Open Side Menu Go to the Top
Register
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post)

10-25-2017 , 05:39 PM
Hey Max,
many thanks for your reply. It is really helpful to have some advice with real life results from more experienced players. I really appreciate it. Cheers.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-04-2017 , 11:16 PM
Quote:
Originally Posted by Max Cut
Thanks for the questions papaisA. Unfortunately I can't offer much in the way of solid answers when it comes to cEV for spins. One point I will make is that the SwongSim cEV simulations are better viewed as a starting point or bound on the cEV swings that can happen in reality.

For example, one simplifying assumption in the simulations is that your actual edge (expressed in cEV) is the same for each and every game. In reality your edge varies due to opponent skill and how well you are playing at the time.
that's good point that estimated edge and reality can be different due to external factors which you mentioned. so is always good to be self-aware and and stop playing when we deviating from our standard games. however is not cEV number which we introduce into SwongSwim the sum of all factors, which include our deviation from our standard games?

as well, concerning your statement:
Quote:
Originally Posted by Max Cut
I've seen real-life result graphs of 70+ cEV over tens of thousands of games that had multiple break-even stretches of like 2k games, which would be very rare in SwongSim simulated results.
if we take for example 7$ spins on .eu and 70cEV, we can see that about 40% of player will have break even stretch of about 2K games in next 10K games. so probability of two 10K games interval will be 0.4x0.4=0.16% ? which means that 16% of players multiple break-even stretches of like 2k games on sample of 20K games? which is imo not that infrequent.
correct me if I am wrong as I am really not familiar with statistics. probably my calculation are over simplified and there is some other factors which need to be taken into account.

At the same time, taking advantage of the opportunity, I would also like to ask you for advice on bankroll management. I am not math geek, so I use your software (which is great, thank you!) for estimation of my bankroll needed. For purpose of calculation, let's say I have 70cEV on 7$ ps.eu and rb is 4%. I play 4000 games a month and my cost of living is about 600$ per months. I have money aside for next 3 months. assuming that I am willing to accept RoR 1%, is about 120 buy-ins should be enough to overcome variance in next three months according to below data?:

