Math Wizards Needed

[jmbreslin 0]

I'm trying to compare potential returns between heads-up NL tournaments and single-table SnGs but I don't have enough math know-how to do the proper calculation. The idea is to get a sense of whether heads-up tourneys have potentially better or worse return rates than the full-table SnGs.Compare $5.50 HU to $5.50 single-table SnGs.Assume an ITM% of 50% in each.Over 10 games your return in HU play would be (5x10)-(5x5.50) = $22.5What would your 1st-2nd-3rd place breakdown have to look like to get the same return at $5.50 STTs assuming an ITM of 50%?

[Dogpatch 2]

Is anybody making ITM 50% consistently?

[troyomac 0]

Is anybody making ITM 50% consistently?

Not that I know of.

[LongLiveYorke 38]

I'm not sure where to start.

Assume an ITM% of 50% in each.Over 10 games your return in HU play would be (5x10)-(5x5.50) = $22.5

Here's your first problem. If we get in the money 50% of the time in a heads up match, it means we win 50% of the time. In your calculation above, you forgot to include our initial buy in of when we win.So, 50% of the time we get a profit of 4.5 (10-5.5) and 50% of the time we lose a net of 5.5.(.5)(4.5) - (.5)(5.5) < 0.If we win only half of our matches, we will lose a lot of money due to the rake.So, if we want to do better by getting in the money 50% of the time in a full table sit n go, we could get all 3rd places and do better since we will be earning money as compared to losing it.

[Chiefclint 0]

try againif in hu u are only 50% then u are losing money. In following ur example 10 games total buyin and fee = 10 x 5.50 or 55 bucks. if you win half or 5, times the 10 bucks each time the that is 50 bucks for a toal LOSS of 5 dollars. And I agree with the others, 50% ITM on single table sng's not happening.

[LongLiveYorke 38]

What would your 1st-2nd-3rd place breakdown have to look like to get the same return at $5.50 STTs assuming an ITM of 50%?

For a given profit margin, there are many ways that we can balance our 1st, 2nd, and 3rd's. Here's why:Let P be the profit that we are trying to match, let A be the percentage of times, conditioned on us getting in the money, that we get 1st place, B being the same for second, and C being for third.We want:P = .5( A(first place money) + B(second place money) + C(third place money)) - buy in.With the condition that A + B + C =1With first, second, and third place money being constants, we have three variables to adjust (A, B, and C) to get the P we want, with our normalization condition. The set of A, B, C for a given P defines a sheet in two space.

[psujohn 0]

The set of A, B, C for a given P defines a sheet in two space.

GEEK!

[LongLiveYorke 38]

GEEK!

Oops, I meant it is a two dimensional surface in three space. Sorry.

[RhinestoneCowboy 2]

Ah-Ha!! Yorke is wrong for once.3rd place is less than $10 in STT, so if he gets ITM 50% of the time he gets 5 3rd place wins which will be less than 5 HU wins.I'm bookmarking this post.

[LongLiveYorke 38]

Ah-Ha!! Yorke is wrong for once.3rd place is less than $10 in STT, so if he gets ITM 50% of the time he gets 5 3rd place wins which will be less than 5 HU wins.I'm bookmarking this post.

Really? I just assumed it would be more than that. I guess I should have checked before hand. So you'd need to throw in a small percentage of 2nd places as well. Or something.

[sixhands 0]

Paging Actuary ... paging Actuary ...Actuary to the general forum please ...

[Potsie P 0]

Agree with earlier post that OP's original math is wrong. 50% wins HU is a $5 loss.To the OP--coincidentally, I've doing this EXACT thing for the last 3 weeks, so your question is perfect timing. Granted, I have a limited sample size (I only play 1 table at a time and only for a couple hours each night) but here are my results to date:$5 HU: 21-7 ($210 winnings - $147 buy-ins = $63 profit)$10 HU: 2-1 ($40 winnings - $31.50 buys = $8.50)Overall HU record: 23-8 (74% win) + $71.50$5 6-man SNGs: 4-4 (3 wins + 1 second vs. 4 no cash = $72 - $44 = $28 profit)$5 10-man SNGs: 2-1 (2 thirds vs. 1 no cash = $18 - $16.50 = $1.50 profit)Overall SNG record: 6-5 (54% win) + $29.50Note: I was EXACTLY 50% cash in 40 SNGs (20 6-mans, 20 10-mans) for the month of May. I actually started this HU thing because I was disappointed in so few first-place finishes during that stretch. This is not meant as a brag post--I only post this because I have to disagree with others that it is very difficult to cash 50% of the time. I'm not that great of a player (really), just a miniscule winning one over 2 years online at the micro-stake levels. Hope this helps.

[sixhands 0]

Agree with earlier post that OP's original math is wrong. 50% wins HU is a $5 loss.To the OP--coincidentally, I've doing this EXACT thing for the last 3 weeks, so your question is perfect timing. Granted, I have a limited sample size (I only play 1 table at a time and only for a couple hours each night) but here are my results to date:$5 HU: 21-7 ($210 winnings - $147 buy-ins = $63 profit)$10 HU: 2-1 ($40 winnings - $31.50 buys = $8.50)Overall HU record: 23-8 (74% win) + $71.50$5 6-man SNGs: 4-4 (3 wins + 1 second vs. 4 no cash = $72 - $44 = $28 profit)$5 10-man SNGs: 2-1 (2 thirds vs. 1 no cash = $18 - $16.50 = $1.50 profit)Overall SNG record: 6-5 (54% win) + $29.50Note: I was EXACTLY 50% cash in 40 SNGs (20 6-mans, 20 10-mans) for the month of May. I actually started this HU thing because I was disappointed in so few first-place finishes during that stretch. This is not meant as a brag post--I only post this because I have to disagree with others that it is very difficult to cash 50% of the time. I'm not that great of a player (really), just a miniscule winning one over 2 years online at the micro-stake levels. Hope this helps.

