Question about SNGs and fluctuation

[Chaostracize]

I have a friend who is a very good SNG player. I say this after having talked about numerous hands with him and seen him play. He is currently playing the 215s.

I was just informed that over 5,000 SNGs he has an ROI of 6%. I think before any conclusions can be made regarding skill that more the person in question should play more SNGs, but he disagrees stating that the 6% is an accurate represntation of his skill and he just isn't good enough to be playing at the level he's been playing at for months.

Is there a satisfactory conclusion that anyone can come to?

I know this really comes down to a simple variance and bell curve post, but I think that this player has simply been getting unlucky and over the course of another 1-5k games his ROI should be easily around 15%.

Thoughts?

[pearljam]

6% ROI over 5,000 tournaments at 200+15's is very good.

[citanul]

[ QUOTE ]
6% ROI over 5,000 tournaments at 200+15's is very good.

[/ QUOTE ]

response 1 for this thread:

no, it's not.

it's making an ok amount of money, it's not even close to "very good"

citanul

[Chaostracize]

That, both player in question and I, agree with.

[citanul]

I think:

a) your friend is "good enough" to play those games, especially considering it's stilll possible for him to improve (clearly)

b) i think that he likely should not be playing those games if he hates large swings

c) i think you are not qualified to be judging what a 215s player's ROI is

d) i think that oh, some other randomly chosen metric, like his ROI, is much more likely to tell you his ROI than your randomly chosen one of "pull a number out of a hat"

e) i think that since (a) implies he can still learn (there are after all players who beat those games for more than 6% and likely (b) is true, that since your friend believss he shouldn't be playing those games, he likely should move down and play lower, and work on his game.

f) it sounds like your friend *is* interested in getting his game better. that's a good thing. if moving down is the way he wants to do it, more power to him. it sounds like he's thought about this possibly more than you.

that's response 2 for this thread.

citanul

[Chaostracize]

50% of the time I like you. The other 50% I think you're a big dick.

[citanul]

[ QUOTE ]
50% of the time I like you. The other 50% I think you're a big dick.

[/ QUOTE ]

which time are we on right now?

citanul

edit; it's funny, because about 50% of the time i'm a nice guy, and 50% of the time i'm a big dick, but sometimes when i am one of those ways, i have to bluff the other way, just to mix it up. keeps them guessing. if your bluffing percentage isn't all that right, your opponents start to peg you pretty good, and that's no good.

[Jman28]

[ QUOTE ]
[ QUOTE ]
50% of the time I like you. The other 50% I think you're a big dick.

[/ QUOTE ]

which time are we on right now?

citanul

[/ QUOTE ]

I almost posted this exact same reply, but I knew that you would.

[Myst]

[ QUOTE ]
50% of the time I like you. The other 50% I think you're a big dick.

[/ QUOTE ]

Agreed. But I think you already knew that... or dont care. One or the other.

[Chaostracize]

This was a good post. Now I like you. You bastard.

[Myst]

[ QUOTE ]
6% ROI over 5,000 tournaments at 200+15's is very good.

[/ QUOTE ]

Its not. It really is not. The variance alone would make any sane man want to jump out the window.

But then again, people play blackjack with worse edges for a living. Meh. Its all perspective.

[jon462]

[ QUOTE ]
I think:
d) i think that oh, some other randomly chosen metric, like his ROI, is much more likely to tell you his ROI than your randomly chosen one of "pull a number out of a hat"
citanul

[/ QUOTE ]
Im really confused. What does this mean. Are you saying OP made up 6%?

6% may not be "excellent" but it certainly seems good enough to me. earning 64k over 5k tournies should certainly be a comfort to someone who doesnt feel they are good enought to play at that level. Anyway it certainly beats 30% ROI at the 20s..

(good grief thats over 1 mil tourney dollars spent)

[Chaostracize]

He's saying I'm an idiot for proposing that player X should be making 15%. Which is 95% true.

[citanul]

[ QUOTE ]
[ QUOTE ]
I think:
d) i think that oh, some other randomly chosen metric, like his ROI, is much more likely to tell you his ROI than your randomly chosen one of "pull a number out of a hat"
citanul

[/ QUOTE ]
Im really confused. What does this mean. Are you saying OP made up 6%?

