do u mean by that? "Variance is often incorrectly blamed for downswings." I don´t think u understand the concept of variance.
what about the numbers in ur model? i understand the concept of hypothetical numbers - they do not fit into hypothetical models. This seems to be the case here. They are obv wrong aren´t they? Or is the model wrong? So if they are wrong, is what u want to show with the model correct still?
u don´t reply here or to any of my statements and questions. what u have to say does not fit the discredit concept u prefer to apply.
ur welcome to discuss ur statements or mine but do so. if there is any wrong in my statements or numbers ur free to show. By generally discrediting me without saying anything about my points u show that they are pretty good.
franxic
Frank wrote:
"First u say variance is lower but downswings are longer and deeper which is obv a contradiction. the bigger ur swings the higher is the variance." "what i also criticize are some statements like "low variance = longer and deeper downswings". What do u mean by that? "Variance is often incorrectly blamed for downswings." I don´t think u understand the concept of variance"
I think it's clear that you don't understand what variance is. Variance is the squared distance of data points from a mean, which in poker terms is either your ROI or average profit depending upon what scale/units you prefer. I teach under grad stats and game theory so I think I understand the concept of variance somewhat better than you based upon your comments. There is a very common misconception amongst poker players about the relationship between ROI, variance and downswings.
Lower variance is entirely compatible with longer and deeper downswings because the main thing that determines the length and depth of a downswing is mean ROI not variance. The lower the ROI the longer the downswing. If you have a negative ROI your on one permanent downswing -that's got nothing to do with variance that's got to do with being bad at poker.
In my example the absolute figures are just an example I can come up with an infinite number of different situations where you should choose the risk averse strategy. Assuming that ROI goes downwing the number of tables you play and that you need a minimum income then the model is valid. BUT the model opnly applies when these situations are true. You keep criticising it by creating situations where it isn't so it's not relevant. There are situations were it is relevant and then it is useful to make these sorts of calculations to determine whether or not to adopt a risk averse or risk prone strategy.
"i dont criticize ur assumptions or ur model, i criticize the conclusion: i point out that under ur assumtions, with ur model threetabling is not an optimal strategy even if u are prone to never lose a month. i dont mind threetabling until brm allows u sixtabling."
I just took ur model and watched it from all sides rather than just from that side u only see "sixtabling could make u homeless". i use ur assumptions to point out that "you can live in a better appartement sixtabling equally save in the same model" without ever discussing the relevance of ur assumptions. The model does not show what u said it shows.
so in short, i criticized something for not doing what it DOES claim right?"
Where do I claim that this model is about optimising profit? The model is there to determine whether to adopt a risk averse or risk prone strategy. By definition the potential beenfits from risk prone behaviour are greater BUT if you must make a minimum amount of profit over a given period then you should check to see whether the risk prone strategy has a chance of not making that amount. So again you are creating a straw man and blowing it down. You are again criticising something that hasn't been claimed.
i am not an under grad, i did some maths.
swings in any direction obv. are directly related to variance since "Variance is the squared distance of data points from a mean" not regarding if this mean (commonly known as expectation) is positive or negative. if swings in any direction get longer and deeper, variance will increase. if there are no swings, variance will be zero. someone with a negative roi is not on a permanent downswing, he has a negative expectation.
i strongly hope that is what u teach the under grads. I would be surprised if that differs from ur book.
"
There is a very common misconception amongst poker players about the relationship between ROI, variance and downswings.
"
Oh, is there?
"
Lower variance is entirely compatible with longer and deeper downswings because the main thing that determines the length and depth of a downswing is mean ROI not variance. The lower the ROI the longer the downswing. If you have a negative ROI your on one permanent downswing -that's got nothing to do with variance that's got to do with being bad at poker.
"
there obv. is...
what do the math cracks out there think?
"
In my example the absolute figures are just an example I can come up with an infinite number of different situations where you should choose the risk averse strategy.
"
oh i dont doubt that. if u assign random numbers to random models u can pretty much prove anything. any model in a world without brm can prove this.
