This comment has intrigued me, variance effect in hands your not involved in. I suppose this happens in DYMs and MTTs more often than I though.
Which leads me to, I forget what it was leading me to, but I'll edit it in when I remember.
I play a few PLO8 DYMs and it's even worse in those, often two players are all in and you're just praying for one of them to scoop. It often ends in a split pot though and you're back to square one. At least with Holdem you rarely get split pots.
To balance it out though I'm sure we've all played MTTs where we get to the FT as a short stack and then someone goes on a rampage and starts knocking out the others left, right and centre and we end up getting second rather than busting in 6th all due to their hard work.
I'll quite often type a 'thanks' in the chat box when someone helps me ladder a few places.
Oh, and do come back when you remember what you forgot in the first post!
Well worth a read for anyone still awake after my posts.
In my own convoluted way that's what I was trying to get across with the whole thread. Imagine the amount of typing you would have saved me if you had of started a thread with that link.
Basically even a top pro with a 80% ROI will lose money over a 1000 MTT sample 13% of the time. That's a bit of an eye opener isn't it?
It works the other way too obviously. Someone could sun run for a couple thousand tournaments, running at 75% ROI, and he and everybody else think he's the don but actually he's a -25% ROI bleeder.
Mind you, since that article, games have gotten a lot tougher and everyone knows nash ranges, so, variance will be a bit closer to some sort of equilibrium and results won't be quite as skewed as suggested.
It works the other way too obviously. Someone could sun run for a couple thousand tournaments, running at 75% ROI, and he and everybody else think he's the don but actually he's a -25% ROI bleeder.
Mind you, since that article, 1. games have gotten a lot tougher and 2. everyone knows nash ranges, so, 3. variance will be a bit closer to some sort of equilibrium and results won't be quite as skewed as suggested.
"Basically even a top pro with a 80% ROI will lose money over a 1000 MTT sample 13% of the time. "
Amazing stat.
If only Team Tin Hat had the sense to realise that, eh?
True, but only to a point. The (considerably) smaller fields on Sky would mean that a top pro (or even MattBates) would be less than 13% probable to make a loss.
"Basically even a top pro with a 80% ROI will lose money over a 1000 MTT sample 13% of the time. "
Amazing stat.
If only Team Tin Hat had the sense to realise that, eh?
True, but only to a point. The (considerably) smaller fields on Sky would mean that a top pro (or even MattBates) would be less than 13% probable to make a loss.
You also need to factor in the fact that MattBates doesn't suffer the same variance as normal poker players as he runs like God.
"Basically even a top pro with a 80% ROI will lose money over a 1000 MTT sample 13% of the time. "
Amazing stat.
If only Team Tin Hat had the sense to realise that, eh?
True, but only to a point. The (considerably) smaller fields on Sky would mean that a top pro (or even MattBates) would be less than 13% probable to make a loss.
You also need to factor in the fact that MattBates doesn't suffer the same variance as normal poker players as he runs like God.
Matt doesn't really run particularly well. He just wants people to think that, as it gives him an (extra) edge. He suffers from variance like anyone else.
Although the thread title and reason for posting is to do with variance, the tin-hat brigade, when they suffer from variance, question the integrity of the RNG and its randomness, when the outcome doesn't suit their expectations , they cry it's rigged.
This sequence..
12,12,12,12,12,12,12,12,12
No one can say with any certainty that the sequence is or isn't random.
"Basically even a top pro with a 80% ROI will lose money over a 1000 MTT sample 13% of the time. "
Amazing stat.
If only Team Tin Hat had the sense to realise that, eh?
True, but only to a point. The (considerably) smaller fields on Sky would mean that a top pro (or even MattBates) would be less than 13% probable to make a loss.
You also need to factor in the fact that MattBates doesn't suffer the same variance as normal poker players as he runs like God.
Matt doesn't really run particularly well. He just wants people to think that, as it gives him an (extra) edge. He suffers from variance like anyone else.
It was the worst 10 minutes of his life....
Pretty nasty of you to remind me of that awful time on a public forum!
There’s also variance around the starting hands you receive, better to accept and move through it and concentrate on making the best decisions possible, the ones that affect your mean performance
These are horrid stats, you should win approximately 80% of the hands with pair over pair AIP and I won just over 50%, even worse is the underpair stat. I haven’t won any out of 17 hands, now I know that I should only win about 20% but I haven’t won a single one since Mid October!