6% may not be "excellent" but it certainly seems good enough to me. earning 64k over 5k tournies should certainly be a comfort to someone who doesnt feel they are good enought to play at that level. Anyway it certainly beats 30% ROI at the 20s..

(good grief thats over 1 mil tourney dollars spent)

[/ QUOTE ]

my point was the 6% is way more likely to be his ROI than 15%, since one was his ROI over a large number of tournaments, and one was a number the OP said he thought was more likely his friend's ROI.

i think that players who have never played higher stakes games, even properly bankrolled, hell, even hugely bankrolled, say 100 buyins, underestimate the psychological effects on even a strong willed person that a 20-30 buyin drop will have on a player at high stakes. yes, i understand that when you multiply 5% x 215 x 5k, you get a large number, but you have to realize how that 5% happens. it doesn't happen by someoen hands you x bucks per tournament. it happens by you lose and you win and then it averages out. with a player hitting 5%, there's more a lot of losing in there, and particularly likely large downswings, even of the nice bleed kind that are just slow drips down in money for say, a week or a month at a time. it hurts. a lot.

say you drop 30 buyins in a week and you're a 5% player, what do you say to yourself in the morning of the next week? "it's ok, in only 600 games, i'll get that 30 buyins back"? (and that's pre rake!) i know this sounds really rambly and everything, but you have to think about these things, even when you have a lot RoR, you have to consider things beyond just ROI, like stress level involved per enjoyment.

this doesn't mean that someone who's rate is say 5% or 6% should just give up and quit, it means i think that someone in such a position needs to work on their game more, as while the money is nice in the long run, it'll drive you nuts. as another poster said, it'll make a sane man jump out windows. or, in my more preferred line, it'd make mother theresa homicidal, or yugo like women, or something like that.

citanul

ps: yeah, i was just noticing yesterday how totally insane it is that say, i who am young and never had a job that paid anything, AND didn't play that much so far this month, AND didn't play for huge stakes this month, can have spent >50k in tournament entries this month.

[Chaostracize]

Another very, very good post.

[citanul]

[ QUOTE ]
But then again, people play blackjack with worse edges for a living. Meh. Its all perspective.

[/ QUOTE ]

i haven't read any literature/empirical work on BJ, and am actually looking to do so. have you got any to recommend? i have this feeling in the back of my head that because good blackjack play involves putting more money out when you believe yourself to be at an advantage, the swings aren't that terrible (though they clearly exist).

also wondering if anyone knows the edges for the scenarios:

1) playing perfect blackjack w.r.t. counting without varying your bets

2) playing basic strategy, no counting, with varying your bets on some reasonable spectrum

3) playing perfect counting strategy with varying your bets on some reasonable spectrum

if anyone has any reference to that that'd be great.

haha, i'm wrong forum / hijack man.

citanul

[jon462]

[ QUOTE ]
[ QUOTE ]
But then again, people play blackjack with worse edges for a living. Meh. Its all perspective.

[/ QUOTE ]

i haven't read any literature/empirical work on BJ, and am actually looking to do so. have you got any to recommend? i have this feeling in the back of my head that because good blackjack play involves putting more money out when you believe yourself to be at an advantage, the swings aren't that terrible (though they clearly exist).

also wondering if anyone knows the edges for the scenarios:

1) playing perfect blackjack w.r.t. counting without varying your bets

2) playing basic strategy, no counting, with varying your bets on some reasonable spectrum

3) playing perfect counting strategy with varying your bets on some reasonable spectrum

if anyone has any reference to that that'd be great.

haha, i'm wrong forum / hijack man.

citanul

[/ QUOTE ]

good sng strategy also involves putting your chips out there when you have an advantage, why do you think there would be less variance with blackjack?

To answer your question, house edge is 2% on #2 (you cant beat BJ w/o counting), and I imagine your edge would be so incredibly small on #1 it wouldnt be worth it.

There's a 2+2 book on blackjack, its designed to be simple enough to be practical to non-rainmen.. but DS does recommend some books with more complicated strategies..

[citanul]

ty.

i own, and have on my desk, sklansky on bj, but hte light is off, and i don't think he answers the exact question i asked.

i didn't mean to compare the variance of bj to that of sngs, just to what some may perceive the variance in bj to be.