I dont plan to set up any sort of war with u but because i think what i state does not mean i´m any sort of too dumb as u suggest.
gl at the tables
"swings in any direction obv. are directly related to variance since "Variance is the squared distance of data points from a mean" not regarding if this mean (commonly known as expectation) is positive or negative. if swings in any direction get longer and deeper, variance will increase. if there are no swings, variance will be zero. someone with a negative roi is not on a permanent downswing, he has a negative expectation."
But swings are not measured againt the mean they are measured against zero. That's why ROI is more important because it determines the distance from zero and thus is the main factor controlling the length and depth of swings.
Variance only really matters when you're switching betwen tournament types. If the comparison was between 6-seaters and 180-seaters then variance would have a huge effect on the extent of downswings.
But when comparing 3-tabling and 6-tabling on the same type of table the difference in variance will be very small. It will be slightly higher 3-tabling if your ROI is higher simply because wins have the biggest effect on variance and you will have a higher proportion of wins with a higher ROI. So when deciding how many tables to play at a time variance is pretty much irrelevant, what matters is ROI.
omg u know thats bullshit.
repeat it as often as u want, i see no flower growing
Why is that bullshit?
A downswing is a period when you lose money so it is measured against zero. i.e a period when your profit is zero.
You don't measure a downswing against you mean. If you have a mean ROI of 20% and you run for 100 games at 15% that is not a downswing. You're still making money.
I've never heard anyone in the poker community call a period of making money a downswing. People plot sharkscope graphs but they are plots of profit against zero not of ROI.
"A downswing is a period when you lose money so it is measured against zero. i.e a period when your profit is zero."
u lose money in a downswing compared to ur expectation, not to zero. u talk about a special, extreme way of havin a downswing everyone can have not regarding his mean roi.
"You don't measure a downswing against you mean. If you have a mean ROI of 20% and you run for 100 games at 15% that is not a downswing. You're still making money."
swings by definition are short-term effects of variance which add to zero over a sufficient sample, so it is. swings have no effect on mean roi, and mean roi has no influence on swings. variance is related to the game u play however good u play it. u can be on a downswing (short term) and win money if u win more by average (long term).
a negative roi is not -as u state- a downswing cos it is by definition a long term result.
dont blame me for that, this is what maths says. period.
"I've never heard anyone in the poker community call a period of making money a downswing. People plot sharkscope graphs but they are plots of profit against zero not of ROI."
thats absolutely true because :
"There is a very common misconception amongst poker players about the relationship between ROI, variance and downswings."
what u say proves this and u should know that if u teach under grads.
"swings have no effect on mean roi"
??? swings have no effect on anything they are caused but don;t have an effect.
"and mean roi has no influence on swings"
Oh dear you really just don't understand poker stats do you. People have greater downswings in turbos compared to regular tourneys entirely because of ROI effects not variance. Turbos run faster so there is a greater element of chance, less time for skill to show through, hence lower ROI. For a winning player variance is actually slightly lower in turbos (it's alightly higher for players with ROI's below about 8-10% depending on the rake). This is another common misconception amongst poker players who often claim variance is higher in turbos because they get bigger and deeper downswings. They get them because of lower ROIs.
Turbos have greater downswings with lower variance. Therefore this must be being caused by a lower ROI.
hard to believe ur serious here.
as i stated swings are a short term result. if someone loses money because of any roi effect this is not a swing. hard to understand for u, but again: don´t blame me. blame maths. or tell me what is wrong about my conception of roi, variance and downswings MATHEMATICALLY. i guess ur phantasy is sufficient for that.
"Turbos run faster so there is a greater element of chance"
"For a winning player variance is actually slightly lower in turbos"
excuse me.. this is the kind of statement which proves u don´t understand variance. if u would, u would understand that this is a CONTRADICTION.
Again, repeating that there is an effect of swings on ur roi or of roi on ur swings is just nonsense (bullshit). if ur roi sinks on the long term this is NOT A SWING. ur EXPECTATION has sunk. there is NOTHING u can say about variance from that fact.