But this based on such a small sample size.Long term it is very difficult to maintain that win rate.

[vbnautilus 48]

Let's say you play 10 $5.50 SnGs. You have spent $55.00 that you need to recoup to break even.With 50% ITM, if they are all 3rd place you earn $45. Since that's not enough, you need $10 more somewhere to break even. Each 2nd place gives you an additional $4.50 and 1st gives you an additional $13.50 each time. So out of those 5 wins, either one has to be a 1st place, or you need at least 3 2nd places. (9 handed Sng pays 9, 13.50, 22.50).

[Dogpatch 2]

But this based on such a small sample size.Long term it is very difficult to maintain that win rate.

Amen.You're just running good right now Potsie. Look at your stats after 1000 games and see what your ITM% is.

[Potsie P 0]

But this based on such a small sample size.Long term it is very difficult to maintain that win rate.

Ah, good friend, we meet again. (no sw...haha)No, you're TOTALLY right. Talk to me in 6 mos. and we'll see if I'm running the same. Note: I actually did a post awhile back asking what a good sample size WOULD be for this exact situation but didn't get any takers. Any thoughts?All I know is, by and large, it's been a pretty fun couple months for me, and I needed it--I'm sure months March and April already offset these stats considerably. It's just more fun to keep stats when you're running good vs. getting killed LOL.

[stevesavi 0]

Oops, I meant it is a two dimensional surface in three space. Sorry.

why dont the three variables A B C map onto three space?

[Potsie P 0]

Amen.You're just running good right now Potsie. Look at your stats after 1000 games and see what your ITM% is.

Thanks for the number, Dog. Just don't hold your breath. With a full-time job and a 2-year-old at home, May was what I would call a "really active" month. You might wait have to wait 2 years for me to get that count!

[vbnautilus 48]

why dont the three variables A B C map onto three space?

3 dimensions along which to vary do create a 3 dimensional space. But any threeparticular points in that space define a two-dimensional surface. Imagine just two points, you have a line (a one dimensional object). Once you add thethird point you have defined a surface (2 dimensional object).

[sixhands 0]

Ah, good friend, we meet again. (no sw...haha)No, you're TOTALLY right. Talk to me in 6 mos. and we'll see if I'm running the same. Note: I actually did a post awhile back asking what a good sample size WOULD be for this exact situation but didn't get any takers. Any thoughts?All I know is, by and large, it's been a pretty fun couple months for me, and I needed it--I'm sure months March and April already offset these stats considerably. It's just more fun to keep stats when you're running good vs. getting killed LOL.

Hello again :)Yes it is more fun to keep stats when you are running well but you LEARN the most from your stats when you are running badly.As for sample sizes, I'm not too sure but I would guess that a decent ITM rate would probably be about 15 - 20% over 1000 and above.Correct me if I'm wrong guys

[Potsie P 0]

Hello again :)Yes it is more fun to keep stats when you are running well but you LEARN the most from your stats when you are running badly.As for sample sizes, I'm not too sure but I would guess that a decent ITM rate would probably be about 15 - 20% over 1000 and above.Correct me if I'm wrong guys

Obv first comment is a good point. In fact, me knowing (without the details) that I was getting slammed in March and April was what caused me to start tracking my play more specifically in May. I know it is not a coincidence that I started playing better once I started tracking where I was exiting these SNGs and making notes as to why. It's such a simple yet effective thing--I don't know why I used to drift in and out of doing this, cuz I'll be doing it from now on based on these last two months.That said, the percentage comment seems low. If you only win (and assuming a first place ITM) 1 in 5 10-man SNGs, you still lose $2.50. Isn't by virtue of 30% of places paying your percentage goal at least 30%?Anyway, thx for the reply.

[dna4ever 2]

[LongLiveYorke 38]

why dont the three variables A B C map onto three space?

Depends on what exactly we're talking about.A, B, C are defined by two equations of three variables (the normalization constraint and the equation for the profit). Now, that leave one degree of freedom between the three of them. In other worlds, if you imagine the normalization constraint and the profit to be two dimensional sheets in three space, the one dimension of freedom will be the intersection of the two sheets (the two dimensional surfaces defined by C = C (A ,B ) for example). So the intersection is a line if the equations are independent. But if we're looking to make a profit more than P in the profit equation, we demand that the normalization equation be true, but the profit equation becomes an inequality. So, the solution is the part of the surface of the normalization constraint that is above the surface of the profit equation. Thus, it defines a two dimensional surface in the three space (RxRxR) of the variables A, B, C.

[CrazyJoe 0]

Hello again :)Yes it is more fun to keep stats when you are running well but you LEARN the most from your stats when you are running badly.As for sample sizes, I'm not too sure but I would guess that a decent ITM rate would probably be about 15 - 20% over 1000 and above.Correct me if I'm wrong guys

ITM 15-20% of single table sng's? even 10 handed ones, top 30% is ITM, so if every game was completely luck based you would be expected to finish in the money 30% of the time, how can 15-20% be good? or do you mean 15-20% ROI?

[sixhands 0]

ITM 15-20% of single table sng's? even 10 handed ones, top 30% is ITM, so if every game was completely luck based you would be expected to finish in the money 30% of the time, how can 15-20% be good? or do you mean 15-20% ROI?

I have dificulty sometimes when it comes to thinking and typing. It's a genetic defect, I plan on passing it on as much as I can.It's called IHAVENOIDEAWHATTHEFUCKIAMTALKINGABOUTSOIJUSTMAKESTUFFUPITIS