/hijack

citanul

[Mr_J]

"To answer your question, house edge is 2% on #2"

No it's not. Basic strategy usually get you down to 0.5% or less (depends on rules).

"you cant beat BJ w/o counting"

There are some games that are beatable with perfect basic strategy, but these games are rare.

"There's a 2+2 book on blackjack, its designed to be simple enough to be practical to non-rainmen.."

Card counting isn't hard at all. Just takes practice. BJ21.com

[jon462]

[ QUOTE ]
"To answer your question, house edge is 2% on #2"

No it's not. Basic strategy usually get you down to 0.5% or less (depends on rules).

"you cant beat BJ w/o counting"

There are some games that are beatable with perfect basic strategy, but these games are rare.

"There's a 2+2 book on blackjack, its designed to be simple enough to be practical to non-rainmen.."

Card counting isn't hard at all. Just takes practice. BJ21.com

[/ QUOTE ]

I was under the impression it was 2% with perfect play, Ill assume you are right though and stand corrected. My point is you arent getting an edge. By "some games" I assume you mean non-blackjack games? There have been table games that are beatable in casinos before, but they dont last very long before some donk posts the game on the internet and people come from all over until the casino figures it out <g>

Card counting is easy for me in a 1-2 deck shoe with grandma dealing.. any faster or more cards involved I get lost <g> Sklansky recommends basic +1 -1 (face cards, cards 6 or under), but he recommends some books with much more complicated strategies that are supposedly more accurate.

[bones]

[ QUOTE ]
was just informed that over 5,000 SNGs he has an ROI of 6%. I think before any conclusions can be made regarding skill that more the person in question should play more SNGs, but he disagrees stating that the 6% is an accurate represntation of his skill and he just isn't good enough to be playing at the level he's been playing at for months.

[/ QUOTE ]

Regardless of what you may read here from a number of posters, ROI is not a measure of skill. It's not the litmus test to see if you should move up. It doesn't tell you if you are the Johnny Chan of sngs or if you need to go back to playing the $1/$2 at stars. It also doesn't predict how well you will run or play in the future.

It just tells you what your return on investment is so far. That's all.

[citanul]

while this:

[ QUOTE ]
It just tells you what your return on investment is so far.

[/ QUOTE ]

this:

[ QUOTE ]
That's all.

[/ QUOTE ]

is not, because:

[ QUOTE ]
ROI is not a measure of skill

[/ QUOTE ]

is totally either disingenous or just wrong, whichever.

surely you aren't going to argue that over a large sample size a player with greater skill is going to outperform, on a ROI basis, a player with lesser skill. that's just silly. and yes, yours truly does believe that 5k games does verge on large enough sample size to start to know things and yes, i'm sure you're willing to throw out facts about how you could just be learning the first 500 games, but yeah, people can learn to throw out the games from their early career, and things like that.

pick a point at which you think you started playing the way you're playing now, play identically for 5k games, and yes, your ROI is a measure of your skill.

citanul

[Myst]

[ QUOTE ]
while this:

[ QUOTE ]
It just tells you what your return on investment is so far.

[/ QUOTE ]

this:

[ QUOTE ]
That's all.

[/ QUOTE ]

is not, because:

[ QUOTE ]
ROI is not a measure of skill

[/ QUOTE ]

is totally either disingenous or just wrong, whichever.

surely you aren't going to argue that over a large sample size a player with greater skill is going to outperform, on a ROI basis, a player with lesser skill. that's just silly. and yes, yours truly does believe that 5k games does verge on large enough sample size to start to know things and yes, i'm sure you're willing to throw out facts about how you could just be learning the first 500 games, but yeah, people can learn to throw out the games from their early career, and things like that.

pick a point at which you think you started playing the way you're playing now, play identically for 5k games, and yes, your ROI is a measure of your skill.

citanul

[/ QUOTE ]

Agreed.

[raptor517]

[ QUOTE ]
[ QUOTE ]
while this:

[ QUOTE ]
It just tells you what your return on investment is so far.

[/ QUOTE ]

this:

[ QUOTE ]
That's all.