"Turbos have greater downswings with lower variance. Therefore this must be being caused by a lower ROI."
my 12-year-old nephew understands that roi is a long term result not influenced by any variance. swings are short term results caused by variance. They have no effect on mean roi since they add to zero. if any effect does not add to zero over a sufficient sample, IT IS NOT A SWING. GREATER DOWNSWINGS with LOWER VARIANCE again are a contradiction as i showed before.
well at least it is kind of funny to read the different versions of ur misconception.
i suggest u repeat it just often enough till i won´t answer anymore or the math changes to ur favor. i guess the second won´t happen, but i won´t answer anymore before u tell where eternal truths like "This is another common misconception amongst poker players who often claim variance is higher in turbos because they get bigger and deeper downswings. They get them because of lower ROIs." are to be read.
What i state is written in any stats book, ur issues seem to be a pretty private invention.
try "The Poker Mindset" (Taylor & Hilger) page 105: "A downswing is a phrase poker players use to describe a period when they earn significantly less than their expected win rate over a number of sessions."
or "Secrets of Sit´n´gos" by Phil Shaw (page 22): "It should also be noted that standard speed sitngos will yield you a far higher ROI than turbos (...) with far less variance (...)."
There is NOT A SINGLE correct statement in ur post, and i suspect u know that.
"There is NOT A SINGLE correct statement in ur post, and i suspect u know that"
Why on earth would write rubbish??
"or "Secrets of Sit´n´gos" by Phil Shaw (page 22): "It should also be noted that standard speed sitngos will yield you a far higher ROI than turbos (...) with far less variance (...).""
Just because it's written in a book doesn't make it correct. As I said there's lots of major misconceptions held by poker players. For a winning player variance is lower in turbos. So that statement above is complete BS.
This is something I'm thinking of publishing a journal article on because of the widespread misunderstanding amongst poker players.
Here's some data showing the variance from the type of table that I've played the most of with identical pay-offs for regulars and turbos. I usually play turbos so I don't have that much regular data but it's enough if there's a huge difference in variance as claimed by Phil Shaw.
This data is for $12+1 6-seaters on Stars.
I've played 405 turbos and 81 regulars.
Here's my finish positions from sharkscope for turbos
and regulars
The pay outs are $46.80 and $25.20
The variance on my turbos is 373.8
The variance on my regulars is 441.5
i.e. exactly as I said it is lower for turbos because I'm a winning player and my ROI is lower on turbos (27% as opposed to 42% on regulars). Therefore I get a lower proportion of wins on turbos. Wins create the greatest affect on variance as they are the data points furthest from the mean. I have been amazed at the of supposed respected poker pundits who make this claim that turbos have higher variance without ever actually doing the maths to back up their incorrect claim. It's obvious theoretically that a winning playing has lower variance in turbos for the reasons I've explained - less wins equals less variance. But I've now showed it's true for a real dataset.
Feel free to redo the calculations and see for yourself that Phil Shaw is talking absolute BS.
I just did it with some of your data. The $2 180 turbos and the $4 180 normals. I have halved everything for the $4 tables - i.e. assumed they were $2 tables to make them comparable.
You have played 797 turbos and 1514 regulars
The variance on your turbos is 124.5
The variance on your regulars is 164.4
i.e. the variance is lower in turbos because you have a lower ROI (26% in the turbos and 42% in the regulars). So even your own stats show that turbos have lower variance.
absolutely. publish that. great sample size. expert conclusions. this will get u prizes.
"For a winning player variance is actually slightly lower in turbos"
i guess because
"Turbos run faster so there is a greater element of chance"
i´m looking forward to read that in any journal. please inform me.
i would be extremely keen on learning what the distribution of ur finishing positions has to do with swings since it says nothing about winning/losing streeks. what does it say about short term results compared to ur expectation? i don´t get it...
why didn´t these donks ask U before they wrote their joke books?
oh now i got it: it says nothing. even if sample size was sufficient (which is not) it says nothing about any swing. and exactly the same about variance. what u show here is a long term result over all ur tourneys. u have more wins than last places. this is because ur a winning player. congrats. but do u believe this is a random variance effect? this effect is stronger in regulars. congrats. ur roi is even higher here (what a coincidence), which means u win even more compared to non-cashes. this is not a random distribution, which means the distribution of ur finishing positions is determined mostly by skill and of course not by variance.
to prove that variance is lower in turbos take a million tourneys turbo and standard and compare 100-tourney-samples. the added squared difference from the mean gives u a variance value. nothing else does, the sample sizes are discussable, but not the way how variance is determined. hard to accept for u, but repeating bullshit does not make it any better.
btw "u show" that "variance" for ur six-players turbo is three times higher than for my 180-players. another interesting subject where the books are wrong and u could write in a journal about.