Dominating hand v dominated hand (sample size 58) win rate 77.6% (expected 70%) Dominated hand v dominating hand (sample size 39) win rate 23.1% (Expected 30%)
Now this surprises me as I expected to run badly here as I dread it every time my opponent turns over a dominated hand however I am a little over ev here as I should win about 70%. Interestingly it is balanced out by running slightly below ev when dominated.
Pocket pair v one overcard (sample size 40) win rate 65% (expected rate 70%) One overcard v pocket pair (sample size 24) win rate 25% (expected rate 30%)
Running above ev here but an extremely small sample size in the undercard v overcards so winning 3 wins out of 5 in that match up skews the figures a bit.
1st and 3rd cards against 2nd and 4th (AQ v KJ etc) (sample size 20) win rate 50% (expected rate 60%) 2nd and 4th against 1st and 3rd (sample size 13) win rate 15.4% (expected rate 40%)
Quite a way behind expected wins here, but sample sizes not huge.
1st and 4th against 2nd and 3rd (sample size 37) win rate 67.6% (expected rate 58%) 2nd and 3rd against 1st and 4th (sample size 17) win rate 23.5% (expected rate 42%)
Slightly ahead on the first match up but quite a way behind on the reverse fixture
Pocket Pair v dominated hand (AA v A6 etc) (sample size 13) win rate 92.3% (expected 86%) Dominated hand v pocket pair (sample pair 8) win rate 12.5% (expected 14%)
Not too far off expected given the small sample size
Also (for those of you that have added up the total number and found it 11 short there were 11 ties at the river, these were mostly hands like A3 v A2 etc which are technically dominating but result in a high percentage of ties)
Hi @Enut I hope you take this the right way. The variance thing is just as it is, I do not have a strong opinion on it. However I do notice how often you are getting it in ahead.
This may sound strange, but if I can be so bold to say, you are getting it in ahead "imo" way too much
Getting it in ahead as often as you do according to your stats, means that you must be folding or just calling too much pre. I would say getting it in ahead 50%-55% of the time is great, but more than that and it means you are folding too much or just calling when you should be 3bet jamming. Even if you got it in ahead only 50% of the time, it would still be +ev because of the uncontested pots you win when you 3bet shove pre.
Of course, "waiting" and picking off overly aggressive opponents with hands that you feel may dominate them is something we all should be trying to do, but not at the expense of passing other good opportunities.
Lets just say, the effective stack is between 15BB's and 20BB's, and "hypothetically" that you always know if they call, it's going be a flip, and will work out even long term. You should in that case 3bet jam all the time as whenever they fold you make a profit. Also bounties make that play even more +ev if you have them covered.
I would recommend widening your calling range of all ins and 3 bet jamming wider with the right stack. You will take down more uncontested pots and won't be folding the best hand so often.
If you try to get it in ahead, all the time, you are playing too passively.
Just to put another perspective on it.
Hope this makes sense.
Cheers,
Graham.
P.S. I have highlighted in bold were the stats are, "too good" if you catch my drift.
These are horrid stats, you should win approximately 80% of the hands with pair over pair AIP and I won just over 50%, even worse is the underpair stat. I haven’t won any out of 17 hands, now I know that I should only win about 20% but I haven’t won a single one since Mid October!
Dominating hand v dominated hand (sample size 58) win rate 77.6% (expected 70%) Dominated hand v dominating hand (sample size 39) win rate 23.1% (Expected 30%)
Now this surprises me as I expected to run badly here as I dread it every time my opponent turns over a dominated hand however I am a little over ev here as I should win about 70%. Interestingly it is balanced out by running slightly below ev when dominated.
Pocket pair v one overcard (sample size 40) win rate 65% (expected rate 70%) One overcard v pocket pair (sample size 24) win rate 25% (expected rate 30%)
Running above ev here but an extremely small sample size in the undercard v overcards so winning 3 wins out of 5 in that match up skews the figures a bit.
1st and 3rd cards against 2nd and 4th (AQ v KJ etc) (sample size 20) win rate 50% (expected rate 60%) 2nd and 4th against 1st and 3rd (sample size 13) win rate 15.4% (expected rate 40%)
Quite a way behind expected wins here, but sample sizes not huge.