Code:
---------------------------------------------------------------------------------------
Effective	Specified		Simulation
Place	Finish Distribution	Finish Distribution
1	3,8E-05%          	3,3E-05%
2	3,19688E-05%          	5,8E-05%
3	3,00312E-05%          	4,2E-05%
4	0,00114%          	0,001108%
5	0,000959064%          	0,001058%
6	0,000900936%          	0,000742%
7	0,00285%          	0,003042%
8	0,00239766%          	0,002583%
9	0,00225234%          	0,002167%
10	0,038%          	0,037692%
11	0,19%          	0,191008%
12	2,85%          	2,856783%
13	7,581228%          	7,582133%
14	27,336744%          	27,344433%
ITM	38,006572%          	38,022883%
-------------------------------------------------------------------------------
1000  simulations of  12000  games
Expected ROI (with rakeback/bonus/award): 6,27%  (752 Buyins)
--------------------------------------------------------------------------------
99% 	had ROI below   12,72%	(1527 Buyins)
97.5% 	had ROI below   10,00%	(1201 Buyins)
95% 	had ROI below   9,14%	(1097 Buyins)
90% 	had ROI below   8,25%	(991 Buyins)
80% 	had ROI below   7,35%	(883 Buyins)
70% 	had ROI below   6,80%	(816 Buyins)
60% 	had ROI below   6,30%	(757 Buyins)
50% 	had ROI below   5,82%	(699 Buyins)
40% 	had ROI below   5,45%	(655 Buyins)
30% 	had ROI below   5,02%	(603 Buyins)
20% 	had ROI below   4,60%	(553 Buyins)
10% 	had ROI below   3,80%	(457 Buyins)
  5% 	had ROI below   3,27%	(393 Buyins)
  2.5% 	had ROI below   2,78%	(334 Buyins)
  1% 	had ROI below   2,39%	(287 Buyins)
-------------------------------------------------------------------
  99% 	had a downswing greater than   52 buyins
  97.5% 	had a downswing greater than   55 buyins
  95% 	had a downswing greater than   59 buyins
  90% 	had a downswing greater than   63 buyins
  80% 	had a downswing greater than   69 buyins
  70% 	had a downswing greater than   75 buyins
  60% 	had a downswing greater than   80 buyins
  50% 	had a downswing greater than   85 buyins
  40% 	had a downswing greater than   91 buyins
  30% 	had a downswing greater than   99 buyins
  20% 	had a downswing greater than   108 buyins
  10% 	had a downswing greater than   123 buyins
  5% 	had a downswing greater than   137 buyins
  2.5% 	had a downswing greater than   148 buyins
  1% 	had a downswing greater than   165 buyins
---------------------------------------------------------------------------------
  99% 	had a low point lower than   0 buyins
  97.5% 	had a low point lower than   0 buyins
  95% 	had a low point lower than   -1 buyins
  90% 	had a low point lower than   -2 buyins
  80% 	had a low point lower than   -4 buyins
  70% 	had a low point lower than   -7 buyins
  60% 	had a low point lower than   -11 buyins
  50% 	had a low point lower than   -15 buyins
  40% 	had a low point lower than   -20 buyins
  30% 	had a low point lower than   -28 buyins
  20% 	had a low point lower than   -38 buyins
  10% 	had a low point lower than   -52 buyins
  5% 	had a low point lower than   -70 buyins
  2.5% 	had a low point lower than   -88 buyins
  1% 	had a low point lower than   -105 buyins
---------------------------------------------------------------------------------
  99% 	had a breakeven stretch longer than   747 games
  97.5% 	had a breakeven stretch longer than   819 games
  95% 	had a breakeven stretch longer than   893 games
  90% 	had a breakeven stretch longer than   1028 games
  80% 	had a breakeven stretch longer than   1206 games
  70% 	had a breakeven stretch longer than   1370 games
  60% 	had a breakeven stretch longer than   1540 games
  50% 	had a breakeven stretch longer than   1740 games
  40% 	had a breakeven stretch longer than   1978 games
  30% 	had a breakeven stretch longer than   2234 games
  20% 	had a breakeven stretch longer than   2596 games
  10% 	had a breakeven stretch longer than   3201 games
  5% 	had a breakeven stretch longer than   3774 games
  2.5% 	had a breakeven stretch longer than   4407 games
  1% 	had a breakeven stretch longer than   5364 games
---------------------------------------------------------------------------------
as well, if I read the data correctly:
Code:
  1% 	had ROI below   2,39%	(287 Buyins)
there ts =<1% of chance that I will not make at least 287BI in next 12K games (3 months play), so I am relatively safe, to be, financially, in three months form now, at least in the same situation as I am right now after deducting my cost of living for next three months?

I hope my comment is not too inconsistent or unclear as English is not my native language. I will be playing for a living for next couple of months and wish to finally put my finance in order as until now I was messing too much with it and it refrained me a lot from moving up in stake as quickly as I would like. So I will much appreciated any advice concerning proper BR management for spins.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-05-2017 , 11:28 AM
Quote:
Originally Posted by Mirikrom
that's good point that estimated edge and reality can be different due to external factors which you mentioned. so is always good to be self-aware and and stop playing when we deviating from our standard games. however is not cEV number which we introduce into SwongSwim the sum of all factors, which include our deviation from our standard games?

as well, concerning your statement:


if we take for example 7$ spins on .eu and 70cEV, we can see that about 40% of player will have break even stretch of about 2K games in next 10K games. so probability of two 10K games interval will be 0.4x0.4=0.16% ? which means that 16% of players multiple break-even stretches of like 2k games on sample of 20K games? which is imo not that infrequent.
correct me if I am wrong as I am really not familiar with statistics. probably my calculation are over simplified and there is some other factors which need to be taken into account.
Hi Mirikrom --

I was not very clear in my statement. In that response I was writing specifically about simulated cEV results (using "EV Chip Mode" on the visualizer tab). You are correct to notice that the example numbers I gave do not apply to regular simulation results.


