Instead of continuing to flame every bad nit thread I can before the mods nuke it, I thought I'd write one of my own and give some advice for those of you who really aren't happy unless your opening range is JJ+, and maybe including 99/TT/AK.

First, I have some good news. There is a time and a place to play that way live, and if you add in calling behind in LP with pocket pairs (limping if there are limpers, calling a 7bb or less raise when someone else raises, assuming implied odds of 15-to-1), then you actually have a shot at winning at a number of live games.

How do we make money being a nit?
1) Game Selection
2) Hand Selection
3) Value betting/value bombing


Game Selection:
To make this style profitable, you need there to be action. As you want to be a nit, you will be generating no action; in fact, should your villains be decent (rare), they will not give you any action, they will fold their marginal hands to your raises, and they will continue postflop only with hands that beat an overpair. From 2002-2007ish, playing a super-tight, uncreative, super-nit style would get you paid off at almost all times of day; as the games have evolved a bit, this may no longer be the case at your local card room at all times.
So what are we looking for:
--People stacking off very light (especially preflop and flop)
--Drunk people
--gamble junkies

This means that you want to play during sporting events, and weekends, esp Fri/Sat night. If your casino has or is near a concert/show venue, you want to play when the show gets out. To make this profitable, you need a table that isn't paying much attention and wants to gamble.
Even better would be following the WSOP or playing LLSNL in Vegas when the Series is running.


Hand Selection:
The people who start the nit threads seem to think that hand selection is the most important part of winning as a nit. Heads up: its not. Game selection is more important; without a desire to bluff, use razor-thin value bets, and play some suited trash in LP, then you're going to need some gamble-degens as opponents to win.
However, we can be very selective with our hand range. Even still the following hands are always a raise (if nobody else has raised):
JJ+, AK
I would strongly recommend considering raising with 88-TT, however, if you really want to be a nit that might not be your bag.
22-77 we can limp/call relatively profitably, assuming we started with a 80-100bb effective stack. This is because most table go multi-way to every flop with someone going broken whenever they have an overpair.
If your flops aren't regularly 4+ way, and/or there is lots of raising and re-raising preflop, then just fold 22-TT, as a nit brain won't be able to handle playing those hands vs. heat. Should you also be facing these kinds of aggressive conditions preflop from people who aren't just degening-off their buyins with trash repeatedly, ask for a table change.
Playing Top 3-5 hands means you will be folding, alot. If folding for 2 hours frustrates you, this style is not for you.

3) Value betting/bombing
Yay, we finally got a hand. And its KK!
Should someone raise before us, we will 3bet. As a rule of thumb, your minimum three bet amount should be a potsize raise in PLO (call the initial raise, and the raise an amount equal to the money you see in the pot).

Often, though, when you luck into JJ-AA, AK there will be no raise. That means you are going to do the raise yourself. The appropriate raise amount should be 5-10 times the big blind, depending on how gambling the table is (if 10X bb raises get called multi-way, then raise 10X the bb; after folding for the past hour, you should have an idea how people are playing preflop)

But then the flop hits: 9T3ddh
Its a flush draw, and someone bets.
Somewhere in most nits brains goes off a little alarm of fear. Turn that off. You made it to the flop with a big overpair, and you've received action. Better than that, its a drawy flop. In this case we are going to raise, and often raise big. As there was a big raise preflop (you did raise, didn't you?), then this should get us to the stack-commitment threshhold. If all the chips don't hit the middle on the flop, get-em in on the turn.


Final Notes:
--Appropriate nit play in LLSNL should be the next-door-neighbor of super-tight, aggressive play.
--Unless you were an online pro with a good 20bb preflop shove matrix, I recommend buying in for 70-100bb
--When playing LLSNL, OP is a game-building, degen LAG, and as such I hope that everytime a nit has an overpair, it gets crushed by a set.

And be aware, this really IS NOT less variance to play this way. You are playing big pots with one pair against droolers. You will often get coolered by weird 2-p hands. You can still make money this way, but it is NOT low variance.