"Why on earth would write rubbish??"
i´m not jesus, ur the only one to know that.
would u mind me posting the above in the 2+2 forums? the reaction there should give u a rough estimate about ur chances of being published in any journal.
and reading this would definitely make my day..
So you say that variances is much greater in turbos in one post and then that you will need a million samples to show it. How does that follow? If there is a big effect then a few hundred datapoints should be more than enough to detect it.
I've actually done the turbo v regular analysis for several players. Two things always come out the same. ROI is lower in turbos. Variance is lower in turbos. These are facts.
If you're correct that turbos have much higher variance it should be very easy for you to find data to prove your correct. So where is it then?
Landsbanki, how is variance lower in turbos? because you can play more?
over say 10,000 games for both turbo and non turbo, non turbo's less variance right?
Hey phar, welcome to the rather long running disagreement between me and franxic and indeed me and the poker community in general.
There is a long held and widely cited belief amongst poker players that turbos have more variance. This is only true for losing players. Winning players suffer less variance in turbos. With turbos you get longer and deeper downswings (here I define downswing as period of losing, not a period of doing less than your ROI even if you are still winning as Franxic defines it). This occurs because of lower ROIs in turbos for winning players.
Variance is the squared distance of any point in a distribution away from the mean. For poker stats the mean is your ROI. Wins increase variance more than any other outcome (except for DoNs and HUs where every thing works in the opposite way, so for a winning player losses increase variance). Wins increase variance more because that is the outcome furthest from your ROI. It's easiest to understand with examples.
For a pokerstars 12$+1 6-seater turbo you get $46.80 for winning and $25.20.
Lets assume an average ROI of $3.50 per game on these. Which is an average pay out of $16.50 adding in the $13 stake and rake.
The squared distance of a win from the mean here is $46.80-$16.50 squared which equals 906. A second place contributes $25.20-$16.50 squared to the variance which is 75.7. A loss is -$16.50 squared which is 272.25
So the main contributor to variance is wins by a long way. So the more wins the higher the variance. Also the fewer wins the long and deeper your downswings. You win more in regulars so you have higher variance in the regulars.
Where there is no difference in pay out as is the case in the $12+1 regulars and turbos on stars the only difference is ROI, which determines how you sample from the 3 outcomes. With regulars there is a greater bias towards wins therefore more variance. As I show above from both my own data and franxic's data. I've also done this for a number of other players. In all but one case the players has lower ROI and lower variance in turbos. I did find one exception where one player had a higher variance in turbos but they also had a higher ROI in the turbos, which isn't usual but does explain why the variance is higher as it's ROI that drives the variance.
Variance does however generate longer and deeper downswings in MTTs compared to STTs. This is because the variability in pay outs in MTTs is much greater. A win in a 180 seater generates a huge increase in your variance. It's easiest to understand this in terms of how steep your sharkscope graph is at any point. With an MTT win it goes almost vertically upward - i.e. a huge effect on variance compared to a loss in an MTT which is a small step down.
omg. whats that all about? a simple statement about poker that once again turns into complex mathematical equations. i mean is that all really necessary?
KEEP THINGS SIMPLE UNLESS ITS WARANTED.
My friend was wanting to start playing the game and i was going to advise him to read some of the posts on here, pfffff, not anymore, he will think that it is a complicated game that requires you to be a rocket scientist to play.
think i will just tell him to read the harrington books
And Pharreal, you are totally correct $/hr is what is important and there is a cross over point between volume and roi at where you are most profitable. But i also think enjoying the game is very important too(for me), so i would sacrafice some $/hr for a more enjoyable session.
ok u teach the under grads. i hope u dont teach them the kind of bullshit u state here. The number of correct statements does not increase by repeating or adding incorrect statements. it is still zero.