Regarding the simplifying assumption topic, perhaps a toy example will help to explain the idea better. Suppose we are simulating a heads-up game where we only face an equally skilled opponent. We would input 50% for 1st-place in the finish distribution and the simulation would give us some numbers that represent what swings we can expect over some number of games.

Now suppose we are facing two different opponents. One we crush and win 75% of the time. The other crushes us and we win only 25% of the time. We expect to face them an equal amount of times, so our overall expected win rate is again 50%. Thus we get the same simulation results as our prior example.

What we have lost is any information about the distribution of when we face the beast versus when we face the fish. In the extreme situation where we play all games against the beast first, followed by all games against the fish, we will need a bigger bankroll because we are more likely to lose in the beginning.

In reality the situation is not extreme like this, but also probably not a completely smooth distribution. The lobby is randomized, but players go on vacation (just an example).

Anyway, this is one type of thing we can lose with our simplifying assumptions (streakiness in the distribution of our edge). Based on anecdotal evidence (graphs/results of real players), it seems this is more of a factor in "EV Chip Mode" than in regular simulation mode. We also lose things external to the game, like rake changes, or the site shutting down, or whatever. These have some non-zero chance and would effect our real-life results.


Quote:
At the same time, taking advantage of the opportunity, I would also like to ask you for advice on bankroll management. I am not math geek, so I use your software (which is great, thank you!) for estimation of my bankroll needed. For purpose of calculation, let's say I have 70cEV on 7$ ps.eu and rb is 4%. I play 4000 games a month and my cost of living is about 600$ per months. I have money aside for next 3 months. assuming that I am willing to accept RoR 1%, is about 120 buy-ins should be enough to overcome variance in next three months according to below data?:

Code:
---------------------------------------------------------------------------------------
Effective	Specified		Simulation
Place	Finish Distribution	Finish Distribution
1	3,8E-05%          	3,3E-05%
2	3,19688E-05%          	5,8E-05%
3	3,00312E-05%          	4,2E-05%
4	0,00114%          	0,001108%
5	0,000959064%          	0,001058%
6	0,000900936%          	0,000742%
7	0,00285%          	0,003042%
8	0,00239766%          	0,002583%
9	0,00225234%          	0,002167%
10	0,038%          	0,037692%
11	0,19%          	0,191008%
12	2,85%          	2,856783%
13	7,581228%          	7,582133%
14	27,336744%          	27,344433%
ITM	38,006572%          	38,022883%
-------------------------------------------------------------------------------
1000  simulations of  12000  games
Expected ROI (with rakeback/bonus/award): 6,27%  (752 Buyins)
--------------------------------------------------------------------------------
99% 	had ROI below   12,72%	(1527 Buyins)
97.5% 	had ROI below   10,00%	(1201 Buyins)
95% 	had ROI below   9,14%	(1097 Buyins)
90% 	had ROI below   8,25%	(991 Buyins)
80% 	had ROI below   7,35%	(883 Buyins)
70% 	had ROI below   6,80%	(816 Buyins)
60% 	had ROI below   6,30%	(757 Buyins)
50% 	had ROI below   5,82%	(699 Buyins)
40% 	had ROI below   5,45%	(655 Buyins)
30% 	had ROI below   5,02%	(603 Buyins)
20% 	had ROI below   4,60%	(553 Buyins)
10% 	had ROI below   3,80%	(457 Buyins)
  5% 	had ROI below   3,27%	(393 Buyins)
  2.5% 	had ROI below   2,78%	(334 Buyins)
  1% 	had ROI below   2,39%	(287 Buyins)
-------------------------------------------------------------------
  99% 	had a downswing greater than   52 buyins
  97.5% 	had a downswing greater than   55 buyins
  95% 	had a downswing greater than   59 buyins
  90% 	had a downswing greater than   63 buyins
  80% 	had a downswing greater than   69 buyins
  70% 	had a downswing greater than   75 buyins
  60% 	had a downswing greater than   80 buyins
  50% 	had a downswing greater than   85 buyins
  40% 	had a downswing greater than   91 buyins
  30% 	had a downswing greater than   99 buyins
  20% 	had a downswing greater than   108 buyins
  10% 	had a downswing greater than   123 buyins
  5% 	had a downswing greater than   137 buyins
  2.5% 	had a downswing greater than   148 buyins
  1% 	had a downswing greater than   165 buyins
---------------------------------------------------------------------------------
  99% 	had a low point lower than   0 buyins
  97.5% 	had a low point lower than   0 buyins
  95% 	had a low point lower than   -1 buyins
  90% 	had a low point lower than   -2 buyins
  80% 	had a low point lower than   -4 buyins
  70% 	had a low point lower than   -7 buyins
  60% 	had a low point lower than   -11 buyins
  50% 	had a low point lower than   -15 buyins
  40% 	had a low point lower than   -20 buyins
  30% 	had a low point lower than   -28 buyins
  20% 	had a low point lower than   -38 buyins
  10% 	had a low point lower than   -52 buyins
  5% 	had a low point lower than   -70 buyins
  2.5% 	had a low point lower than   -88 buyins
  1% 	had a low point lower than   -105 buyins
---------------------------------------------------------------------------------
  99% 	had a breakeven stretch longer than   747 games
  97.5% 	had a breakeven stretch longer than   819 games
  95% 	had a breakeven stretch longer than   893 games
  90% 	had a breakeven stretch longer than   1028 games
  80% 	had a breakeven stretch longer than   1206 games
  70% 	had a breakeven stretch longer than   1370 games
  60% 	had a breakeven stretch longer than   1540 games
  50% 	had a breakeven stretch longer than   1740 games
  40% 	had a breakeven stretch longer than   1978 games
  30% 	had a breakeven stretch longer than   2234 games
  20% 	had a breakeven stretch longer than   2596 games
  10% 	had a breakeven stretch longer than   3201 games
  5% 	had a breakeven stretch longer than   3774 games
  2.5% 	had a breakeven stretch longer than   4407 games
  1% 	had a breakeven stretch longer than   5364 games
---------------------------------------------------------------------------------
as well, if I read the data correctly:
Code:
  1% 	had ROI below   2,39%	(287 Buyins)
there ts =<1% of chance that I will not make at least 287BI in next 12K games (3 months play), so I am relatively safe, to be, financially, in three months form now, at least in the same situation as I am right now after deducting my cost of living for next three months?