[/ QUOTE ]

is not, because:

[ QUOTE ]
ROI is not a measure of skill

[/ QUOTE ]

is totally either disingenous or just wrong, whichever.

surely you aren't going to argue that over a large sample size a player with greater skill is going to outperform, on a ROI basis, a player with lesser skill. that's just silly. and yes, yours truly does believe that 5k games does verge on large enough sample size to start to know things and yes, i'm sure you're willing to throw out facts about how you could just be learning the first 500 games, but yeah, people can learn to throw out the games from their early career, and things like that.

pick a point at which you think you started playing the way you're playing now, play identically for 5k games, and yes, your ROI is a measure of your skill.

citanul

[/ QUOTE ]

Agreed.

[/ QUOTE ]

ah the best romances begin with dissent. myst and citanul sittin in a tree.. holla

[bones]

[ QUOTE ]
surely you aren't going to argue that over a large sample size a player with greater skill is going to outperform, on a ROI basis, a player with lesser skill. that's just silly. and yes, yours truly does believe that 5k games does verge on large enough sample size to start to know things and yes, i'm sure you're willing to throw out facts about how you could just be learning the first 500 games, but yeah, people can learn to throw out the games from their early career, and things like that.

pick a point at which you think you started playing the way you're playing now, play identically for 5k games, and yes, your ROI is a measure of your skill.

[/ QUOTE ]

Being a mod, I'm sure you read most threads. Even the repeat ones with the same questions, over and over. Who are the people who typically ask ROI questions? People who have played <1k games and haven't been reading/studying for very long. They will almost never play a meaningful sample size at the same skill level. I would certainly hope not.

I'm not saying that there isn't a correlation between ROI and skill level. I'm just saying that it's just not as strong as people make it out to be, and certainly not strong enough to be considered THE metric of skill.

And I'm not disputing that over a course of 5k games by 2 established players in the same buyin level, the stronger player will very likely have a higher ROI. But as a practical matter, this will rarely happen.

[citanul]

[ QUOTE ]
Agreed.

[/ QUOTE ]

damnit, now when i change my location next week you'll think it's because i don't hate you.

(btw, how great would it be to hint at changing my location, and then next week just change it to "I hate Messy_Jesse" or "I hate Myst"?)

crap, i clearly need to go to bed.

[raptor517]

playing at 6% will make you go insane. at any level. you could potentially break even over 2k+ sngs right? ugh. /endlife. holla

[citanul]

[ QUOTE ]
Being a mod, I'm sure you read most threads. Even the repeat ones with the same questions, over and over.

[/ QUOTE ]

pfft, there's some movie with "you assume too much" that i can't remember what movie it is, but yeah:

you assume too much.

mostly i just run around all willy nilly and do what i want.

citanul

[citanul]

oh yeah, it's cuz I'M DRUNK ON POWER!

BOW DOWN!!!

citanul

[Isura]

[ QUOTE ]
i haven't read any literature/empirical work on BJ, and am actually looking to do so. have you got any to recommend?

[/ QUOTE ]

"Theory of blackjack" is the ultimate mathematical treatment of the game.

i have this feeling in the back of my head that because good blackjack play involves putting more money out when you believe yourself to be at an advantage, the swings aren't that terrible (though they clearly exist).
[ QUOTE ]


[/ QUOTE ]

This is really not true. The edges you are pushing in blackjack are very small. It's near impossible to put yourself in a position to have a large edge that is very common in even high stakes poker games. Another factor, is that even professional blackjack players consistently miscalculate or misunderstand their true edge. I dealer dealing 5 more cards (before shuffling) in a 2 deck game can pretty drastically change a counter's edge.

[ QUOTE ]
1) playing perfect blackjack w.r.t. counting without varying your bets

[/ QUOTE ]

Varies depending on the rules. But note that counting without varying bets can never gain a significant edge. The best you could hope for is about +0.5%. Even to achieve this, you would need to use a counting strategy optimized for strategy adjustment (eg. the popular Hi-Lo method is not good enough, it's optimized to utilize betting variations).

Aside: It's possible to gain a 2% edge in a good game using a decent strategy while varying your bets AND playing strategy (both according to the count).

[ QUOTE ]
2) playing basic strategy, no counting, with varying your bets on some reasonable spectrum

[/ QUOTE ]

There has never been found a betting strategy (without counting) that can gain an edge using just basic strategy plus betting variation.