"The squared distance of a win from the mean here is $46.80-$16.50 squared which equals 906."
yepp and if u have an upswing the distance gets bigger and it´s zero if u run average?
what u keep stating remains wrong and gets even wronger every time u try to defend the shit u wrote before. there is no relation to variance which always compares "small" samples with the mean. e.g. if u win 1 of five six-persons by average over some thousend games but only win 10 from a hundred sample without changing ur game, u are on a downswing caused by variance. understand that and dont try to prove ur bullshit with phantasy maths.
it´s just ridiculous and emberrassing what sort of bullshit u state here, and ur only hope is that noone ever reads that.
"So the main contributor to variance is wins by a long way. So the more wins the higher the variance."
that´s funny. if a player wins every tourney he plays what will his variance be? maximal uhuuuu. really funny if u weren´t serious.
btw the correct answer is 0 (zero), any average talented under grad may explain u that. If they don´t talk to u: if u win every game, u therefore win every game in any given sample, so the result af any sample will be equal to ur long term results (u won every game). equal results means distance is 0 (zero), zero times zero remains zero so the squared distance of ur sample result from the mean result is zero. now add the squared distances of every single sample (0+0+0+...=0) and ur result will surely be zero. all without any scientific calculator, how do u like that? feel free to redo the calculation..
why do ur results differ? u just calculated something else u personally assigned the name variance to.
good luck for ur career as a scientific columnist. u might accelerate it if u aggregate ur best statements like "more luck -> less variance" or "less variance -> deeper downswings" in a single article. u could easily fill a whole book with statements as elaborated as these. and ur under grads had something to laugh about. not to forget ur revolutionary new way of calculating variance with a single long term result which cleary proves phil shaw does not know what he is talking about. u must be extremely smart since u are the only human being to understand that.
when will ur article "180 plrs tourney variance broadly misconcepted by poker community (just 1/3 of a six plr)" be printed?
or "The Poker Mindset is wrong - Landsbanki tells the truth"
"Phil Shaw is stating complete bullshit - Landsbanki explains why"
"A Swing is not a short term result - Landsbanki invents math"
"Variance not skill (ROI) is responsible, if u have more total wins than u should have by average - Landsbanki corrects common misconception - Poker IS like rolling a die"
waiting so desperately...
i´m outta here, good luck at the tables and with ur nobel prize
"if a player wins every tourney he plays what will his variance be?"
I agree it will be zero but this doesn't happen in the real world does it. In 6-seaters or above wins always increase variance - show me one person's stats where this doesn't happen.
You shown no evidence to back up your argument, neither has Phil Shaw or anyone else. People make statements without doing the analysis. The analysis shows I'm correct.
hil-la-ri-ous
no evidence? i found easy examples which contradict ur statements, quoted books, told u that it wasn´t variance u calculated (u really think it was mua ha ha) and ur results therefore are bullshit. Any "analysis" based on them also is.
for a person with a brain that is a lot of evidence.
Something written in book is not evidence. LOL. Is everything ever written in a book true? Of course not. So both Darwin and the Bible are correct by your claim.
Show me some real data from real poker players which backs your claim. I've done it. You haven't. End of.
franxic wrote:
"
hil-la-ri-ous
no evidence? i found easy examples which contradict ur statements, (...) told u that it wasn´t variance u calculated (u really think it was mua ha ha) and ur results therefore are bullshit. Any "analysis" based on them also is.
for a person with a brain that is a lot of evidence.
"
Show me some real data from real poker players which backs your claim. I've done it. You haven't. End of.
was about to go off about so many people believing in god(s) and the bible but
maybe i should save my religion rant for another thread in a later day.
"So the main contributor to variance is wins by a long way. So the more wins the higher the variance."
"I agree it will be zero but this doesn't happen in the real world does it. In 6-seaters or above wins always increase variance - show me one person's stats where this doesn't happen."
any player with any roi in any tourney reduces his variance with a win if the win moves the roi of the actual sample closer to mean roi. this is true for appr. 50% of the wins which do appear. again an under grad can explain u that.
ridiculous and embarrassing.
Exactly pharreal, sometimes this is very hard to explain to the new ones, they only think in ROI