How 8 tabling is more profitable in TLR than 4 tabling.

I've been doing some thinking on how the winrates between 8 tabling and 4 tabling differ. I believe this makes some sense. Keep in mind I'm no math major.

Say you could kill the LHE game 4 tabling, that would include finding the perfect table selection and seat selection, while also playing perfect. I will use 3/6 as an example, with no rakeback and no bonuses, although these would come into the equation later. So, your winrate is a function of Cards, Your Play, Table Selection, Seat Selection. Otherwise seen as f(C, P, T, S), whereas if one of these variables increase, so does your winrate. You also cannot control the cards you get, thus, C is a variable we cant change.

So, consider that P, T, and S are all perfect, then by killing a LHE game you 'could' potentially make 4 BB/100 over 40k hands, would net you approx. $9600 but you'd have to play 167 hours that month to do so, making your hourly winrate around $57/hour. Not too bad. However, even the best LHE player does not play this perfect, whether this is because we don't know the most optimal seat selection or table selection, or we just don't play 100% perfect over that many hands consistently. So, your winrate could be around 3 bb/100 and you still can't complain.

What happens as you increase tables?

Theoretically, if you could maintain T, S and P to be the best, you would have the same winrate and thus increase your hourly winrate by the number of tables. This is rather impossible. As you increase the number of tables played, P will undoubtedly decrease as you are forced to make more decisions in the same amount of time, while the quality of the reads you have also decreases. T will decrease as you must spend less time finding better tables and still keep up the increased number of tables. Finding good tables become increasingly difficult the more decisions you have to make in that amount of time. And, not suprisingly, S will also decrease as T decreases.

How much these variables decrease is dependant on yourself. If you could stil play 100% optimally, then this is one variable you don't have to worry about and could be accomplished with enough practice. T and S, on the otherhand must still suffer. Critics argue that T and S are important to maximizing your winrate, and I agree. But at some point the pure value of getting as many hands in as possible counterfeits the added profitability of good table and seat selection, as long as P is as high as possible. Granted, having both good table and seat selection is optimal while still 8 tabling, it just isn't feasible.

In doing the calculations for 8 tabling, we increase our hands from 240 per hour to 480 per hour. I would say the 'best' a person could do is 1.5 BB/100 while 8 tabling (as seen over a few hundred thousand hand sample from 2+2), but, this winrate is at 30/60 and not 3/6 so i imagine it could be as high as 2 bb/100 and I wouldn't be surprised. So, over the same 40k hand month you would actually earn 1/2 the amount, in 1/2 the time and thus, your hourly rate is higher. The hourly rate would go from $57 to $114, and the time it takes to make this would half from 167 hours to 84. Meaning, you could theoretically play the same number of hours per month 4 tabling but make twice the money.

The conclusion could go wrong, however. It depends on how you define your month, most people have a goal of hands per month and how many hours they can play per week (if you dont you should). If you have unlimited hours to play that many hands per month and you are satisfied with that, then by all means continue to use T and S as optimally as possible and continue to 4 table. But, if you are like most people then hours are hard to come by and getting the most hands in during the same time period is the best way to make the most money.

Realistically, its impossible to know your winrates without a large sample size of hands both while 4 tabling and 8 tabling. But, as long as you can 8 tabling more than 1/2 the winrate you can 4 table at, then it is more optimal to 8 table than 4 table. The only way to find this out is to do it for yourself. If you have never tried it, then you can't knock it.

Not to be a jackass, but it seems like all of your points are pretty obvious

Play more tables = harder = lower BB/100
More hands per hour = more BB/hour
Most profit = combination of above to facts that works best

I tried quite a while ago to start playing more than 4-tables at once back when I was at 3/6 and over a pretty large sample size(I think like 60,000 hands, but I don't have poker tracker handy), my BB/100 dropped from the mid-to-upper 2's down all the way to about -0.5(about even counting rakeback)....Not the largest sample size ever, but it's gonna be a while before I try more than 4-tables again

I've tryed doing this multitabling. I ran 1.4BB/100 over 30K hands at 2/4$ my winrate 3-4 tabling was around 2.5bb/100 over 10K hands so none of these samples are very big but they still show how your hourly BB/100 is indeed bigger when multitabling.

@Jeff. I think theres a calc fault in your post as you think the hourly winrate go from 57$ to 114$

4 tabling:
doing 240 hands an hour is 2,4BB * 3BB/100 = 6,8BB/hour = 39$

8 tabling:
Doing 480 hands an hour is 4,8BB * 2BB/100 = 9,6BB/hour =57$

Say you keepup the 4BB/100 while 4 tabling then theres no difference in the hourly rate.

It's pretty simple:

Hand/hour * winrate/hand * = winrate/hour

If hands/hour doubles and winrate/hand is cut in half then we get the same result. So you need to be able to play twice as many tables without cutting your winrate in half.

Also, realize that by sacrificing win-rate for volume you will have more negative 200BB+ swings.

Originally Posted by Fnord

Also, realize that by sacrificing win-rate for volume you will have more negative 200BB+ swings.

Yes and No. yes because in the same period of time you'll play more hands and thus the frequency of getting Aces cracked also increases. But thats only because you see more aces. No because I would have to say all 200+ BB swings i've ever seen were magnyified because of the tilt factor. Playing more tables does not increase your variance per hand just variance per a given time period, as long as you can play with mental competence then the magnitude of the swings shouldnt increase, but the frequency would. I also believe that more tables allows less time for tilt after losing a big hand because you have to forgive and forget to remain successful.

To my knowledge you've never tried 8 tabling, if you have it wasn't over a significant sample size to really matter.

Nehmer: You're right. But some people dont think about things.

