yeah top regs will finish top 3 much more then other positions. like labax said these tournaments are very top heavy and we should be aiming to take them down rather then trying to ladder up.

from your finish results it seems you need to work on your short handed and HU game. but dont discount earlier play aswell. so long as you are constantly improving you'll get there! good luck

i have 12% over 3600 and had a downswing of ~300BI

It's great that you do post game analysis and test various situations in HRC. Keep going!

Well, imo, it wouldn't have looked promissing if you were on a hot streak like you were in the 3.5r's
It should statistically go up now

Your overall ROI seems fine, don't worry about the 2.5s. It has been balanced untill now with the good 3.5r streak.

I think you're on a good path.

I don't think you quite realise how heavy the variance is in these 180's.
For a 1BI (incl 0.09 as rake/juice) your variance per game (for 7% roi) is about 25.1

If you play Y games the variance is then Y * 25.1, so for 5000 games it is:
Var(5000 games) = 5000 * 25.1 = 125500

So the standard dev = SqRt(var) = SqRt(125500) = 354

Your 7% would be centered on 0.07 * 5000 BI = 350BI
but you would expect a 'bell curve' with about 70% of times being anywhere between +/- 1 sd from this or in other words 0% to 14% with a ~70% Confidence Interval (CI).

The variance per game for a 40% roi sample is a bit higher (33.8) so for 1000 games the variance is 33800 and the sd is sqrt(33800) = 183

For this 'bell curve' it is 1000 *.4 = 400 +/- 183 (for an approx 70% confidence Interval)
or about 20% to 60% roi.

(Note this is from a frequentist stand point and a more sensible Bayesian approach would indicate that very high scores are not sustainable long term).

It takes loads of results to get any real accuracy with roi's of 180's, you can run really hot and really cold for any batch.

This is something that makes 180's difficult, you don't get good feedback from your results, many thousands are needed to spot leaks/problems just via the results.

Keep studying individual hands, or situations, icm etc. and let the results form, use good bankroll management in them.

As far as icm goes I would stick with ranges suggested from icm tools.

Icm is pretty good usually on FT, I think those that suggest simply aiming for top 3 places are wrong and they don't use the tools correctly. If you don't put opponents on sensible, thought out ranges but leave Nash ones (not often played) it can seem you should play very tight, put in the 'correct' ranges and often icm tools suggest wider push ranges to the chip ev ones.

Although sometimes the chipev with bad ranges gets close to correct icm sometimes it is well off whereas the icm using good ranges adjusts to nearly all situations.

Thanks for all your answers guys. But now i am a bit at a loss. After the first few posts i was confident in my own analysis that i put too much emphasis on ICM at the final table (especially the beginning). But the last two posters just flat out advise against that.

Lets look at a practical example. Suppose there are 9 man left. Typical FT stack distribution. 1-2 big stacks, 2-3 short , and the rest med stacks. We have 10bb utg. If you load such a situation in nash you most like will get a nash shove range of around 15% ( i dont have acces to HRC atm, but i think its close enough). But if you then look at the nash call ranges i think they are not realistic ( often too tight ). If you loosen up the call ranges for the other players to more realistic values you suddenly get a utg shoving range of around 5%. Huge difference.

Problem with playing so tight is that you most often bleed out, and ladder up a few places but end at 4-6th place. So it might be better to take direct -$E shoves who in turn become +$EV because your chances of becoming 1-3 increase.

I really dont know. Only know i now have a roi of 7% and with all the work i put in to it ( more then 2 hours study per day) i dont think that is enough. As the last poster points out i could be running really bad, and should not alter my strategy.

Are there any 180 regs with >10% ROI who can advice me?

Thanks!

The issue is that an icm calculator thinks your stack is worth $X and then if you call/shove it will be worth less than this figure so it tells you to fold . But by continually folding your stack just ends up losing value to the point where you would have been much better off just calling in the first place
The calculations are based on the assumption that you are able to realise your equity of the prizepool . If you are supposed to continually fold with a medium stack however then you soon see that the equity figure is artificially high if you are forced to play this way . So if you consider that the current equity is artificially high and actually the equity of your stack if you call and win will probably be higher than icm thinks , then that leads you to be able to take supposedly -$ev spots because you know they are better than the alternative of continually folding
I'm talking here of a situation where you are going from a "boxed in" medium stack to a bigstack who can icm abuse others

This is a really interesting insight, thanks.

Their nash ranges against your 10 bb utg shove don't look realistic as the player pool of both regs and recs are shoving wider then Nash from that spot. The recs aren't aware of correct Nash ranges from early positions and regs know you can shove a little wider then Nash if we think that a slightly minus ev shove is our best chance of maintaining Fe. Although I think at 10 be this is probably a mistake especially as player pools even reg on reg are already accounting for these wider range shoves
.

