Database Examination and Leak Finder
Posted by James Hudson
Video Player is loading.
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Resume Video
Start from Beginning
Watch Video
10
Loading 65 Comments...
Great video, please make more of this.
Looks like I am loosing A LOT when cold calling as well. Should I just cold call less or do some other adjustments?
HERE is a better resolution of the image.
Really nice video, hope too see more like this in future.
Loved the vid and followed along in my own HEM database. I have just recently bought HEM so it was very helpful for me. The only issue is that I was unable to find a number of the statistics used in the video. Many statistics have different names in PT4 and the HEM stat searching system is not great.
You mainly covered preflop situations. What ways do you use HEM's filtering capabilities to analyze your post flop play?
I think I found everything except for XF flop, turn, and river as PFR. What are those called?
Thanks, found them.
Yes, please, more of this.
Poker tracker 4 is so much better than Holdem manager.
Please not the PT4 > HEM2 or vice versa debate. Lots of people agree with you and lots of others (including me) disagree.
Please, please, please - can we just leave it at that and let what looks like a useful thread develop?
Nice video james. Would like to see more in the future please.
could you help me with my big blind numbers?
over 140k hands I was losing 50bb/100 from it and wasn't defending wide; defending vs steal ~35% and 3betting 5%
This month I started 3betting ~8% and defending vs steal 45 - 50% and am losing more money, its now -63bb/100, but when I filter my database none of my 3bets or flats seem to be costing me more than 100bb/100 which folding would cost me. (Im looking at this by clicking the small multi-coloured hand grid). Am I missing something? This month sample is 38k hands btw.
fwiw I adjusted my sb strategy to 3b/fold at the same time and that looks good so far as has went from -25b/100 to -7bb/100.
Any insight would be appreciated, thanks!
ah yes, that makes sense! Thanks a lot yeah I will do that.
always like these videos James...in the future tho can you please do them at a higher resolution, its really annoying not to be able to read the numbers that you are looking at (you do a good job of reading them, but its still annoying).
hah, im so dumb, I didnt even realize you could change the resolution :/ Just one more thing this video has taught me! :) sorry and thanks
Haha, no problem.
Once again, a great video from you James. I hope to see more of this format in the future. Any chance this video will be submitted to HM2 to be placed on the startup page of the program?
Great stuff, thanks! fwiw I think hm2 cold call stat doesn't include hero overcalls (you talked about how wtsd from blinds shouldn't be higher than other positions at around 29min partly due to more multiway pots)
Probably me being dumb but i cant find the check fold flop as pfr you answered " I believe those are skip cbet as pfr and check fold oop." but i still cant find them. Thanks a lot
ok thanks a lot this video relly helped me i have hardly ever used HM2 to go over my game finding massive leaks all over the place !! Once again thanks
Very nice review. I like how you suggest how to fix those leaks. Will you be making more videos like this in the near future?
Wow, fantastic video. I can see this content being super useful to a lot of people, and I'm one of em! A+ content, and A+ video production quality as well. Your preparation prior to the video shows up. And those RIO headphones are coming up big on the audio quality!! Haha.
Really good video.
Cbet average : 52%
x/fold as pfr : 86%
Flop cbet success : 41%
Vilain is printing money when i don't cbet. How can i fix that ? Just watch all my missed cbet and answer to this question : could I cbet here ? could I x/raise this flop ? x/call ?
Yes, maybe more bluffs and more x/call with strong and mid-strong hand and let vilain bluff
I have a 320k samplesize for NL50 Zoom...
My CC EV bb/100 are
SB: 23,7 (2693 hands)
BB: -23 (4290)
MP: 136 (710)
CO: 27 (1140)
BTN: 68 (2259)
The WTSD% you mentioned later in the Video... I have the same percentages for SB and BB, and they are also 2% higher than average, so not sure if that is the reason why his winrate is so low there!
My 3b bluffs as you filtered it...
SB: 97 (1865 hands)
BB: 116 (1793)
MP: 340 (99)
CO: 146 (313)
BTN: 56 (1063)
Called Preflop 4b turns from -918 EV bb/100 when folding to 392 EV bb/100 (275 hands) when calling
My called 3b stats from SB and EP are horrible i guess!!!
SB: -484 (174 hands)
EP: -258 (568)
MP: -206 (528)
CO: -81 (472)
BTN: -37 (627)
This video explains really well the concept of cold calling and 3-betting for value. I had to be careful not to put the conclusion on hero on my own game. This video does teach you how to analyse your HEM, which is very helpful.
I for sure need to add more valuehands in my 3-betting range and lose bluffs. Also, cold calling less in MP, CO, BTN will help my winrate.
Thanks man, great video!!
My 3b succes is at 53%, is that still low?
Only 125 hands, but I am losing on the BTN with cold calling. How can I keep an eye on that since it's a small sample size but might be a problem?
About opening wider from the SB: Considering the high rake that you need to pay to play a pot, wouldn't it be better to not make this range to wide?
Aurelius - I tend to open very wide from the sb mostly because BB's tend to over fold at the stakes I play ($100nl and lower) and you only have 1 player to worry about. If there is no flop there is no rake! Below $100nl I used a 2.5x open from the SB which, if my math is right, if villain folds 57% or higher you can literally open any 2 and just muck everytime you are called or 3b and still turn a profit. The math: Risk/(risk+reward) {Risk=2 Reward=1.5} 2/(2+1.5) = .5714 or 57.14%. If it works this often you break even w/ any 2, if it works more you turn a profit w/ any 2.