Beaver: I think we are talking about the same thing, just that I was using 4 bb/100 for my hypothetical hourly rates while 4 tabling but was talking about 3 bb/100, whereas i did the calculations for 2 bb/100 while 8 tabling.

Anything more than 3-4 tables on my 1024x768 resolution makes me want to cry. Especially when I am tabbing through tables all awaiting actions and I nearly auto-fold a big pocket pair.

The most I like to do (comfortably) is 3 tables, more than that I don't find the time to observe others' play, let alone add notes to them.

Originally Posted by elipsesjeff

Playing more tables does not increase your variance per hand just variance per a given time period, as long as you can play with mental competence then the magnitude of the swings shouldnt increase, but the frequency would.

A reduced win-rate increases the chance of a big downswing over N hands. However, over time is another interesting variable.

Originally Posted by elipsesjeff

To my knowledge you've never tried 8 tabling, if you have it wasn't over a significant sample size to really matter.

I know what moving up to 5 and 6 did for my game. Comperable argument. My biggest problem was following the action after switching metally from another window. I would see raised pots with good hands and assume I had the pre-flop lead. I estimated I was making around one mistake like that per hour. Also, I would get stuck in broken games and suddenly post in a 3 handed game.

Rakeback is a big part of the picuture as it allows us to profitably add tables as there is less pressure to beat the rake.

Finally, endurance might be an issue. If you can put in more hours at fewer tables it might be far better on a week to week basis.

All this really boils down to "whatever works for you."

Endurance and burnout are big factors in this issue.

While 7-tabling on one screen I did suffer some burnout.

My level of thinking decreased and autopilot took over whenever I had 3 or more hands going at the same time.

Also you need to do all them quick default decissions and this leads to burnout I think.

Another thing to consider is how much your learning curve suffer form multitabling..?

The good things in multitabling: You can play a lot of hands on short time thus you can get a good sample size in little time. This is good for confidence to you know theres a very good chance that you are a winner in a particular game.

If you can do it, then it makes wonders for building a bankroll

Originally Posted by Fnord

A reduced win-rate increases the chance of a big downswing over N hands.

I believe the reduced winrate is the effect of the big downswing, not the other way around. Dont mean to be picky but, a winrate is a measure of the past that we use for future estimates. By seeing a lower winrate we can assume that previously there may have been a large downswing (or even larger breakeven stretch) but can not and shouldnot make any inferences in the future.

My point is that if you could 8 table at just over 1/2 the rate at which you could 4 table for the same amount of time, you would win more. The winrate is statistically accurate of the past, with all downswings and upswings included and should not be used to estimate any future potential swings.

And finally, the likelihood of a 200 BB downswing isnt as high as a 200 BB upswing during that same stretch.

Originally Posted by elipsesjeff

My point is that if you could 8 table at just over 1/2 the rate at which you could 4 table for the same amount of time, you would win more.

No argument, I'm just saying that playing at 75% of your 4 table win-rate will make it more likely that over the course of N hands you will run really bad. The math isn't trival, but it's an obvious and well documented result. Also, if a downswing makes you play worse...

Anyway, like I said it's whatever works for you. I probably would eventually cross into 5 or 6 tables if I played enough, but relating what happened when I tried it. For now, I can play 3-4 full or 2-3 short tables, play my A+ game, beat the crap out of the table and have fun.

Originally Posted by elipsesjeff

I believe the reduced winrate is the effect of the big downswing, not the other way around. Dont mean to be picky but, a winrate is a measure of the past that we use for future estimates. By seeing a lower winrate we can assume that previously there may have been a large downswing (or even larger breakeven stretch) but can not and shouldnot make any inferences in the future.

You're confusing true win-rate with a statistic that attempts to estimate your true win rate. If your true win rate goes down, then (assuming the same variance), the likelihood of a given downswing increases.

The actual occurrence of a downswing only affects the statistic (though it may also affect the true win rate due to tilt, but let's ignore that for the moment).

I think it's safe to say that there is a range of tables at which an individual can achieve his highest true win rate. As he increases (or decreases) outside of this range, the likelihood of large downswings increases.

Originally Posted by elipsesjeff

My point is that if you could 8 table at just over 1/2 the rate at which you could 4 table for the same amount of time, you would win more. The winrate is statistically accurate of the past, with all downswings and upswings included and should not be used to estimate any future potential swings.

Unless a player has a record consisting of several hundred thousand hands, the confidence interval suggested by his statistical record probably only tells him enough to know whether the data was likely to be generated by a player with a positive win rate. This is certainly true for DB's on the order of 50K hands.

Originally Posted by elipsesjeff

And finally, the likelihood of a 200 BB downswing isnt as high as a 200 BB upswing during that same stretch.

Only if you are still a winning player as you increase the number of tables. Since it is difficult to tell this a priori, it is probably more important to judge whether multitabling is successful by the number and magnitude of the mistakes that are directly attributable to multitabling.

One more thing to add. I think anyone striving to increase earn is better served by trying to double limits (play higher) as opposed to doubling the number of tables.

Increasing tables should be done when you are soundly beating the level you are striving to play. JMO.

Originally Posted by koolmoe

One more thing to add. I think anyone striving to increase earn is better served by trying to double limits (play higher) as opposed to doubling the number of tables.

Increasing tables should be done when you are soundly beating the level you are striving to play. JMO.

Good point.

It's pretty easy to get stuck with loads of tables at any limit where you know you'll be a winner in the long run. And you would not benefit from just opening 8 tables at a new limit. I thought I could manage 4 tables of 6max when I came from 7 full, but since I was new in this game it was to big a mouthfull and I know now that many players don't like more than 3 tables of 6 max at a time.