And I agree with base mental post on Variance. You shouldn't be sweating a 7 roi if you have less then a 5k sample. Even then if your playing loads of tables it's not bad. Play with an roi simulator a bit to realise how significant variance can be over surprisingly large samples. If you use a 95% confidence interval a 1K sample is just spit in the wind.

Yeah just want to kind of back up what others have said.

You don't "have" an roi of 7. You have an roi of seven over a relatively small sample when you look at some calcs.

Don't get twisted up over your roi and think all the work I've put in has only given me x% roi over this sample, just focus on improving

Hi lads,

Been inactive for a while and didnt post cuz i was playing on .es ( fishier and passive as hell so paradise on earth, id stay there forever ) mtts and did fairly good ! now that I'm back on the .it I'm playing my 27's and some 45's ( when traffic allows so ) .... I recently sticked only to 3e with 120bi and i have been shot takin at 5e with more or less 70bi.... i have been doing great the last 3-4 days, i played those 3e for a while and i can beat them, the 5e on the other hand i find the best players of the 3e plus some other good regs with some kkkkk winning on sharkscope !! So i feel the 5e its all about having a good heater between your legs since everyone is playing the same way!!

I have been playing 4-6 tables as usual combining 3e-5e 27's... to focus as much as possible on villains and the way they play their hands, most of the best regs here play 15-25 tables on autopilot and i have never been a fan of it!

My question here is what u think of my aggro brm ... i was planning moving down to 1e 27's if my roll drop to 50bi 3e ... so that 'd be 150bi 1e

Lately i have the feeling that an aggro brm, moving up and down is the best way to speed up my game and reach higher stakes, i don't need 5k tourneys to realise that I'm beating the game !

thoughts ?

^I agree. You don't need to prove beyond doubt you are beating a level to move to the next one, just move up when you are rolled for it and down when you are not.

thanks lektor... always much appreciated your help !

As tittle says im wondering what is the current average hourly for an average turbo 180 regular? I know it depends on several factors like ROI, ABI, daily volume and how long are your sessions but this is exactly what i was wondering about so to all the 180s regulars out there...

1. What is the best time to start your 180s sessions?
2. How long on average are these? 10 hours? 12 hours?
3. How many games can you play in those sessions?
4. Is best to just stick to $3r/$8 and $15 or would you recomend to load also $2s offpeak to fill volume?

Thanks in advance¡!¡

I've come to the conclusion that it's almost impossible to know what your true ROI is in 180 man sit and go's

I remember looking at a sharkscope graph of a high stakes tournament player. He's up 320k over a very large sample playing tournaments.

I noticed that back in the day he played 6000 $2.50 180 man sit and gos and was break even!!

Just be happy that your making money. Play your best, make sure that your EV bb/100 is positive, put in massive volume and pray to the baby jesus that the money goes your way over the long term

More than happy to go over hands with you, not a crusher or anything but maybe i can pick up on things you missed.

First, are those stats based on sharkscope? Is that ROI at $3r adjusted for average rebuys? I think they´re not.

Second, here´s a varianca simulation based on $2.5 ROI (7%) and sample size (5,000)



According to this variance simulation the worst peak would be at 364.8 buyins with a 99.7% confidence interval.

Small sample for sure but im more intrested to know whether or not those rebuy results are adjusted for average rebuys.

This is obviously important but i think is also important to have like a study buddy or an active skype study group and have regular live HHs review sessions with other players.

Bottom line: I think my shoving and call ranges, and icm adjustment at final 10 are pretty rock solid. I see regs with good roi's at final table making horror calls with KJs vs a 10bb MP shover ( which is absolutely fine in chip ev, but most often not so in icm mode).
At that ROI a 350 BIs downer is expected. What is not expected for a solid player with rock solid FT and ICM game to have a 7% ROI, i think you´re either running below EV on 2s and above EV on 3r and/or both.

This is also a posibility and i think this could be the case for some of the hands you´ve posted on the MTTSNG sub forum (an AQ hand comes to mind)

Especially when I have a medium stack at a final table. If you play according to icm you really cant do much in that situation. Your stack has a lot of value, and busting before shorter stacks would be bad. Of course you have to call a lot tighter under icm pressure, but you also cant shove that much. If you are in EP you can only shove TT,AQ+ for instance. While the nash shove ranges are a lot wider. But because people are calling looser then nash (only at FT, in chip ev it is just the opposite) according to HRC you have to shove a lot tighter. SO i think i might be trying to ladder up too much. While ignoring the big prizes at the end. So maybe i should shove wider there ( like nash ).