[ QUOTE ]
3) playing perfect counting strategy with varying your bets on some reasonable spectrum

[/ QUOTE ]

About 2% is the best you can hope for. Better edges used to be possible with advanced techniques such as dealer manipulation, dealer hand reading, etc but that is a whole complicated subject that is not really applicable to today's game (ie casino's caught on, and have taken safety measures).

[ QUOTE ]
if anyone has any reference to that that'd be great.

[/ QUOTE ]

The following are very good. Also, search around www.bj21.com. (http://www.bj21.com.)

Professional blackjack by wong (http://www.amazon.com/exec/obidos/tg/detail/-/0935926216/qid=1125126306/sr=2-2/ref=pd_bbs_b_2_2/002-4469676-1428801?v=glance&s=books)

Theory of blackjack by griffin (http://www.amazon.com/exec/obidos/tg/detail/-/0929712137/qid=1125126306/sr=1-3/ref=sr_1_3/002-4469676-1428801?v=glance&s=books)

World's greatest blackjack book by humble (http://www.amazon.com/exec/obidos/tg/detail/-/0385153821/qid=1125126372/sr=1-14/ref=sr_1_14/002-4469676-1428801?v=glance&s=books)

[citanul]

isura,

thanks for your detailed reply.

looking back at my post, i don't even know what i was thinking i was writing when i wrote up #2. like i said, clearly nap time.

citanul

[testaaja]

[ QUOTE ]
isura,

thanks for your detailed reply.

looking back at my post, i don't even know what i was thinking i was writing when i wrote up #2. like i said, clearly nap time.

citanul

[/ QUOTE ]
Here is a link for you http://www.blackjackforumonline.com/ . You can find some neat stuff bout BJ over there. About card counting etc.

[Isura]

[ QUOTE ]

looking back at my post, i don't even know what i was thinking i was writing when i wrote up #2. like i said, clearly nap time.

[/ QUOTE ]

Yeah, the whole riding streaks, martingale system type approaches have been tried before. And some people like Jerry Patterson, Gregory Manarino have tried to sell books and course packages promoting these non-mathematical approaches (they are an interesting read atleast).

Anything in print by Jerry Patterson, Ken Warren, or Gregory Mannarino is pretty worthless, but a fun read. /images/graemlins/smile.gif

[tjh]

[ QUOTE ]
[ QUOTE ]
50% of the time I like you. The other 50% I think you're a big dick.

[/ QUOTE ]

which time are we on right now?

citanul

edit; it's funny, because about 50% of the time i'm a nice guy, and 50% of the time i'm a big dick, but sometimes when i am one of those ways, i have to bluff the other way, just to mix it up. keeps them guessing. if your bluffing percentage isn't all that right, your opponents start to peg you pretty good, and that's no good.

[/ QUOTE ]

Wow...
Something is up... the big dick to nice guy ratio used be 90/10 ..

[Mr_J]

"By "some games" I assume you mean non-blackjack games?"

No I do mean blackjack. Some single deck games with favourable rules have a very small edge off just perfect basic strategy. Haven't played one of these, just remember hearing on BJ21 that they do exist.

Better to just play poker though /images/graemlins/wink.gif

[jon462]

[ QUOTE ]
"By "some games" I assume you mean non-blackjack games?"

No I do mean blackjack. Some single deck games with favourable rules have a very small edge off just perfect basic strategy. Haven't played one of these, just remember hearing on BJ21 that they do exist.

Better to just play poker though /images/graemlins/wink.gif

[/ QUOTE ]

what favorable rules? Have their been casinos that offer 3:1 on blackjacks? any special rules I have seen (like spanish 21 or blackjack\3 card poker combo) are just extra sucker bets. Im not arguing with you that just sparked my curiousity.

[byronkincaid]

[ QUOTE ]
playing at 6% will make you go insane. at any level. you could potentially break even over 2k+ sngs right? ugh. /endlife. holla

[/ QUOTE ]

Can you put me in your will for the $9K rakeback? /images/graemlins/smile.gif

[rvg72]

I wrote this ROI simulation program:

Link to forum thread (http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Board=singletable&Number=321 8857&Forum=f22&Words=rvg&Searchpage=0&Limit=25&Mai n=3218857&Search=true&where=bodysub&Name=&daterang e=1&newerval=1&newertype=w&olderval=&oldertype=&bo dyprev=#Post3218857)

You could use this to do some analysis on the numbers. If he was really a 15% ROI player then there is less than a 1 in 10,000 chance he would run at 6% or lower over 5,000 SNG's... Even a 9% ROI player would have only an 18% chance to run at 6% or lower over 5K... With that sample size chances are very high that he is a 5 to 7% ROI player.