Agreed. I just figured I'd actually give some advice on how to be a longterm +EV nit for those (AlmostPro, and the donk who started 'nits united'), who seem dedicated to the style.

Before I chime in on the matter; what do you find to be the lowest Var style?

I don't know live. On-line, I analyzed it and the aggressive 12-16 VPIP with a PFR a couple of points lower is the lowest variance.

Reason: They mostly have the initiative on the flop, the flop is 1-2 villains and they miss a lot.

Not sure where it would be live, but probably not too far away from this.

Yeah, I figured OP knew that. I just thought I should point it out to the potential nits. I find that a LP style has the least variance, but also the lowest win-rate (if it's even positive) . Generally, looseness lowers variance, and aggression increases it. Win rates go WAY up with aggression, though,

I think the best way to reduce negative variance is to play a style that is 5+bb/hr at your game. If you play a style that is ~ 20bb/100 (or 10BigBet/100) winner, than losing becomes far less likely. While your true variance may have increased, playing a lower variance style near or below EV is bound to show significant downswings.

When I first started playing alot trying to make money, I played a style that while not quite this nitty was its slightly looser, more aggressive cousin. And my swings were pretty big. I ground out (including paying back all initial money 'invested), then busted something like 3-4 times b/t mid 09 and Feb 10 (playing mostly 1/3 with a small amount of 2/5 mixed in). St. Louis WSOP week it all 'clicked' for me. I started playing a slightly looser, slightly more passive version of this, where I would give people cards, and induce a decent amount of LAG bluffin, while making sicko bluffs that were NEVER called given my image; and my w/r went from small to like 5-10bb/hr (figuring out true w/r live is LOLhumorous from a math perspective)--large amounts of 2+2 reading and posting helped immensely (special thanks to DGAF, Limon, and Kablooey).

My variance went way down as my w/r went way up. My lowest VAR is obvi game dependent (playing no-foldem degen tables where you auto-stackoff QQ+ preflop is high vol, but highest WR), but at your avg, passive LLSNL East Coast table, I think my lowest vol style is the spewy looking 30/10-15/5-8ish style that I discuss and have the most fun playing. However, I massively change styles depending on game and table conditions (the moment I think I'm in a shark tank, I drop to under 15% VPIP). So for me, thats an LP preflop with LAG postflop style. However, everyone is different. The top winners I know at LLSNL (with the most consistency) play brunsonesque styles. Where that would be highly vol at 5T+ in a big sino, so few villains play back effectively that I think those styles are consistently profitable at an LLSNL game.

Quite honestly, I've also met a couple highly profitable good nits. They play very tight, and relatively aggressive, and are aware of their image (sometimes loose b.c they are young, sometimes nitty, b/c they're nits), and depdending on how the table is viewing them depends on if they are bluffing or not (decent lags basically can't continue against their raises, and they know this). However, their range is wider than this, and their postflop play is far stronger.

Curious any thoughts relating to this as it pertains to development of a players skills.

An interesting thing I see with a certain type of Nit (some TAG some TP). At the risk of being too general it's often the late 30s to early 50s white guy. He may dabble in the low BI short/fast tournaments too.

A while back I was thinking of these players when re-reading Harrington on Cash. He suggests if you have a below average postflop game to go with the strict starting hand requirements and bigger bets etc. A potential problem seems to me that with that style the post flop game develops less because this style is really winning pots on flops a lot or bet bet betting and hoping to hold up. This is fine when shorter but disasterous when deep.

I see these type players who, I think, because they play so many less hands to the river and so few hands in general really struggle post flop as OP indicated in number 3. The two vulnerable sub-catagories I see are a) those who are married to their flopped TPTK and overpairs when fairly deep. "how could you call with that?" and b) Those who are sort of opposite and will lay them down a little too much on scary boards.

So these players inevitably wind up getting stacked or floated and bluffed as the case may be a little more than the average player. It seems to me that variance would be high for the former type of Nit and WR low or negative for the later.

