I don't think you're going to get a clear trend on session types, largely because session lengths vary so widely among players. Further, game styles with very different standard deviations can have similar overall expectations.

Clide: 1/2 stats can be deceiving, Some 1/2's are home games (big difference). Some 1/2's have large buyins/match the stack, etc.

Squid said it best/ ~65% winrate, big wins>>>big losses. Different styles win big.

Thanks Squid, thats just the kind of sample I was wondering about. I wonder if other 1/2 winners at this rate have a similar experience.

Garick, Thanks for the input.

I’m not expecting it to be an exact science. With so much of poker result interpretations being so ambiguous and subjective, I’m just looking for some clues to glean some insight.
I think in general, just considering session to session should give an idea in spite of some outlying session lengths.

Im just considering live casino 1/2 games 9-10 handed.
I think you are right that many things go into how big a game really is but I think getting into the weeds on that will just lead down a rabbit hole. I think as experienced poker players most on this forum are likely to be, we can make some general assumptions about what a live casino 1/2 game will typically be. Enough of these games will be enough alike that I believe we can learn some things from that information in spite of the outliers. After all, accepted win rates are based on some of these same generalizations.

Well, if you search this thread, there have definitely been discussions of percentage of winning sessions. IDK if there are also some of win/loss size ratios.

I have not seen too much detailed statistical analysis of session length/result ITT.

When I'd looked at my results I had something that was mostly normally distributed with a few "buy in sized" artifacts.

This is from ~1000 sessions and probably 3000 hours or so (don't remember exactly, it's from a few year ago), all $1/2:

The spikes are going busto from some $100 BI increment. There are some similar spikes at cashout from a combination of playing extra hands and tipping out the dealer whatever random extra white chips I had. The WR here isn't super great, but it's a lot more data than I typically see at this stake.

Clide: Yes, you're right about thinking in terms of a 'budget' instead of a bankroll to start. Ideally you'd want to have a fund that you can play out of continuously and never need to "pay in" to play. But when you're first starting out you don't need to save up $4k to sit down at $1/2. Set aside about $1000 so you're not getting tapped out and running to the ATM after 2 busts, and start playing. Track your sessions and your spending, then see where you are.

About 2500 hours of 1|2. Table winnings only (promos/comps excluded). 11BB/h for the sample. If a higher resolution would be interesting I can adjust it.

2500 hours in 1/2 playing in Florida…are you allergic to money or something?

Not sure what this means. Are you criticizing me for not playing higher stakes?
I also have over 2k hours at 2|5 in Florida. About half my 1|2 hours are in MI and about half are in FL.

It was an "easy" $30/h (after adding promos bad in) when I was having a hard time for reasons I don't want to disclose to strangers. It wasn't the primary reason but there would have been way less 1|2 hours if casinos closing due to covid.

Brown, this is great! Thanks
Im not sure what the numbers at the bottom of the graph mean. I understand they represent 100bb but I don’t understand the designations of (2,3] etc.
Can you please explain? Sorry if it should be obvious :/

Think I got it.
1 to 2
2 to 3

(1,2]
(2,3]

They are intervals. A parenthesis means the range doesn't include that endpoint and a bracket means it does. I think this meaning is universal in math but I'm not sure. It's not important at all but it would bother me if that information weren't included.

If you missed it the units are in 100BB "buyins."

Thanks for making that clear.
Do you know what your win/loss percentage was for this stretch?
It makes me wonder how impactful those very large anomalous wins are on a win rate. Though they are so few, they are really big. Im talking about the 7-12 range.
Would you say that you rely on these kind of sessions from time to time to maintain a ~10bb/hour win rate?
I’m trying to figure out if these kind of sessions are just anomalies or if they should be a feature of a winning style’s session history. Maybe they seem like a lucky night but are actually a necessary “side effect” of a winning strategy. Its hard to get a feel for how impactful they even are without doing the math. I don’t feel people find these distributions important or interesting but I really do think they offer some insight. I’m curious where you would say your playing lies between conservative and aggressive/volatile?

Meh, that's nothing. I've got 365 sessions / 1354 hours of Live 2/4 *Limit* under my belt.

Glol@me,ldo;48%sessionswon@$1.96/hour,winningatlife!G

Also, for those guys who posted those nice win/loss distribution giraffes: is there a simple way I can do something similar based on a single results column in Excel?

ETA: Currently looking into some Histogram Chart tutorials. Hopefully my pea sized brain can figure this out...

GcluelessExcelnoobG

It's pretty straightforward: https://www.excel-easy.com/examples/histogram.html I do mine in Matlab because that's what I deal with every day for work.

Mine was raw win/loss in dollars, so it'd be interesting to see the difference between absolute $ and $/hr. I've got a pretty wide distribution of session lengths but I didn't see any pattern with WR vs. session length.

My version of Excel doesn't have the easy peasy Histogram chart option so I had to fart around with it for a bit, but I think I got this...



So that's over 627 sessions at 1/3 NL limit (6.87 bb/hr winrate over 4959.5 hours and, sigh, holding).

I broke it down into 66bb = $200 BIs even though I've only been doing that for the last ~1/3rd of my hours (first ~2/3rds I did 100bb = $300 BIs).

ETA: Didn't do the correct fancy smancy brackets, but [0,1] means $0 to $199 (inclusive) in my giraffe.

GIwasted*way*toomuchworktimeonthis,lol@meG

Interesting quick takeaways from my giraffe:

42% of the time I fall within a $200 (1 BI) win/loss. Sounds nitty (especially considering my average session time is 7.9 hours)?

67% of the time I fall within a $400 (2 BI) win/loss. Gotta be nitty?

