Sng Variance- This Is Ridiculous

[hockeylax3 0]

Are these all single table SNGs with 9 or 10 players or are they larger fields? I've developed a spreadsheet that gives me some insight into the expected variance of SNGs and if you give me the number of players I can give you some insight into the the likelhood of failing to cash in N tourneys etc. Also if you could include your ROI for these tourneys if you have it (or an estimate) I may be able to provide some insight or at least attempt it.

They are about 2/3 9 players and the other third is 18's or the new 45's. The one win was in a 45 person for 279 so that helps my numbers but not enough to make up for the rest.

[hockeylax3 0]

lol.

In the 50 or so before that I was up a bit but nothing amazing. Like 14 buy ins (+280 - rake so prob not much) . I dont have poker tracker so it's all in excel so its hard to be exact past that bc theres just too many lines.

[hockeylax3 0]

In the 50 or so before that I was up a bit but nothing amazing. Like 14 buy ins (+280 - rake so prob not much) . I dont have poker tracker so it's all in excel so its hard to be exact past that bc theres just too many lines.o its hard to be exact past that bc theres just too many lines.

[psujohn 0]

they never fold. thats why i moved up to where they respect my raises.

True it's far easier to beat better players. Those ignorant low stakes guys love to push in with 72o and suck out on your aces.

[IQCrash 1]

that reminds me of this one time where i had aces hold up for the past ten time i got them. so im on the button and five people limp and i say to myself, self, youre due for them to lose. so i folded them. and guess what? somebody with q4o hit two pair on the flop and i would have lost a ton.

I really hope you're just bad at the whole funny thing and that wasn't serious.

[scottyno 0]

how many of those tournaments you busted in were you all in earlier and survived though. You cant just say "I lost 32 where I got all in with the best hand and lost", if you get it all in with AK vs JT 3 times in a sng and you're the shortstack every time odds are that you'll bust.

[antistuff 0]

I really hope you're just bad at the whole funny thing and that wasn't serious.

i really hope youre just bad at the whole math thing when you tried to call me out and you werent serious.

[hockeylax3 0]

how many of those tournaments you busted in were you all in earlier and survived though. You cant just say "I lost 32 where I got all in with the best hand and lost", if you get it all in with AK vs JT 3 times in a sng and you're the shortstack every time odds are that you'll bust.

I dont have an exact number for that, but its not me sucking early n being a shortie on the bubble. I dont aim for 3rd in these bc its not that much money, i have lost as a top3 stack plenty of times to higher stacks. I just cant figure out if i should just keep playing, or stop and change it up for a while?

[fleung22 1]

i really hope youre just bad at the whole math thing when you tried to call me out and you werent serious.

don't fall for it Crash...if I didn't know any better I'd think this was another one of AcesUp's characters. no way this guy is that stupid.

[dunkirk 0]

I'm going to run your numbers as follows:9 player tourneys (50%, 30%, 20% ) payout structureand I'm going to say that in the previous 50 tournaments you are plus 14 buy-ins (which is very good a solid 28% ROI). I'll let you know what I find out.

[hockeylax3 0]

I'm going to run your numbers as follows:9 player tourneys (50%, 30%, 20% ) payout structureand I'm going to say that in the previous 50 tournaments you are plus 14 buy-ins (which is very good a solid 28% ROI). I'll let you know what I find out.

That +14 was the 50 previous to the 38 i posted about originally.

[Zach6668 513]

there is a way to figure out how unlikely it is to lose that many in a row. i dont how to do it, but im pretty sure that losing that many is approaching it doesnt happen land.you are right, but what you said has nothing to do with what i was saying.

.35^35

[IQCrash 1]

.35^35

.35^35 = 1.1025075 × 10-16

[fishoutawata 0]

Actually, the odds of him losing the next 65/35 after the very last one he lost are... you guessed it, 35%.

shhhhh, let you all keep playing your next hand (or 65) based upon the last. next you are going to be telling people it is impossible to have the nuts pre-flop

[UncleCollins 0]

It might be bad variance.. but it might not be either. Have you considered tweaking your sng strategy? I'm not saying you're playing bad or anything like that at all, but if I didn't cash for 35 sngs in a row than I'd have to reevaluate my game. Just because you're losing doesn't automatically mean it's because of bad variance.