I have a friend who is very recreational (plays a couple times a month) and varies between both these subcatagories. He usually buys in for between 70 and 80 BBs. I suggested to him that he should consider a double up and rack up policy at his current level and really start paying attention when he is not in hands to learn as much as he can while at the table to spot what different signs of agression from various players means post flop.

I realize there are very competent post flop players who also play a winning nitty style. I am not talking about them but I think the ability to protect a stack post flop is part of the variance equation. One way to guard against that is to simply limit exposure.

Re: Developing Skills. I highly recommend (in this post BF world), that anyone starting out should find a 20 or 50NL w/ 100bb cash homegame to play. And then they should play >75% of hands preflop. Which will prob be like 80/20, because nobody 3bets in a game like that and most pots are limped.
It will give you LOTS of experience playing pots with fishies at a stack depth you will see live. Even better if 50 means alot to them, because they will have the same scared money effect of a live nit with 250 in front of them.
I did this back in 08/09 and it really helped develop my postflop skills from nonexistent. It takes awhile to get used to calling with middle pair, and multi-street bluffing, etc.

Note: this only works if you're trying to win while playing that loose. Quickly you will discover that just playing your cards w/ an 80%VPIP is a losing proposition.

Great discussion so far. I like where this is going.

I'll add some of my own thoughts on the topic:

1. Image

If you're young (and this is especially true if you look young), people will think you're FOS most of the time. Remember, poker is all about doing the opposite of what your opponents want you to do. So if your opponents expect you to be aggro lag-tard Tom Dwan-on-crack, then you need to play tight and value bet big.

This means bombing 3 streets with AQ on the Q8626 board.

Don't think that just because you play 'TAG', people are going to respect your bets. No one gives a **** that you know how to play TAG. No one cares that you've only voluntarily played 3 hands PF in the last 2 hours, and that you've only shown KK at showdown.

Donks want to call you. To them, you're just a punk kid. They don't like you, and they certainly don't respect you. Don't try to bluff them. Only bluff the true nits.

2. Stack depth

You need to be cognisant of your stack size, as it should affect the hands you choose to play. If your table is playing deep, hands like 78ss lose value, while hands like As3s skyrocket in value. Premium hands also lose their relative value, as they begin to obtain some reverse implied odds.

For instance, A3ss is a fairly easy hand to play deep-stack poker with. We're only trying to make nut hands here. We can get away from the AJ9 board quite easily. The same goes for small/medium pocket pairs-- we're strictly set-mining. But playing 250BB+ deep, a hand like KK can become dangerous, as it is less clear when we should be stacking off with it post flop.

On the J84 board, we level ourselves into thinking our opponent's c/r could easily be a draw, and all of a sudden we're stacking off 250 BBs with just a pair.

I'll just simply note that variance =/ losing. A breakeven player who's variance is +1BB/hr to -1BB/hr has far smaller variance than a 10BB/hr player that is 19BB/hr to 1BB/hr, even though the higher variance player never loses.

As Harrington pointed in HOC, higher variance play results in higher win rates in the long term. Seeking to reduce variance is not what a poker player should be doing to increase winnings. The caveat is that if you are a losing player, no amount of bank roll strategy can help you. It is harder to play a winning high variance strategy than a low variance strategy.

The other thing I'll note is that as one moves up, it becomes harder to have a measurable edge against the average opponent. The effort to be the best at your table moves up exponentially. The effort continues to get harder even at the same level. It was less than 10 years ago that making a cbet with air was "high level" play that only a few top pros understood. Today, Norman Chad has beat it into the most casual viewer of the WSOP going into a 1/2 game for the first time what is involved with it.

Agree that ganes are harder and mathematical Var is higher in the 1-19bb/hr case. However, imho what effect and crushes most live-players results based poker psychology is losing sessions, imho. When most players discuss Var they are referring to negative Var or simply bad play in some cases, even surprisingly good ones (dgaf has some good discussion on variance in his mhfr 2k well). I agree w you on the technicals of your post, but the mental impact is the revers of the math there.