You can flip this around when you are in the BB to see how often you need to defend by taking villains risk/(risk+reward) and then take the answer (1-a) and the resulting number is how often you need to defend at min. to prevent him from profitably opening any 2.
When you are looking at a BTN open it is a little different because the SB and BB combined need to defend enough to not let BTN profitably open any 2. When you are on the btn you can take your risk to reward and see how often both players need to fold to make opening any 2 profitable. I would think it is rare these days that both players would be folding enough to make opening any 2 on the btn profitable but it does still occasionally happen.
And if my math is wrong I really hope someone will correct me (they probably will) but I am almost 100% sure it is right.
Who is Lefort?
What is a good winrate for the SB, and BB?
Somewhere in this video you are analyzing how the player is doing vs 3 bets. You are saying that if he is opening to 3bb and his winrate is > -300bb after calling a 3bet he can start to call even more. However once you have been 3 bet it is a new situation which should be analyzed from scratch. So folding after being 3bet has an ev of 0bb thats why his winrate in this spot should be positive no?
17:50 - You saw his river call eff. was low and immediately assumed he was CALLING too much. River eff defined is $$spent/$$won so couldn't this also mean he might be over FOLDING?
15 min in - When looking at the cold-call filer how do you determine that if HERO was to get he's cold-calling EV up 0 he would gain +~2 bb/100 in EV overall. How did came to that conclusion?
Hi James,
First of all, great video and it showed me many leaks in my game.
I got some questions regarding the database analysis and comparing plays by EV.
In your analysis you said, that when we open the Button to 2bb and get 3bet, we are looking to lose with our defends against 3bets less than -200bb/100.
Does this apply to other situations, too? Because usually people say the EV of folding is 0. But in your analysis the EV of folding would be -2bb right?
So lets say, our defends against 3bets (either by calling or 4betting increase our winrate to -100bb/100. I wanted to compare this to defends against tight 4betting ranges.
For example, CO opens to 2.5bb and we 3bet AKo in the BB to 9bb. Now we are getting 4bet from a 4bettingrange in the CO of 2.6%(samplesize 10k+hands) to 20bb.Overall 4bettingrange was 3.6%.
Basicly, all his 4bets are for value. (although villians fold to 5bet is 37% over 14 samples, but never in the CO or Btn)
My question is now, when I shove the EV should be(I hope the calculation is correct, assuming 100bb stacks)
EV=38,82%*200,5bb-61,18%*200,5bb
EV=-38,7bb
When I compare the EV(shove) of AK against villians 4betting range(here 2.6%) it is at -38,7bb. Which equals to -3870bb/100, instead of just folding to the 4bet which would be -900bb/100.
Can you compare those results? Or is my thinking wrong?
I would appreciate your help.
Thanks
By my calculations you should lose about 12.8bb by jamming in this situation so you should probably go ahead and call if villain is strictly 4 betting for value.
Your calculations are a little off. 38.8% of the time we're going to win 110.5bb (the money that we've already invested in the pot via the three bet, villain's stack and the blinds) and 61.2% of the time we lose 91$ which is the amount that we're risking seen as the 3 bet chips are already in the pot and no longer ours to risk.
0.388x 110.5= 42.874
0.612x -91= -55.69
42.874-55.69= -12.8
I hope that clarifies things a little.
Thanks for your help with the EV calculation. -39bb seemed a little bit off :)
So, if his 4bet range wouldn't be so tight and our EV of shoving gets to lets say -5bb instead of -12.8bb and assuming EV(call) is less than -5bb/100(for example EV-call is -7bb), a shove would be better than a call or fold? Because EV(fold) would be the highest with 0 EV, I should fold. Correct?
About -200bb/100 because I got 3bet, you wouldn't compare this to 3betting to 9bb and folding in the BB, or are those winrates comparable?
I just mean, that if I 3bet to 9bb and get 4bet I should fold,call or shove which ever is more +EV. And I don't include the -9bb from my 3bet in the calculation? Then lets say in my example above folding is 0 EV(instead of -9bb) shoving would be -12,8bb and calling is maybe +5bb. I am choosing the most EV play and call. Is that correct?
Ok, thank you very much. That made it more clear. :)
Hey James, This is a great video and one that has helped me identify and plug several leaks. What do you think good winrates should look like when open raising the sb and when defending the bb vs a sb steal. My samples are still too small to be relevant so I'm just trying to anticipate future holes in my play. Thanks for your time.
Stuff like this is going to depend a ton on the frequencies you use and how your opponents play. Try comparing your results in these different spots to regulars who are doing well in your games. If you don't have enough hands on a specific regular then combine a few regulars into an alias and compare your numbers vs their cumulative stats.
will be great if you could do a new vid but talking about tournaments which could be the +ev stats, guess will help a lot :)
Sorry, I'm not making videos anymore but there's tons of other pros who've made great tournament videos. Maybe try checking out the learning paths for stats related tournament videos.
Excellent video. Can anyone point me to a discussion on how each of the metrics are calculated?
Be the first to add a comment
You must upgrade your account to leave a comment.