1st and 4th against 2nd and 3rd (sample size 37) win rate 67.6% (expected rate 58%) 2nd and 3rd against 1st and 4th (sample size 17) win rate 23.5% (expected rate 42%)
Slightly ahead on the first match up but quite a way behind on the reverse fixture
Pocket Pair v dominated hand (AA v A6 etc) (sample size 13) win rate 92.3% (expected 86%) Dominated hand v pocket pair (sample pair 8) win rate 12.5% (expected 14%)
Not too far off expected given the small sample size
Also (for those of you that have added up the total number and found it 11 short there were 11 ties at the river, these were mostly hands like A3 v A2 etc which are technically dominating but result in a high percentage of ties)
Hi @Enut I hope you take this the right way. The variance thing is just as it is, I do not have a strong opinion on it. However I do notice how often you are getting it in ahead.
This may sound strange, but if I can be so bold to say, you are getting it in ahead "imo" way too much
Getting it in ahead as often as you do according to your stats, means that you must be folding or just calling too much pre. I would say getting it in ahead 50%-55% of the time is great, but more than that and it means you are folding too much or just calling when you should be 3bet jamming. Even if you got it in ahead only 50% of the time, it would still be +ev because of the uncontested pots you win when you 3bet shove pre.
Of course, "waiting" and picking off overly aggressive opponents with hands that you feel may dominate them is something we all should be trying to do, but not at the expense of passing other good opportunities.
Lets just say, the effective stack is between 15BB's and 20BB's, and "hypothetically" that you always know if they call, it's going be a flip, and will work out even long term. You should in that case 3bet jam all the time as whenever they fold you make a profit. Also bounties make that play even more +ev if you have them covered.
I would recommend widening your calling range of all ins and 3 bet jamming wider with the right stack. You will take down more uncontested pots and won't be folding the best hand so often.
If you try to get it in ahead, all the time, you are playing too passively.
Just to put another perspective on it.
Hope this makes sense.
Cheers,
Graham.
P.S. I have highlighted in bold were the stats are, "too good" if you catch my drift.
I agree with you 100% and had noticed the same myself. I think that I fold small pocket pairs and suited connectors for example to shoves too much, probably because I think that I am, at best, in a flip with the PPs and almost certainly behind with suited connectors. I am probably not factoring in the money already in the pot and the opponents whole ranges as much as I should be.
I am certainly trying to widen my pre flop calling/jamming range a bit and it seemed to work last night as I managed to take 9 bounties in a 77 player MTT. I still went out before the money which was disappointing but I had fun getting the bounties! I have posted a hand in the tourney section if you fancy taking a look.
I watched a little of the same tourney after I was out and when the bubble had burst I noticed a player in the bb call an early position shove for 12 bbs with J10, he had 13 bbs so could win the bounty but certainly I would never have made that call (approximate details given as I don't have the HH). His SS rating is VERY high though so I guess his call was correct and his thinking was that winning that hand would give him a bounty AND a much better chance of winning the tourney. He was behind but did win the hand btw.
I probably wouldn't call a 13BB effective shove myself with the JT that often, however I may well 3bet jam with it against a button or small blind raise, if the opener is someone who folds to 3bets a lot. I am certainly re-shoving all my pocket pairs.
Most the time you'd take the pot down and if called you are probably 40%-45% against his range, which is plenty when you factor in the uncontested pots you win. Obviously this works even better if there are anties, but I wouldn't let the lack of anties put you off too much.
So it would be good to start re-shoving wider over a raise, although do keep an eye out for those that may min/Raise to induce when they have a monster.
Fold equity is key, which clearly you have when you re-shove over a raise, and don't have when you call an all in.
I am guessing you probably also need to be straight shoving 12 BB's or less wider than you currently do.
It looks "to me" like you need to widen both, but, I would suggest, particularly your re-shove range, especially against aggressive openers that fold to 3bets a lot. Clearly there is ICM to consider, but with the Sky structure (money mainly up top) and bounties "usually" it usually encourages the more aggressive play too.
Comments
This comment has intrigued me, variance effect in hands your not involved in. I suppose this happens in DYMs and MTTs more often than I though.
Which leads me to, I forget what it was leading me to, but I'll edit it in when I remember.
I play a few PLO8 DYMs and it's even worse in those, often two players are all in and you're just praying for one of them to scoop. It often ends in a split pot though and you're back to square one. At least with Holdem you rarely get split pots.
To balance it out though I'm sure we've all played MTTs where we get to the FT as a short stack and then someone goes on a rampage and starts knocking out the others left, right and centre and we end up getting second rather than busting in 6th all due to their hard work.
I'll quite often type a 'thanks' in the chat box when someone helps me ladder a few places.
Oh, and do come back when you remember what you forgot in the first post!
Well worth a read for anyone still awake after my posts.
In my own convoluted way that's what I was trying to get across with the whole thread. Imagine the amount of typing you would have saved me if you had of started a thread with that link.