I hope my comment is not too inconsistent or unclear as English is not my native language. I will be playing for a living for next couple of months and wish to finally put my finance in order as until now I was messing too much with it and it refrained me a lot from moving up in stake as quickly as I would like. So I will much appreciated any advice concerning proper BR management for spins.
On a technical note, I suggest increasing the number of simulations from the default 1k to something like 20k or more.

The numbers and logic seem correct. As I mentioned before, use this as a starting point. Build in a buffer (because simulations use simplifying assumptions) and contingency plan (because rare events do sometimes happen).

glgl
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-05-2017 , 12:52 PM
Hey Max Cut,
big thanks for answer! a lot of valuable explanations!
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-08-2017 , 06:00 AM
Is there any possibility of getting Swongsim updated to PartyPorker SNG hero 3max??
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-08-2017 , 10:42 AM
Quote:
Originally Posted by sutiplay
Is there any possibility of getting Swongsim updated to PartyPorker SNG hero 3max??
Since they are very very nearly identical, you can use the PokerStars presets. Just use whichever buyin level has the same rake. The only difference is for the top tiers that pay 85/7.5/7.5 instead of 83.3/8.3/8.3. I've added it to the update list.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-08-2017 , 06:58 PM
Hey sir, what is the latest swong out atm? And have you added ignition/bovada?

Thanks
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-08-2017 , 07:12 PM
Quote:
Originally Posted by Nat3
Hey sir, what is the latest swong out atm? And have you added ignition/bovada?

