My 10NL experience (August, 54k hands) - systemic mistakes or just "variance"?

My 10NL experience (August, 54k hands) - systemic mistakes or just "variance"?

I have moved up to 10NL at start of month and after 54k hands played these are the results:

Total = -46.64$ (-0.86 bb/100)
EV diff. = +86.49$ (+1.59 bb/100)
Total adjusted = +39.85$ (+0.73 bb/100)

I am more-less satisfied with the results since I consider the 1st month to be an adjustment period, especially since:
- getting continually coolered is not fun (I did spew some due to frustration but not that much)
- I had grinded out a promotional bonus + I have 15% rakeback so overall my bankroll actually went a bit up

Now, there is a big systemic issue that I have identified and have now dug into the database to see what exactly was going on.

They say the picture is more eloquent than words, so... look at the bottom

The order is by size of pot, so you can imagine the carnage (some of the big pots are 150-200 bb, and these small ones on the bottom that I won are like 50bb) and the insane impact on the win rate this scenario has.

I excluded:
- hands which didn't actually come to showdown (I folded one bottom set to an almost certain overset)
- flopped quads vs my turned set/FH (there went another buyin )
- rivered set over set situations since most of the money goes in by the turn and river suckouts are not strategically relevant

So, this raises some questions:
1) Is this a normal frequency for set over set situations (18 over 54k hands)?
2) Are low sets really that freaking terrible? Should I just dump that crap out of my ranges and refuse to float flops even when I am getting a great price (or fold turned sets to a turn shove)? Should I just pot control/call down with them if the opponent is showing aggression?
3) Should I just shove flops with such small sets to avoid getting destroyed by turned bigger sets? I can't imagine that is really optimal but I am seriously unsure WTF is going on here.

Strategic impact of this is that set mining as a strategy is completely non-viable regardless of price, position etc.

This is by far my largest BR drain at the moment, especially since getting coolered (see EV diff.) and this together make me seethe in rage and that is not the optimal state of mind for sure. I do take breaks after these things to cool off but it doesn't help since it just keeps happening.

BTW I looked into the database to see what the best players do and mostly I see a distinct lack of getting coolered in these situations (though I only have a few k hands on each of them)... so they are all in nice profit

Opinions?

I have 47k hands at nl10 this august, so i'll share what I've felt.

I don't use HEM or any tracker, and honestly, in nl10, people tend to fold post flop to any sign of strength. Set mining usually just gets me that continuation bet of theirs, and that's it. RARELY do I get many chips at nl10, they don't fight for the pot if they don't have it.

ALTHOUGH, I do certainly make good profit by check-calling with low pairs when I feel they missed their pairs (low cards come), and it seems suspicious. I'll get a barrel or two there quite often before they give up and check it back to me.

Small pairs started to become profitable for me when I started 3betting with them 100% of the time (Granted this was on zoom, so no one had history on my betting patterns). I'd take the pot preflop most of the time, and still found it easy to outplay the regs postflop when starting with a pair.

Another trick that works well at nl10 (funny enough) is when in position and checked to (and I have a monster, with no scare cards possible on the turn), i'll min bet. You'd be surprised how everyone thinks a min bet = weakness or some pseudo blocker bet for the turn. I think at least 75%+ of the time, they check raise a min bet. Then I just play the calling game on them, getting big profit.

I made about 8.9bb/100 over that 47k sample and am now on 25nl. Started today, not sure how hard it is yet compared to nl10.

sample size, 54k hands may as well be 540 hands for comparing what you're trying to compare.

AIEV would suggest you're b/e or winning, probably should be happy with that if you've just moved up. I guarantee set mining isn't the reason you're not 7bb/100 or whatever.

those set over set must be extremely tilting to get over and over again...

well... that was not a great result, but by the way you said it, it was the best you could do... next month will be better i'm sure! good luck!

acky100 - Well, due mainly to this scenario pairs 22-77 are net money losers, -138.5$ which is around -2.5 bb/100. If I just dumped set mining with them this is how much my win rate would improve. I didn't bother to add up everything but I have lost around 15-20 buyins in these set-over-set situations, obviously I know I should get a few of those but 15-3 is imbalanced.

When I add the psychological strain I feel from the constant losses I am sure I lose probably another 1-2 bb/100 on top of that and that is a pretty big cumulative effect. I can't imagine I have another bigger leak (than around 3.5 - 4.5 bb/100) in my game at the moment, obviously I must have quite a few but I would guess individually they are not that big.

Now, if this is really variance then who cares, but if it is strategic then it is a major leak that needs plugging (that is my concern). I can't play TAG style when the "agressive" part of my game (with 22-77 hand group) is getting me net losses.

bjj_mat - Thanks for sharing. Well, it depends on the player, I basically know which regs are sticky and which fold to aggression unless having something big... non-regs are usually sticky/spewy.

3betting small pairs is something I do rarely and only vs players who fold a lot to 3bets (making 3bet itself break even or profitable), I am in position + I think they won't fold easily if they didn't fold preflop (so I can get the $$$ if I hit the set, in theory at least). This isn't effective since obviously small sets are getting utterly annihilated.

Floating with small pairs is as effective as with any other cards, obviously there are problems when I actually "luck into" a set on the turn and again I get destroyed since I am calling down cbets (or vice versa, I have a flopped set vs his overpair and then he gets a set on turn and I lose another stack).

The min bet thing - the regs know me and there is no way I call sell this non-regs, yeah, but even then I make a smallish "blocker" bet (1/3 pot) and the effect is the same.

Very nice win rate, I can only dream about it...

stevenright - Yeah, not to mention I am getting paranoid thanks, we will see how the next month goes.

Honestly, your point is valid if there isnt a sample size concern, but the sample is not 54k hands because you're trying to compare the 18 times you've had set over sets or something.

Also, you are really really really over estimating the amount of bb/100 you're losing by keeping this leak, if it is a leak then its a small leak at best due to how rare the situations are, nowhere near the magnitude of even 2bb/100, if it EVEN is a leak, i've done quite a lot of database work myself but i still can't even check the profability of myself playing small pairs with any decent reliability, and i've got over 300k hands at zoom, and that's just overall profitability. You're narrowing down the analysis from the profitability of small pairs, to set over set spots with small pairs, and if you can't already see what im getting at, its super meaningless - 18 times in 54k hands, id guess its just as likely that you'd create a leak by stopping set mining at these limits, to be honest.

Im not saying there isnt times to drop small pairs, infact theres tons of them, so work on folding them when there are too many players to act behind, there is a big squeezer behind.. etc, but folding them for the most part especially in late position at nl10, or vs nits, is so much more likely to create a leak than calling with them, as they tend to be very profitable hands at the micros for most people.

I understand your point, but getting a statistically valid sample of set over set scenarios is gonna take me a lifetime (probably even then I wouldn't have big enough sample ), so using hands played and the frequency of this happening is the only practical approach I could think of.

I guess I can give it another month (that should be around 40-50k hands too) and then see what happens, even though I am not exactly looking forward to this trend continuing.

The problem with playing small pairs from EP/MP while also getting destroyed when hitting a set is just too much to overcome... but if I just dump all that from my already narrow ranges then I am face up every hand against regs (fish don't notice anything anyway) and they know exactly which boards I can't possibly hit. It's a conundrum