I´ve been building a sharkscope players group for just Turbos 180s for $3r, $8 and $15with just the best of the best over a decent sample size, right now this players group adds up to 77k+ games played 18% ROi and 17% ITM, note that i use a 2.35 as average rebuy to adjust rebuys results, so checking this player group results by finish position we have..

Anybody?

1) Peak times obviously which for me is 5pm-12am UK time. traffic is DEAD outside of this time

2) The longer your sessions the better because you will get more tables/games in per hour since you have to wait a while to get a full session of tables going and you obviously play fewer and fewer tables when ending your sessions. Thus the longer the better... as long as the session length does not adversely affect your performance

3) I find i play on average ten tables per hour when i play ten tables at once.

4) Play the highest game you are rolled for and then you think/have evidence suggesting that you can actually beat those stakes. Again since traffic is dead outside of peak hours you would have no choice but to play the $2.5 games

What is a big enough sample to determine if my current ROI is actually my true ROI or at least very close to it?

ive heard there are solid regs (20+ roi) who went on break even stretches for 10k games. So you can imagine you need ALOT to be sure (30k isch?) But those are extremes of course. If you dont grab a really bad (or good)piece of the "normal (gaussian) distribution" around 15k games should be enough.

I don't think you can really get much accuracy on 180 results for most people.

As a rule of thumb it takes about 12 times as many 180 results to get the same roi confidence level as 9 seat STTs. 5k games isn't all that accurate for STTs so 5 x 12 = 60k also isn't all that good for 180s!

For 30k games with about 9% rake and about 10% roi the 90% confidence interval is +/- 4.9%.

Every time you increase the number of games by 4 you half the confidence spread so CI of 30k x 4 = CI of 120k = +/- 2.45%.

The (10% roi) 90% CI of 2.5k is +/- 17.12%
so
The 90% CI of 10k is +/- 8.56%
The 90% CI of 40k is +/- 4.28%
The 90% CI of 160k is +/- 2.14%

It doesn't matter very much about the roi size the 90% CI's will still be roughly the same so for a 0% roi 90%CI for 160k games is still about +/- ~2.14, it is actually +/- 2.02%. The 90CI spread gets just a little larger as your roi goes up (because the variance raises very slightly with roi).

This shows that if you make a change to your game it would take many many, many, thousands to be confident of it being an improvement if judging only by your roi. These, and all MTTs, are quite tough games to play.

Oh and it doesn't matter over the speed, the CI will be the same for the same number of games played regardless of turbo, normal or hyper.

its even worse then i thought.and i thought it was chernobyl to begin with :P

Just move up and down depending on the level you are rolled for. It's nonsense. for example, to play 60K $0.50 180 mans before you can be confident enough move up to the $1.50 level.

I started with $300 bankroll for 45p turbo. (Have another $300 kept to insurance)
I'm not a massive grinder, so far I only played 245 games
Down $75 (50 bi)

Is this ok for a winning player? Does it happen too often?
I'm not also a well experienced mttsng player, I came from 10NL 6max cash games just to try another format of game.

This is my chipGraph so far:


Net Won Chips: 43,920
Chips EV adj: 86,961

I'm pretty ok with coolers, bad beats, etc.
I'm enjoying playing these games although I'm loosing.
So, my question is, should I continue to play and wait for further results, or just move back to cash game?

___

EDIT:

PS: Does ChipEV matters?
I'm having trouble in searching new material 'cos new games have 'antes' since first levels

Hi, yeah a 50BI downer can easily happen, even bigger downs will happen sooner or later. I think with your $300 bankroll you did a pretty good job in your calculation. 100BIs can easily happen, 200 shouldn't.
Chip EV does matter, but that will only be some indicator with a much bigger sample. Better go with bbEV\100 for now.
Overall I'd recommend you going back to NL10 cash though if you say yourself you don't play much volume @ $1.50 SNGs, then NL10 will surely be more profitable for you.

Yeah that can be just variance.

The lack of non-laughable material for MTTSNGs and MTTs in general is a reason why there is more potential in them rather than sitting at a table in NL10 where 5 out of 6 players have read the same very accurate book

But nobody can tell you if you should. Do what you enjoy most (you probably have to ship a couple of 45s and see the positive side of variance before you can answer that) and you'll find that's probably where you personally can also do best.

And +1 to mixing up the games. I played a bit of HU Razz recently which is totally different to normal and a lot of fun.

I don't use 3rd party software but I would assume chip EV is not that accurate. You may be calling too wide on the final table if it seems higher than your win rate.