5NL analysis session, could use some help

After a busy 2 weeks I am starting to dig into my stats on 5NL (45k hands played with poor results, loss of -1.65 bb/100 which is EV adjusted -0.65 bb/100) and trying to figure out what works and what doesn't, what needs fixing or a new approach etc.

I am especially looking into 17k hands played over the last week, which have been utter carnage (-6 bb/100). These came after I made some adjustments that were supposed to be positive but turned out badly, although I can't tell if I am just running really horribly or this is a systemic problem.

Anyone who beat up that level up to looking over these hands and giving me a few pointers? I am interested in big pots of course but overall play is probably in need of a re-haul.

So far stats look pretty normal except that my positional awareness seems to be too high (!). I am playing on FT and can't really post hands in HH forum since it appears none of the hand converters work with the new format.

I am looking at my opponents that have over 1k hands and are winners and mostly they just seem to be running obscenely well (incredible set% stats etc.)...

without some stats and a graph it's hard to tell what's good/wrong in your game. You said that you made some adjustments. Well maybe you searched for the wrong articles for you. I mean you can't read high stakes articles and apply them to your 5nl game, or maybe you did not understood them, ooor maybe you are on a simple donwswing/break even.

Adjustments were mostly a bit more stealing, exploiting preflop tendencies of other players etc. This increased the VPIP/PFR slightly and the positional awareness went up because of it.

Stats don't really say that much by themselves since a lot of things end up hidden. HEM calls my stats normal, and I am slightly losing.

I need someone to look at the hands and see what kind of systemic mistakes I am making.

Edit: OK, fair enough, I do think I am running like crap but that is a dangerous belief. I do worry that the stats might be "tainted" by the "runbad" thing if it is real (!), but how can I be sure? I can't wait another 50k hands to get a larger sample while bleeding money.

P.S. Why is the forum refusing to make paragraphs when I use Firefox, but makes them when I use Chrome?! WTF

May I suggest you record a session of your playing so we might see it and comment on it, we might see things there that we don't see by looking at stats. Also post your stats for the stats guys though obv.

This is basically the same as reviewing the hands from the DB, except for the hassle of making a movie (which I don't know how to do ) and then everybody needs to strain their eyes watching it unless I pick just one table.

I finished some early scans, so far it seems the problem might be on the turn but I am still trying to figure it out.

Turn cbet% is 29.8 but still the success is only 19.4 (!!!). Either people are massively floating me to the river or I am running REALLY ****ing horrific.

I did finish the All In analysis of the last 17k hands:
All In Preflop = -10.32$
EV difference = 12.99$

All In on Flop = -5.29$
EV difference = 7.84$

All In on Turn = 11.83$
EV difference = 3.47$

All In on River = 23.15$

Well, I suppose this is pretty funny doesn't matter WHEN the money goes in, there will always be suckouts. But it seems overall I am not making horrible mistakes when getting the money in (I try to keep track of criteria players use for stackoffs and try not to get it in too lightly, not that it helps).

All In at any point = 19.37$
EV difference = 24.30$

Total loss is -54.94$, so after adjustment there is still -30.64$ unaccounted for.

LD1977 said:

This is basically the same as reviewing the hands from the DB.

Click to expand...

No it's not. We'll be able to see what you see on their HUD, see real time how you make decisions, and what those decisions are. If you talk at the same time you can even tell us why. We might see things you are doing that you aren't aware of. You can't expect us to trawl through your db looking for those loose -EV calls in the SB for example, but we'll see them clear enough if you do it on video

LD1977 said:

hassle of making a movie (which I don't know how to do )

Click to expand...

Camtasia.

LD1977 said:

everybody needs to strain their eyes watching it unless I pick just one table

Click to expand...

I managed to put out a little 10NL video with 6 tables running, and if you full screen it and put it in HD you can see it fine.


Up to you though, I was just trying to help.

Dunno, so many decisions are situational... looking at hands in the DB also gives the HUD stats, it is just that my HUD is probably different from yours etc. (this would be a problem in the video too).

Anyway:

1) The lack of turn folds even though I have severely reduced frequency of cbets seems to be mostly incorrect calling from villains and of course there are some turn raises (a very popular move with nut type hands!). Since my AI stats are more or less OK it seems I have a fair grasp on relative hand strengths BUT I will need to look into this deeper after I clear out the bigger stuff.
*** There were a LOT of suckouts where I felt that the turn call was mathematically incorrect... maybe I am wrong or maybe I am running really badly.

2) Light 3betting IP does NOT work and I need to reduce it.

3) Cold calling preflop needs severe reduction, although stats seem to be tainted.

3a) I am losing money while set mining, main reason being:
- I am hitting sets less than 10% of the time (normal is almost 12%)
- I am getting raped when I actually do hit sets (!!!). Bad beats galore (some truly incredible stuff there).
*** This seems to be the "runbad" thing since the set mining calls are made with proper stack size implied odds. Paradoxically that is also the reason why the losses are so heavy!

4) I call too much in both blinds, 3betting is profitable so probably I need to find a better balance between 3betting and folding.

5) Calling donk bets IP is losing me money. Raising also loses but at much lower rate. It seems on average when I face a donk bet the EV is negative and there is not much to be done except try to minimize the bleeding.

LD1977 said:

Dunno, so many decisions are situational...

Click to expand...

Exactly.

LD1977 said:

looking at hands in the DB also gives the HUD stats

Click to expand...