[SCYUKON 0]

Have you been reading Harrington on Holdem or something? I read those books in November and December of 2005 and I had a run of 33 tourneys in January 2006 where I got sucked out on each time. Always in as the fav, and never with an M less than 10 (learned about M in the Harrington books - yeah!). Not always 65/35, but always a fav.The experience killed most of my drive to try and improve further (although I did give in and read HOH III when it came out too - I scored pretty high on all of the questions but big whoop I live in fear of the suckout now). In fact, I have not been the same since - that really killed my zest for the game. Play great for 2 or 3 hours then busto - what type of fun is that?So in short, you must be improving technically cuz it seems the better you get (according to the HOH style) the more you get sucked on, cuz you are getting it in with the best of it more often.That is why if I ever do return to playing a reasonable amount, I want to learn the DN smallball strategy so that you rarely have all the money in unless you have the absolute nuts. More $ and less badbeat stories.Hope you hang in there.D

[Actuary 3]

I know this probably shouldn't be in general, but it will get the best response there.

How'd that work out for you?***********IQ meet sarcasm. Or perhaps you knew and I'm oblivious.*******************Losing 32 out of 34 with 65/35 advantage is 34! / (32! * 2!) * 0.65^2 * 0.35^32Now, I"m not an Appointed Actuary, so that may be wrong.********************OP,stop focusing on the last handstart repecting 3rd place. In 9 man STT, cashing is your priority

[SuperJon 175]

How'd that work out for you?***********IQ meet sarcasm. Or perhaps you knew and I'm oblivious.*******************Losing 32 out of 34 with 65/35 advantage is 34! / (32! * 2!) * 0.65^2 * 0.35^32Now, I"m not an Appointed Actuary, so that may be wrong.********************OP,stop focusing on the last handstart repecting 3rd place. In 9 man STT, cashing is your priority

I couldn't agree with that more. It irks me to no end when people say they play 9 or 10 person SNGs for 1st place. I play them to cash, and make money. Whether it's 3rd, 2nd, or 1st, it's more than I bought in for.

[dunkirk 0]

I don't know if this helps but I ran your numbers on my analysis spreadsheet and a couple of significant things came out. What I did was measured the natural variance of SNGs for a 50 tourney run (our initial sample where your scored an impressive 28% ROI). I estimate the natural variance as follows. I assume an average player (equal number of 1sts, 2nds, 3rds, 4ths, ... 9th ) place finishes. Against a no-rake game we would expect his ROI to be 0 %. Then I ran a 1000, randomized 50 tourney stretches of our average player and record the ROI for each run and then I compute the standard deviation of this distribution. What I discovered was the standard deviation of the ROI estimate on 50 tournaments is about 20% which is really quite large. This tells us that 50 tournaments is not nearly enough to make a decent prediction of your "TRUE" ROI. (In fact it takes a 1000 tournaments to estimate your "true" ROI within a standard deviation of 5%). I think most people underestimate the significance of this.Your ROI of 28% on 50 tourneys is good and suggest that you probably are a winning player but we really can't say this with great certainty. If you are really a break even player at this level there is about an 8% probability that you may go +28% ROI on 50 SNGs at this level. Or course you may be much better than 28% ROI as well. All this is a wordy way of saying that a particular ROI in 50 SNGs is not a good estimate or "true" ROI or what you might expect on the next 50. I don't know what your ROI was over the bad run but it looks like it must have been horrific and makes you a net loser over the composite stretch of 85+ tourneys. This suggests that most likely you are considerably less than a true 28% ROI player and probably better than your "cold" streak. Don't rule out the possiblity that you are not yet a winning player at this level. Keep in mind that the best data we have so far 80 tourneys suggest that you are losing. This is not meant to be critical in the least - it takes a lot of patience to "prove" that you can beat SNGs at any level for hundreds of tourneys.

[Socrates 0]

discussing or bringing up "variance" when you have an inadequate sample size is -EV as well as ridiculous.

[fleung22 1]

I couldn't agree with that more. It irks me to no end when people say they play 9 or 10 person SNGs for 1st place. I play them to cash, and make money. Whether it's 3rd, 2nd, or 1st, it's more than I bought in for.

then I shall irk you. i'm never going to be the guy who sits back during bubble time just to ensure i squeeze into the money.the times i win will make up for the times i'm bubble boy. i may up my attempt to steal blinds when it's near the bubble because it's usually when the blinds are significant and the play is softest. if i'm sitting with a table full of people holding their breaths till they're in the money i'm going to have a massive stack by the time the bubble bursts. i consider this one of the most crucial times in a tournament.