My winning percentages at various BI (66bb) wins versus losses:

-1 (263): 61%
1-2 (154): 65%
2-3 (94): 61%
3-4 (52): 77%
4-5 (11): 55%
5+ (27): 93%

GcluelessnumbersnoobG

GG, that's more than clear enough to make the point. Binning at a full BI increment makes for a nice trend without the artifacts I have in mine.

It's hard to tell visually, but I don't think your distribution is much tighter than mine or Browni's. Although maybe we're all three nits.

What's the story on the +11 session?

I'm guessing if you binned then you'd have the same shaped giraffe?

I have no clue if binning at a bigger BI makes a difference or not? And then on top of that could there be a difference in topping up to the BI after every hand (which I do) versus not? It's also possible I should have broken my stats into 2 batches (the first 2/3rds results using the $300 BI versus the latter 1/3rds results using the $200 BI)? But I'm too lazy to do that.

It's funny, looks like both me and Browni have a single way outlier largest win. I suppose somewhat to be expected since I'm guessing most of us have a built-in stop-loss due to limited money-in-pocket (my largest loss being $1200 which was my maximum money-in-pocket at the time), whereas obviously we don't have that for wins. And yet I literally have zero recollection of it at all; just know for sure it happened and logged it in multiple sources afterwards. I do recall one session where I busted like a couple guys in a row and broke the only table running at the time, and then lol moved all my chips into play at the 2/4 Limit table while waiting for the 1/3 NL table to re-open, although I have no clue if it was that particular session or not. Although if you ask me about my largest losing session, I can tell you exactly how the last 2 BIs went into the pot (and in the last hand even give you the name of the villain, in spite of that hand happening like a ~decade ago).

Gpain>>>>happinesswithregardstomemory?G

GGs graph looks a bit tighter. It would be a lot of tighter if he had bucketed by 100BB. Standard deviation is higher in buyin units for smaller buyins, but lower in BB units.

I don’t think there is anything to interpret from my graph. It’s always been my opinion that these statistics are not useful. It looks really close to a normal distribution to me. Might look nice to overlay the bell curve onto my graph.

My biggest outlier is from a short 4.5h session playing with a couple whales at the table where my winrate was likely $100+/h and I ran good.

I believe I play looser and more aggressive than most forum users advocate for 1/2. Less aggressive than optimal because I still (intentionally) underbluff a lot of spots but the idea that you can’t bluff in at 1/2 is a pretty bad one. There are tons of weak/tight and scared money players and tons of players who split their ranges everywhere so they’re really weak in certain spots.

I'd agree 100% that there's not much to take from analyses like this. The sample sizes aren't large enough, there's too much variation in conditions (I'd beat the piss out of 2012 me if we met on the felt), and there's no way to aggregate between players in a meaningful way. But I still like looking at them.

Same with the "day of the week" analysis. By gut and my eyes tell me there's a difference, but it doesn't shake out in the limited data.


Not to detour into strategy too far, but that's a great point about bluffing at $1/2. There are plenty of good spots for it, especially if you have the right image and use it sparingly.

I want to start taking this hobby more seriously and do it right. It’s not going to be my job or anything, but I hope to get a consistent ~5 hours a week of play. Can I please get your opinions on the following?

1. I initially planned to start at the lowest stakes in LA, but apparently the buyin structure is less than 100BB at everywhere except The Bicycle’s 2/3 game. I donÂ’t know if I’m a winning player yet, which is why I wanted to start at the bottom. However, I’m closer to other casinos and could start at 2/5 or 5/5 too. Are ~$500 buyin games much tougher than the 1/2 to 2/3 games in LA? https://www.pokeratlas.com/poker-rooms/los-angeles

2. Is 20 buyins enough? This would be $6k or $10k, depending which blind level I decide to start at. This is the only money I plan to invest, and if it turns out I’m a losing player, then I won’t be topping up/trying again.

3. I think it’s strongly recommended that I subscribe to a training site. I’ve searched around and found a bunch: upswing, raise your edge, run it once, solve for why, red chip poker, and crush live poker. What differentiates them? Which one do you suggest?

4. I’m guessing it’s generally true that it’s easier to play on Friday night and Saturday. True? Are games that much worse at other times or day?

5. Besides a training site, I’ve read a bunch of poker books including Modern Poker Theory, The Grinder’s Manual, Applications of NLHE, and others. I think the last studying to add would be a solver. Does one come with the training sites? Is there anything else?

6. Which casinos are recommended in LA, and why? I’ve been watching the Hustler’s live stream and it seems pretty cool.

7. Any other suggestions/advice?

1. From lives streams and hand histories I’ve seen, I’m not sure tough games exist in LA in any capacity.

2. If it’s a hobby and you can replenish the money, 1 BI is enough. 20 is plenty.

3. red chip and CLP are geared towards live poker while the others, I believe, are geared more towards online. I was a CLP sub pre-pandemic and liked their content.

4. Personally I find weekday games “easier”. They are reg filled and typically many of the players play a very straight forward style. I tend to have a fairly good idea where I’m at. Weekend games are typically more actiony. People are drinking more so the gamble gamble comes out. All this is to say the games are generally “easier” in different ways.

5. You can use a solver when studying live poker but at lower stakes it’s not going to be super important. It’s good to know the information but relying on its results to try to play more GTO in LLSNL is just pointless IMO. Many pots go multi-way for many streets, so pots get bloated quickly. IMO, understanding combo metrics and pot odds is of much more value at these steaks

6. Never been to the area.

7. Good luck. Stick with the studying, LLSNL isn’t too difficult to beat with a little bit of studying and some stubbornness to not quit if you lose in the beginning