4 tabling the $215's at 6% isn't too shabby is it? About $50/hr or so but there would be some big swings... Over 1000 SNG's he would likely (50%) have a 40 buyin or more drop at some point.

rvg

[Mr_J]

*only* an 18% chance if his EV is 9%?? That's still a reasonable chance to run significantly worse. Kind of surprising, then again the figures are low so 1% makes a large difference.

6% at the 215s is more like $90 an hr /images/graemlins/wink.gif

[rvg72]

[ QUOTE ]
6% at the 215s is more like $90 an hr /images/graemlins/wink.gif

[/ QUOTE ]

Really? You get in 7 games an hour at the $215's? I figured 4 an hour if you're playing in sets and taking pee breaks is about right.

[Mr_J]

Seriously who 4tables in sets /images/graemlins/tongue.gif

Also remember rakeback adds around $20.

[Myst]

[ QUOTE ]
[ QUOTE ]
6% at the 215s is more like $90 an hr /images/graemlins/wink.gif

[/ QUOTE ]

Pfft, im getting that right now at the $55s.

[Mr_J]

If I was heavily cashed (ie 6%+RB over 5k $215s) 4 tabling 215s is more attractive than 6-8 tabling 55s. Then again if I was hitting 6% at the $215s I could probally make as much at the $109s. So yeh, 4 table the $109s thanks.

[Chaostracize]

My friend 8 tables.

I think he is a good player, at lease he can beat the rake, which means he is well above average. but a top-notch SNG player should be able to get a 12%+ ROI over a large sample. yeah, i think 5000 SNG is a large enough sample to calculate you ROI. but maybe your friend is still improving his game. i think on PP 215, 15%-16% ROI is about the max ROI you can maintain in a long run especially if you are multi-tabling.

[Myst]

[ QUOTE ]
i think on PP 215, 15%-16% ROI is about the max ROI you can maintain in a long run especially if you are multi-tabling.

[/ QUOTE ]

So states the mysterious stranger. HIS WORD IS DOCTRINE AND LAW!!!!!!!

[pearljam]

[ QUOTE ]
[ QUOTE ]
6% ROI over 5,000 tournaments at 200+15's is very good.

[/ QUOTE ]

response 1 for this thread:

no, it's not.

it's making an ok amount of money, it's not even close to "very good"

citanul

[/ QUOTE ]

But I read in the FAQ that 6% was the highest a good player could sustain over a long period of time in the 200+15s.

thanks. hehe
cetainly higher ROI is achievable if you play single table ,choose your game very carefully and only play in peak hours. but you wont make a lot of money playing that way.

[Isura]

[ QUOTE ]


what favorable rules? Have their been casinos that offer 3:1 on blackjacks? any special rules I have seen (like spanish 21 or blackjack\3 card poker combo) are just extra sucker bets. Im not arguing with you that just sparked my curiousity.

[/ QUOTE ]

- splitting pairs
- dealer standing on soft 17s
- double down after splitting pairs
- splitting aces
- multiple cards when splitting pairs
- early surrender
- late surrender (NEVER offered anymore, very favorable to player)

[jon462]

[ QUOTE ]
[ QUOTE ]


what favorable rules? Have their been casinos that offer 3:1 on blackjacks? any special rules I have seen (like spanish 21 or blackjack\3 card poker combo) are just extra sucker bets. Im not arguing with you that just sparked my curiousity.

[/ QUOTE ]

- splitting pairs
- dealer standing on soft 17s
- double down after splitting pairs
- splitting aces
- multiple cards when splitting pairs
- early surrender
- late surrender (NEVER offered anymore, very favorable to player)

[/ QUOTE ]

oh, ok. I thought those were all standard except for the surrenders (never seen either at a casino) and the soft 17\hard 17 rule (ive seen both about equally)