Yes, so are you going to make a video going through hands post session? Fine. Although still not what I was going for. When I had someone on here sweating me, he pointed out I had a timing tell that was easily exploited by any regs paying attention. That's not something we can see by looking at your DB. Just saying, there will be things you haven't considered that might be obvious to people on here, and even to you when you play it back and think, oh... I shouldn't have done that.

LD1977 said:

it is just that my HUD is probably different from yours etc. (this would be a problem in the video too).

Click to expand...

Then just tell us what it is.

Listen, I know some people don't like making vids for whatever reason, so that's cool. I was just thinking it might be helpful for you.

LD1977 said:

1) The lack of turn folds even though I have severely reduced frequency of cbets seems to be mostly incorrect calling from villains and of course there are some turn raises (a very popular move with nut type hands!). Since my AI stats are more or less OK it seems I have a fair grasp on relative hand strengths BUT I will need to look into this deeper after I clear out the bigger stuff.
*** There were a LOT of suckouts where I felt that the turn call was mathematically incorrect... maybe I am wrong or maybe I am running really badly.

Click to expand...

Presumably you are only value betting the turn if your cbet turn is v low. So, I would assume you are just running into better hands I guess. When you talk about turn cbet 'working' I guess you mean getting folds. If your turn cbet is value based, then calls is what you'll be wanting, not folds.

LD1977 said:

2) Light 3betting IP does NOT work and I need to reduce it.

Click to expand...

What sort of hands are you talking about when you say light? AQ/AJs/KQ or A2s/K9s/98s or something else? When you say it doesn't work, I suppose you mean they are not folding pre or post. You might want to make sure you are 3betting wider for value and lop off your bluffs vs villains who don't fold.

LD1977 said:

3) Cold calling preflop needs severe reduction, although stats seem to be tainted.

3a) I am losing money while set mining, main reason being:
- I am hitting sets less than 10% of the time (normal is almost 12%)
- I am getting raped when I actually do hit sets (!!!). Bad beats galore (some truly incredible stuff there).
*** This seems to be the "runbad" thing since the set mining calls are made with proper stack size implied odds. Paradoxically that is also the reason why the losses are so heavy!

Click to expand...

Play more hands. Also, 10% and 12% are almost the same, stop worrying about that kind of detail.

LD1977 said:

4) I call too much in BB, 3betting is profitable so probably I need to find a better balance between 3betting and folding BB.

Click to expand...

Yes, probably.

LD1977 said:

5) Calling donk bets IP is losing me money. Raising also loses but at much lower rate. It seems on average when I face a donk bet the EV is negative and there is not much to be done except try to minimize the bleeding.

Click to expand...

Work out what kind of hands they are donking with and act accordingly. Make notes.

I am not sure how much you can tell from the timing when someone is multitabling. He might appear to "think" but he is actually spending time on another table and on the table you are watching he is making instant decision. I would never rely on something like that.

I heavily reduced turn cbets (since people don't really fold much if they have ANY equity) but I still do some bluffs there on scare cards etc. Mostly I bet for value and usually I am still ahead on the turn (unless I get checkraised, and sometimes even then) but rivers hate and despise me for some reason. It is pissing me off since I believe that these people are calling with bad prices and yet everybody seems to be in God mode and these are all big pots!

Light 3betting is something I do vs people who fold to 3bets a lot and those who raise light themselves and DON'T fold to 3bets. "3bet light" is a filter in HEM and it seems to include everything except 99+, AQo+. I didn't get deeply into the problem but I suppose there are some kicker issues there and of course the "run like crap" thing is possible but I really can't say now.

Donk bets seem to be divided by sizes (observation while playing):
1) weak made hands, weak PPs - small bets (up to half pot), fold to raises a lot (those that don't I take notes of)
2) draws (OESD, FD) - usually 2/3 pot+, do not fold to raises much (depends on quality of player, better ones seem to recognize pot odds and implied odds as a concept)
3) monsters - these vary a lot (from very small to overbets), can't really categorize these... usually sets, made straights, pocket AA (although this is usually an AI check raise) etc.

Timing tells can be crucial. Especially when you are playing against heavy multi-tablers. I just ordered myself a headset 5 minutes ago and I'm going to see if I can get a sweat going, I advise you do the same. There is so much more that can be picked up from a session where you explain what you are doing and why right at the moment you are doing it.

That would be great but I dunno how to go about it.

- I would need someone who is much better and willing to watch me play?
- Also I am not sure how can he see my HUD, or I have to describe what I am seeing? That might be difficult to do unless playing just a few tables and then the samples are too small.

Well I played like crap today... really a few super stupid plays.

Good thing is that I noticed a player who seems to be doing really well in 5NL so I have a chance to see his playstyle.

Whats your name online? I play 5nl and might have seen you, or have notes on you that I could maybe point out?

Sure, I play on FT though and you seem to be PS?

OK so after doing manual analysis I have tried Leak Buster to see what it can say about my game.

Overall Grade = B
Total Score = 231

This seems decent enough for the first run. I have a lot of A/A+ grades in preflop play could use better balance from EP to LP (I am slightly too LP heavy).

Leaks found: 86
Critical: 5
Very Important: 5
Important: 8

Since I am losing at around -2 bb/100 clip this is not too surprising, but I did hope there will be one humongous leak that could be fixed and that would be enough. Oh well.