[TWCReborn 0]

I had the same problem as you mention here. I had AA cracked (I bet preflop a standard amount and 60% of the pot post flop on a flopped set) to a guy early who paid to make a flush. I folded when he moved all in on the turn showing me the flush. 2 hands later I reraise with jacks (at this point it is for a third of my chips since I lost a ton in the AA hand). The maniac puts me all in and shows A6. Somehow I knew an A would hit (and it did). Next SNG, 3 left, 2 paid, my aces were cracked to a preflop 88. Next SNG 1 from the money I raise with 88, and loose reraiser moves all in, and I call (she showed 22 and hit a 2 on the river). Next SNG, 5 left 3 paid, my set is cracked to a gut shot drawer when I get it all in with the best of it. ETC ETC. I will spare you the details.Upon analysis I have found that I was not playing too tightly bubbling out. I also found that besides 1-2 cases where I was a 49% dog, I was always the favorite when I put in all my chips. Part of it is variance. However, to treat an entire SNG as one unique and discrete and homogeneous unit is not correct. The variance you see on 1000 SNGs will vary depending on how you play each hand in each of those 1000 SNGs.First off, if you don't already, you probably need to distinguish between chip expectation value and tournament equity. I suggest SNG wizard for this. It is a bit expensive but worth the money. Also Moshman's book on SNG's might also help. The question you have to ask yourself is: If you are getting involved with bad players too frequently for all your chips, who benefits? The answer, most often, is the players NOT involved in the hand. That is not to say you can't bet without the nuts, but I have realized that I can make subtle adjustments to my game to decrease the variance of my chip stack going from hand to hand. The further a hand goes, the further the win percentages move apart (there is usually one winner and one loser-- 100% and 0%). Earlier in the hand, percentages are usually closer, and even AA vs high suited connectors is only 4-1. If you have to survive 2 races for all your chips where you are at a 70% advantage, your chances of survival on average is 49% (with just a tad bit less variance). Would you race for all your chips as a medium stack where you are a 49% dog? Probably not? The players on the sideline benefit the most when players are involving too many chips in showdowns early.There are other things to consider. When you are pushing in high blind situations, are you pushing into call stations? This might be a positive chip expectation play, depending on where you think your hand stands, and depending on the range of hands you expect him to call with, but call stations late stage are liable to bust themselves or bust others who play aggressively so I tend to decrease my aggression especially if these call donkeys will act after me. Raising too much preflop, raising too much postflop, pushing for all your chips with high pairs-- these are all things that even if you are a favorite when the money went it-- they will hurt you unless you are pushing with 1 foldish player to act. Why? Because if you are involving too many chips when you play, you are bound to have up and down chip swings, and consecutive losses are highly likely to knock you out.In tournaments, sometimes it is best to look for ways to reduce the number of chips you have to play for. If your table is loose and aggressive, a lot of equity is gained from letting bad players have showdowns. If you're table is very foldish to raises, you can play more aggressively, but I would still look for the minimum bets to get people to fold. The key to SNG's as I am beginning to learn is to avoid showdowns for a ton of chips until the blinds are getting high. And even then, if I think player A is going to raise every pot, and player B is going to call every raise, and they have similar stack sizes, I most probably won't call or reraise even sitting on JJ unless I am severely short-stacked because player A is going to knock player B out or vice-versa. The higher the blinds, the more I have to gamble, but equally as important, the more likely I am to get called, the less likely I am to push against him. Suppose you have 3 players left and 2 get paid. What if someone raises in front of you all in and you have pot odds to call with AJ, but there is a big stacked call station behind you? I would tend to fold here unless I were seriously short-stacked. This would change if 3 places were paid. If I have the first player covered, and the player to act after you to act has you both covered, and I think my hand could be the best hand, I might push depending on how high the blinds are.Overall there are many parameters to consider: CEV (pot odds and win%), the callishness and range of raising hands for a particular opponent at a particular moment, your calling and raising range (should be optimized to the other parameters), the tournament payout structure, and stack sizes (relative to each other and to the blinds). All of these things combined together will give you an estimation of what the most profitable plays are (and late stage, your pushing and folding ranges in particular situations-- it should vary with opponents and not be static). Take a look at SNG wizard, and you will get a better feel for what I am talking about. Moshman's book also talks about these things. SNG Wizard also has a quiz mode where you can cycle through hand after hand answering "What would you do here?" SNG Wizard however DOES NOT train you on how to put your opponents on pushing/calling ranges-- and this is more art than science (frequency, their image, your image, tilt, fatigue, impatience, quality of hands shown, etc. all give pieces of the puzzle). SNG wizard will give you the rest. If you can put opponents on an accurate pushing/calling range, SNG Wizard will give you the best tournament equity (the most PROFITABLE) play which is different from the play which on average will give you the most chips (chip expectation and equity is distinct). Hope that helps. If you have the money, buy the book, read it, then get SNG Wiz, then play 1000 games and see if that made any difference.

[chgocubs99 0]

Weirdest bump and first post ever.

[Elo and AG 0]

Weirdest bump and first post ever.

Blah, blah, blah (SPAM) blah, blah, blah (SPAM) blah, blah, blah (SPAM)Nice FIRST post

[timwakefield 68]

Weirdest bump and first post ever.

Agreed, but it's also a great post, albeit 10 months late .Welcome TWCreborn! I hope to see you around the strat section (I'm a sng player too).