Thanks
Download link: www.dropbox.com/s/ukpadqy24se12th/SwongSim.exe?dl=0


No update for bo yet. It's on the list.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-10-2017 , 09:23 AM
Quote:
Originally Posted by Max Cut
Since they are very very nearly identical, you can use the PokerStars presets. Just use whichever buyin level has the same rake. The only difference is for the top tiers that pay 85/7.5/7.5 instead of 83.3/8.3/8.3. I've added it to the update list.
Thanks man!
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
12-19-2017 , 06:47 AM
Thank you Max cut, love yout software.

Some of SwongSim users might be interested by this thread :

My modest contribution to the study of spin structures
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
12-19-2017 , 09:33 AM
Quote:
Originally Posted by Pad
Thank you Max cut, love yout software.

Some of SwongSim users might be interested by this thread :

My modest contribution to the study of spin structures
Very cool and insightful. Thanks for linking.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
04-03-2018 , 02:53 PM
hey was trying to look up if the jackpots on ignition are even do-able. I tried to put the table I found on their website into swongsim but it s says -98% pre rakeback EV with 36% winrate. I'm confused how it works then, I copied the settings I changed here, and the rake 7% across all stakes.

[SwongSim Simulation Settings]
Simulations=1000
Games=10000
Players=3
PlayerFinishDistribution(1st)=36
PlayerFinishDistribution(2nd)=33
PlayerFinishDistribution(3rd)=31
Frequency1=1
Frequency2=3
Frequency3=9
Frequency4=105
Frequency5=24887
Frequency6=74995
Frequency7=0
Frequency8=0
Frequency9=0
Frequency10=0
BuyinMultiplier1=1200
BuyinMultiplier2=240
BuyinMultiplier3=120
BuyinMultiplier4=15
BuyinMultiplier5=5
BuyinMultiplier6=2
BuyinMultiplier7=0
BuyinMultiplier8=0
BuyinMultiplier9=0
BuyinMultiplier10=0
PrizePool1(1st)=83.333
PrizePool1(2nd)=8.333
PrizePool1(3rd)=8.333
PrizePool2(1st)=83.333
PrizePool2(2nd)=8.333
PrizePool2(3rd)=8.333
PrizePool3(1st)=83.333
PrizePool3(2nd)=8.333
PrizePool3(3rd)=8.333
PrizePool4(1st)=100
PrizePool4(2nd)=0
PrizePool4(3rd)=0
PrizePool5(1st)=100
PrizePool5(2nd)=0
PrizePool5(3rd)=0
PrizePool6(1st)=100
PrizePool6(2nd)=0
PrizePool6(3rd)=0
PrizePool7(1st)=100
PrizePool7(2nd)=0
PrizePool7(3rd)=0
PrizePool8(1st)=100
PrizePool8(2nd)=0
PrizePool8(3rd)=0
PrizePool9(1st)=100
PrizePool9(2nd)=0
PrizePool9(3rd)=0
PrizePool10(1st)=100
PrizePool10(2nd)=0
PrizePool10(3rd)=0
PlayerFinishDist1(1st)=36
PlayerFinishDist1(2nd)=33
PlayerFinishDist1(3rd)=31
PlayerFinishDist2(1st)=36
PlayerFinishDist2(2nd)=33
PlayerFinishDist2(3rd)=31
PlayerFinishDist3(1st)=36
PlayerFinishDist3(2nd)=33
PlayerFinishDist3(3rd)=31
PlayerFinishDist4(1st)=36
PlayerFinishDist4(2nd)=33
PlayerFinishDist4(3rd)=31
PlayerFinishDist5(1st)=36
PlayerFinishDist5(2nd)=33
PlayerFinishDist5(3rd)=31
PlayerFinishDist6(1st)=36
PlayerFinishDist6(2nd)=33
PlayerFinishDist6(3rd)=31
PlayerFinishDist7(1st)=36
PlayerFinishDist7(2nd)=33
PlayerFinishDist7(3rd)=31
PlayerFinishDist8(1st)=36
PlayerFinishDist8(2nd)=33
PlayerFinishDist8(3rd)=31
PlayerFinishDist9(1st)=36
PlayerFinishDist9(2nd)=33
PlayerFinishDist9(3rd)=31
PlayerFinishDist10(1st)=36
PlayerFinishDist10(2nd)=33
PlayerFinishDist10(3rd)=31
OutOf=1000000
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
04-03-2018 , 03:11 PM
Change "OutOf" to 100000 (it's at the very bottom in your post). glgl
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
04-03-2018 , 07:47 PM
Quote:
Originally Posted by Max Cut
Change "OutOf" to 100000 (it's at the very bottom in your post). glgl
tyty
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
04-13-2018 , 03:03 PM
It's probably my worst run in 1K games in about 20K sample across stacks, so not sure if it's pretty normal downswing or ridiculous one(comparing with others), can you shed some light please ?