I take advantage of the above, but in an opposite way. I'm a 49 year old guy who is normally fairly quiet at the table--and I'll have a cup of coffee in front of me if it isn't yet well into the evening. Strangers might see me being "Old Man Coffee," whereas the regulars who can see beyond their own cards probably see me as being a solid TAG.

Consequently, within a typical 4-6 hour session, I can often pick up one or two decent size pots heads-up with a semi-bluff (against non-stations), in addition to also stealing some smaller, multiway orphan pots.

Playing a nit style is THE highest variance style of play. Obviously you can beat LLSNL using this style, but buckle up for a wild ride.

I personally find that the better I get at reading my opponents the less variance I have.

for example, was playing 2/5nl the other day and raised $25 from the CO w QJo and was called in MP from a rec-fish ABC player who I knew put me on "AK" but would only call my c-bet if he hit. SB (super weak/tight fit-n-fold player) calls, MP calls, 3-way action

Flop(75) 9 4 2r
checks to me, I c-bet 40 SB folds, MP calls.

Turn(155) 7
I make a big show of grabbing chips to bet and freeze MP from betting, he checks, I check back.

River(165) K
He tanks with a disgusted look on his face and checks, I bet $100. He shows me 98s and says, "You got lucky, I had you till the river" and then he folds.

I had zero doubt that if an A or K hit on river I could bet him out for $100 but if a Q or J hit I could bet $65 for value and he'd call thinking I was bluffing with AK...

Conversely, I've been in spots with trips medium kicker where I snap fold to a villain's shove because I know this particular villain is NEVER shoving without a better kicker or boat.

Which is basically Tommy Angelo's principle of Reciprocity. Every time we make a +EV play that our opponents would not make, we win. Every time we make a money saving fold our opponents would not make, we win. Imo, this helps lower variance.

Sure, anything that increases your win rate reduces your variance. But even if you assume equal win rates, say 5bb an hour, the guy who wins 5bb an hour playing VPIP/PFR 8/4 is going to experience higher variance than the guy who wins 5bb/hr playing 15/10.

I agree, but the advice of "crush the game and don't lose" isn't particularly useful. That said, I agree that a lot of players don't understand that the way to avoid the "poker" variance is to increase the "math" variance of your game.

QJ is one of my favorite hands at a table I can get HU on the flop. If the flop has an A or K, you win with a cbet. If it has a Q or J, you value bet the crap out of people. You'll have a J or better flop over 50% of the time.

Another classic post from venice. But you're right you taught me along time ago about fundamentals and variance. You need math and the psychology side. But the math work away from the table is most important to combat situations when you are pushing edges. Pokerstove and plo has helped me put myself in the best situations. Where I'm bluffing or setting up bluffs with back up plans. Without pokerstove I would be clueless and rely on wishful thinking like timing. Board texture and action determines my timing.

Remember high school? Everyone sat in the same class, studied the same books, and listened to the same teacher, and yet, some get A's and some get D's.

It's not the theory that makes you a winning player; it's how well you are able to apply it.

It is different because you said you wanted to do this to REDUCE variance. It won't. It will INCREASE variance. Have you not been reading the thread?

I play a style very similar to this, but I buy in at about 40bb. I would think the smaller buy-in would reduce variance, but if not, it has at least increased my WR.

The smart players usually fold their small PPs and SCs if I raise preflop since they don't have odds to play me, that leaves me with just the droolers and gamblers post flop.

If I have been folding for an hour or two I will open with a speculative hand in position to take advantage of my nitty image.

Good deal. Variance isn't evil, and can be positive or negative. One can still make money playing this style, as our opponents suck. Just remember what mpethy said, "buckle up for a wild ride."

It has mostly been ignored ITT, but one of the biggest contributors to positive winrate when playing a nitty style is to exploit your image to bluff in appropriate spots.