Basically even a top pro with a 80% ROI will lose money over a 1000 MTT sample 13% of the time. That's a bit of an eye opener isn't it?
"Basically even a top pro with a 80% ROI will lose money over a 1000 MTT sample 13% of the time. "
Amazing stat.
If only Team Tin Hat had the sense to realise that, eh?
It works the other way too obviously. Someone could sun run for a couple thousand tournaments, running at 75% ROI, and he and everybody else think he's the don but actually he's a -25% ROI bleeder.
Mind you, since that article, games have gotten a lot tougher and everyone knows nash ranges, so, variance will be a bit closer to some sort of equilibrium and results won't be quite as skewed as suggested.
The (considerably) smaller fields on Sky would mean that a top pro (or even MattBates) would be less than 13% probable to make a loss.
He suffers from variance like anyone else.
It was the worst 10 minutes of his life....
This sequence..
12,12,12,12,12,12,12,12,12
No one can say with any certainty that the sequence is or isn't random.
This may sound strange, but if I can be so bold to say, you are getting it in ahead "imo" way too much
Getting it in ahead as often as you do according to your stats, means that you must be folding or just calling too much pre. I would say getting it in ahead 50%-55% of the time is great, but more than that and it means you are folding too much or just calling when you should be 3bet jamming. Even if you got it in ahead only 50% of the time, it would still be +ev because of the uncontested pots you win when you 3bet shove pre.
Of course, "waiting" and picking off overly aggressive opponents with hands that you feel may dominate them is something we all should be trying to do, but not at the expense of passing other good opportunities.
Lets just say, the effective stack is between 15BB's and 20BB's, and "hypothetically" that you always know if they call, it's going be a flip, and will work out even long term. You should in that case 3bet jam all the time as whenever they fold you make a profit. Also bounties make that play even more +ev if you have them covered.
I would recommend widening your calling range of all ins and 3 bet jamming wider with the right stack. You will take down more uncontested pots and won't be folding the best hand so often.
If you try to get it in ahead, all the time, you are playing too passively.
Just to put another perspective on it.
Hope this makes sense.
Cheers,
Graham.
P.S. I have highlighted in bold were the stats are, "too good" if you catch my drift.
It recommends that the optimal range of winning when it goes to showdown, including all ins pre, is 49%-54.9%
If you are winning more than 55% it's a sign you are playing too passively.
From your figures, the number of times you are getting it in good, with normal variance would lead to a figure greater than 55%.
Cheers,
G
Hi Graham and thanks so much for posting.
I agree with you 100% and had noticed the same myself. I think that I fold small pocket pairs and suited connectors for example to shoves too much, probably because I think that I am, at best, in a flip with the PPs and almost certainly behind with suited connectors. I am probably not factoring in the money already in the pot and the opponents whole ranges as much as I should be.
I am certainly trying to widen my pre flop calling/jamming range a bit and it seemed to work last night as I managed to take 9 bounties in a 77 player MTT. I still went out before the money which was disappointing but I had fun getting the bounties! I have posted a hand in the tourney section if you fancy taking a look.
I watched a little of the same tourney after I was out and when the bubble had burst I noticed a player in the bb call an early position shove for 12 bbs with J10, he had 13 bbs so could win the bounty but certainly I would never have made that call (approximate details given as I don't have the HH). His SS rating is VERY high though so I guess his call was correct and his thinking was that winning that hand would give him a bounty AND a much better chance of winning the tourney. He was behind but did win the hand btw.
I probably wouldn't call a 13BB effective shove myself with the JT that often, however I may well 3bet jam with it against a button or small blind raise, if the opener is someone who folds to 3bets a lot. I am certainly re-shoving all my pocket pairs.
Most the time you'd take the pot down and if called you are probably 40%-45% against his range, which is plenty when you factor in the uncontested pots you win. Obviously this works even better if there are anties, but I wouldn't let the lack of anties put you off too much.
So it would be good to start re-shoving wider over a raise, although do keep an eye out for those that may min/Raise to induce when they have a monster.
Fold equity is key, which clearly you have when you re-shove over a raise, and don't have when you call an all in.
I am guessing you probably also need to be straight shoving 12 BB's or less wider than you currently do.
It looks "to me" like you need to widen both, but, I would suggest, particularly your re-shove range, especially against aggressive openers that fold to 3bets a lot. Clearly there is ICM to consider, but with the Sky structure (money mainly up top) and bounties "usually" it usually encourages the more aggressive play too.
Hope this helps.
Cheers,
Graham.