https://i.imgur.com/blBwFK7.jpg
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
04-13-2018 , 04:11 PM
Looks pretty horrible to me. My best guess would be that it's not a frequent occurrence, but not super rare. Anecdotally, I can find an ~1k sample in my personal graph that is a similar amount under cEV and have seem a couple other graphs just as bad. So not exactly "pretty normal" and probably not "ridiculous" either.

In simulations these events are more rare, probably due to limitations in the cEV mode.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
04-14-2018 , 01:02 AM
Quote:
Originally Posted by Max Cut
Looks pretty horrible to me. My best guess would be that it's not a frequent occurrence, but not super rare. Anecdotally, I can find an ~1k sample in my personal graph that is a similar amount under cEV and have seem a couple other graphs just as bad. So not exactly "pretty normal" and probably not "ridiculous" either.

In simulations these events are more rare, probably due to limitations in the cEV mode.
Thanks for replay Max ,It is good to know that at least it's not that normal, because it really is getting into my head this days and preventing me from putting enough volume,hopefully it will turn back from somewhere
Getting loads of cooler + everyone binks like 2/3 outer ,literally I won like %5 of all flips this month ,lol.

Went out last night & paid a guy to cry for my running in spins

Thanks again
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
05-20-2018 , 11:14 PM
Hi there. I do respect Max Cut, his job and knowledge so I want to post some idea.
I did the research of streaks. For example the chance to see the eagle of a coin 5 times in row 0.5*0.5*0.5*0.5*0.5 = 3.125%. I took my ITM% and saw my streaks. I play about 2 tables at the same time. This is wrong but it works perfectly for normal contribution (for example to win 3 Spins in row). And it has "long tales". The streaks to win 8 Spins in row or to win just 1 in 20 Spins happened the way much more often. I have small database (3-5K tourneys) so to proof this fact , this research has to be made on bigger database.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
08-29-2018 , 01:15 PM
Download link: www.dropbox.com/s/ukpadqy24se12th/SwongSim.exe?dl=0

  • Added "Bust Chart" feature to graph percentage of players that bust given bankrolls (assuming no cashouts).
  • Added preset structures:
    • Party Poker Sit&Go Jackpots (3 max).
    • PWL Jackpot Poker (Bovada / Ignition).
    • Chico WindFall Sit&Go (BetOnline / Tiger Gaming).
  • Removed redundant "Run" button from settings page.
  • Removed "Text Only" check box. "Run New" on Text Output tab will run only text. "Draw New" on Visualizer tab or on Bust Chart tab will draw new visualizer graph, bust chart, and text output.


SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
08-29-2018 , 01:26 PM
Noticed a great free video about Spin variance out by Coffeeyay which serves as a nice tutorial on SwongSim. Thanks, Adam!


Quote:
Originally Posted by SpinLegends
How to master variance in Spin & Go?

In a free 40 minute video our SpinLegends Head Coach - Adam "Coffeeyay" Sobolewski - shows you how to handle and understand variance in Spins.

Watch it here: Variance in Spin & Go's




Mods, please let me know if quoting that post from marketing is not appropriate in this thread. I have no association with any of the involved parties.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
09-13-2018 , 03:30 AM
Quote:
Originally Posted by Max Cut
Noticed a great free video about Spin variance out by Coffeeyay which serves as a nice tutorial on SwongSim. Thanks, Adam!

Mods, please let me know if quoting that post from marketing is not appropriate in this thread. I have no association with any of the involved parties.