I want to say this as politely as possible, but this is wrong on both counts.
First, variance and winrate are independent variables by definition. Variance isn't some nebulous concept related to the size of downswings, it's a tangible mathematical concept.

On the second point, from my POV it seems intuitively obvious that playing tighter results in reduced variance, because you play more hands where your win/loss is zero. Just for fun though, I took a look at my HEM db, filtered for players on whom I had >5k hands of full ring, and looked at the SD of players with VPIPs in the vicinity of 8 verses those in the vicinity of 15. I learned two things, first almost no regs play that tight (had 4 players in the 6-10 VPIP range vs 78 in the 13-17 range), and second, the average SD of the tighter group was slightly lower than that for the higher VPIP players, which surprised me. (when I actually go deeper in my DB and look at the 20bb shortstackers who used to play an 8/6 style, they actually have a much lower SD, but I believe this is mostly due to stack size)

From my POV, I ignore the hands that are folded preflop. I suspect the higher variance styles are those which build bigger pots when you see a flop.

At times, I've thought of myself as a smallball nit.

You are correct, but 90%+ of poker players define variance as losing.

Meowmeow:

Win rate is not an independent variable in any utilitarian examination of a poker player's stats I have ever performed. It is the dependent variable.

Just think about the database analyses that I perform for players. They are sort of informal regression analyses that take winrate as the dependent variable, and examine the other stats that affect win rate. So if I ask the experimental question "do cold calling frequency, c-bet frequency and attempt to steal frequency affect win rate?" I have set up a question in which win rate is the dependent variable.

I imagine it is possible to construct an experiment in which win rate is studied as an independent variable, also. If I ask the experimental question, "do win rate , vpip and pfr affect standard deviation?" i have used win rate as an independent variable and standard deviation as a dependent variable. In no way can win rate be considered an independent variable "by definition" as you contend. The design parameters of the experiment determine what are dependent and independent variables.

As for the results of your HEM research, I don't think they contradict my assertion that nits will experience more variance than any other style of play. The reason is that standard deviation is a description of the shape of a distribution after it has been created. But it doesn't tell us anything about how it is experienced while being created. So, if you look at a nit style of play, 92% of all inputs will be non-VPIP events. But those 8% VPIP hands are going to be a wild ride. You'll win a big pot with KK, then get your aces cracked in another big pot, then flop a set and stack a guy. It'll average out to 5bb/hr, but it'll be experienced as big swings.

Compare that to a tag who is playing twice as many hands, with a much weaker holding on average. His sample will include some hands where he picks up a few small pots along the way to smooth out his in session results.

One thing that confuses a lot of people when they think about standard deviation in poker is that a winning player's results are not normally distributed. A winning player's session results are skewed to the right.

The other thing most people don't think much about is that standard deviations in poker are HUGE relative to win rate. For example, my WR at 1/2 online was $4.80/100 hands. My standard deviation was $140/100 hands.

Ok, put those together. 68% of my 100 hand samples were somewhere between -$66 and $74. You see how the sample is skewed to the right of $0?

Now, suppose I am SABR42, and I crush 1/2 at $14/100, and my SD remains $140/100. Now, 68% of my hands are between -$56 and +$84, which means that SABR will experience fewer losing 100 hand samples than I will.

Translate this to a live player, who records results by session, and it means that the higher your win rate, the fewer losing sessions you have assuming SD remains the same. When I said "anything that increases win rate reduces variance, this is part of what I meant--that you'll necessarily have fewer losing sessions.

Looping back to what I said earlier, but didn't make clear: when I said that a nit and a tag with the same win rate would experience different variance--that the nit's results would be swingier--I meant basically, susceptibility to being influenced by luck. A nit's session results basically depend on whether he wins big with big hands. If yes, be books a big win. If no, he books either a small loss or a big loss, depending on whether his aces got cracked or just didn't get paid. By contrast, the Tag has other options. By playing more marginally profitable hands, he can turn those small losses the nit takes when he doesn't get paid into small wins.

I have more I could say, but this is already tl;dr.