Hello, first of all thanks for creating SwongSim, very eye opener even for beginners when it comes to variation.

One thing that I don't understand and that the video doesn't touch is why in the text output you have: 14 places + ITM when spins are a 3 person format.

For instance:

Effective Specified Simulation
Place Finish Distribution Finish Distribution
1 3.6E-05% 3.6E-05%
2 3.3E-05% 3.6E-05%
3 3.1E-05% 2.7E-05%
4 0.00108% 0.001098%
5 0.00099% 0.000997%
6 0.00093% 0.000935%
7 0.0027% 0.002706%
8 0.002475% 0.002481%
9 0.002325% 0.00235%
10 0.036% 0.035825%
11 0.18% 0.180215%
12 2.7% 2.700561%
13 6.642216% 6.644087%
14 26.437968% 26.436073%
ITM 36.006784% 36.007426%


Also, why do you use scientific format on the first 3 places?(3.6E-05% is actually 0.000036?)

Thank you in advance!
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
09-13-2018 , 11:37 AM
Quote:
Originally Posted by llllllll
Hello, first of all thanks for creating SwongSim, very eye opener even for beginners when it comes to variation.

One thing that I don't understand and that the video doesn't touch is why in the text output you have: 14 places + ITM when spins are a 3 person format.

For instance:

Effective Specified Simulation
Place Finish Distribution Finish Distribution
1 3.6E-05% 3.6E-05%
2 3.3E-05% 3.6E-05%
3 3.1E-05% 2.7E-05%
4 0.00108% 0.001098%
5 0.00099% 0.000997%
6 0.00093% 0.000935%
7 0.0027% 0.002706%
8 0.002475% 0.002481%
9 0.002325% 0.00235%
10 0.036% 0.035825%
11 0.18% 0.180215%
12 2.7% 2.700561%
13 6.642216% 6.644087%
14 26.437968% 26.436073%
ITM 36.006784% 36.007426%


Also, why do you use scientific format on the first 3 places?(3.6E-05% is actually 0.000036?)

Thank you in advance!
We break the specified finish distribution into effective places for each multiplier frequency.

There are up to 30 possible effective places (10 multiplier tiers X 3 finish positions). For example, PS has 14 effective places (3 tiers pay 3 places and 5 tiers pay 1 place, for 3 X 3 + 5 = 14).

Each particular number is specified by %frequency X %finish. Entries resulting in 0% are ignored.


=========
Example (PS with 36/33/31 finish distribution):

(tier 1)
1st effective place: 1/1000000 X 36/100 = (0.000036%)
2nd effective place: 1/1000000 X 33/100 = (0.000033%)
3rd effective place: 1/1000000 X 31/100 = (0.000031%)

(tier 2)
4th effective place: 30/1000000 X 36/100 = (0.00108%)
5th effective place: 30/1000000 X 33/100 = (0.00099%)
6th effective place: 30/1000000 X 31/100 = (0.00093%)

Likewise for the rest of the multiplier tiers.
=========


The "effective places" are displayed mostly as a quick check that the simulation distribution is close to the specified distribution


The scientific format must be due some default formatting done by the programming language I use because I do not specify it. It's useful since there could be input settings that result in very long regular format. It could be made to trigger more leniently (for example, so that all preset inputs fit in regular format). I've added it to the list to look at in the future.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-27-2018 , 07:11 AM
hi can someone tell why numbers on winamax.es are wrong:




wich one should i use?

Last edited by ElMarginal; 11-27-2018 at 07:26 AM.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-27-2018 , 09:12 AM
Quote:
Originally Posted by ElMarginal
hi can someone tell why numbers on winamax.es are wrong:




wich one should i use?
Looks like there was an update to their structure at some point. The one on the left is close, but the 3rd tier pays out 100 buyins (not 200).
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote
11-27-2018 , 09:12 AM
Download link: www.dropbox.com/s/ukpadqy24se12th/SwongSim.exe?dl=0

  • Updated Winamax Expresso preset payout structure.
SwongSim -- ROI/variance simulator for lottery-payout SNGs (2k post) Quote

      
m