Yes of course haha. You got to give me more credit than that Garick! I didn't multiple to get sd/100. That's set by site. I also couldn't change BB/100 WR for calculation purposes, but I did convert it only for presentation of the data.

Sent from my Pixel using Tapatalk

Multiply by sqrt(4) = x2

Just checking. I have seen that mistake made ITT.

Regardless, 100bb/100 std is more indicative of a 6m online game which doesn't go multiway to the flop every ****ing time etc etc

Not sure about that one because a common down swing might not be the ones that most affect the bottom line. Sounds like it would be dominated by a few stale patches. I mean the app is the app but I would rather ask it "what is the avg # of 200bb losses a 10bb winner will experience of 100k hours?

This may or may not be interesting to someone.

In 2016 I played 173 hrs of 2/5 on Tuesdays at a win rate of -4.24/hr. My worst day of the week by a very large margin.
In 2017 Ive played 57 hrs of 2/5 on Tuesdays at a win rate of $118/hr. My best day of the week by a very large margin.

That is a meaningless sample as evidenced by your range of results. You're confidence interval on that has to be something like 12%.

That's not the point of the post. I posted it for people to drawn observations to help or understand BRM better and to understand what's normal when it comes to variance. Meaning, assuming the 100/BB SD is a close proximity (just assuming, not saying it is), then you can expect at some point to experience a 1,000BB downswing if your a 5BB/hr winner.

Yeah, I know...but there are some days of the week in certain industries that are worse than others. Im sure Tuesdays are pretty bad in the restaurant and movie theater business. That's why you see coupons for those days. To drum up business.

Last year it really seemed like Tuesday was a really slow day for action in the poker room. Fish and rec players gamble it up on weekends. Not so much on Tuesdays. I started specifically scheduling errands, going to the beach ect, on Tuesdays. This year has been totally opposite but like you said the sample size is clearly way to small to make any observations other than its a crazy anomaly from 2016 to 2017. It was like someone flipped a switch.

Maybe you need to play better poker when the fish and rec players aren't gambling it up.

Maybe I am and that's why I went from -4$/hr to $118/hr?

Game conditions are totally relevant. Interestingly I hit the casino last night at 10PM and played on an insane table for 4 hrs. Tuesday could be the new Friday.


I made $44 which was triple my current hourly rate for the year.


Spikerunningbreakevenraw22

+1

It's taken me 7 years to get to my 3285 hour sample in my 1/3 NL game, which is probably in the 100K hand ballpark.

First 2000 hours: 9.4 bb/hr (I am the best)
Last 1285 hours: 4.3 bb/hr (I am the worst)

Moving forward, what confidence does knowing my overall winrate is 7.4 bb/hr give me? Absolutely none whatsoever.

Geverythingchanges,renderingpastresultsalmostmeani ngless,imoG

FIFY.

Btw, I just followed my largest upswing (20-3 over 23 days +7500 including 14 straight wins) with my largest downswing (8 of 9 losing sessions for -$3,825).

Poker is funny.

First. A professional can achieve these results in 2 years playing 40 hours weeks or 3 years at 25 hours a week.

Second, the post intentions where to wrap our minds around expected volatility that great, good, and ok player should expect. Not how much money you expect to earn given current WR.

Third, your post about how you earned a lower BB in the last third of your 1,285 hours doesn't mean your WR of 7.4 bb/hr is worthless. Yes, since your ss goes away back, your 4.3/bb hr is prob closer to reality. But, you should expect to go on long deviations from your true mean. 7.4 BB/hr is simply an average. To many people think in binary. 100% meaningful or 0%. My point is, there is definitely value, meaning, and conclusions that could be drawn from the WR.

Ha, honestly I'm totally cool with any + winrate and don't think it should be undervalued, especially as conditions worsen. Little tough to stomach personally due to where I've been, but overall it's just a stupid game I play for fun, so I'm cool with it.

My winstreak record is 32-3, over a 6 month period in 2012/2013, back when I knew how to play poker, ldo. I can't even conceive of accomplishing something like that now.

GcluelessNLnoobG

There haven't been enough Tuesdays in 2017 to derive anything from results. 2016 sounds like feedback where you cut out some of the 52 Tuesdays. I use the same circular logic with Fridays. But you may be right.
IMO the first glance test of game quality is more accurate than a sample of 12 sessions.

During past 1,100 hours, my Tuesday hourly is $101.85.

Guess I should switch up my schedule, or could be because total Tuesday hours during that time is 5




Sent from my Nexus 6P using Tapatalk

It's all statistical noise. I am crushing Thursdays in like 200 hours or so and it's the best day of the week for me.

Since people talk a lot about how many hours you need in order to get an accurate picture of your win rate, people should look at their weekday subsamples in order to get a glimpse of alternate starts in their poker playing career. Over a hundred hours on Sundays I am doing -7bb/hr for example. I know for a fact that I am not a losing player, but conceivably, that's how my first 100+ hours could have gone. And so on.

What also may be interesting is this: I think a reasonable assumption is that for most serious players, Saturday should be the most profitable day of the week followed by Friday, then probably by Thursday or Sunday and then by Wednesday, Tuesday, Monday.

On Saturdays and Fridays you should be winning more than average and on Mondays, Tuesdays, Wednesdays lower than average.

When you have a relatively small overall sample, like let's say 1000 hours, those weekday subsamples should be all over the place, in all probability not conforming to the assumption I laid above. But as those weekday subsamples get big enough, they should start conforming to that assumption. So maybe after you have 500 or 1000 for each weekday, you may start getting a decent picture of the true difference among days.

I think my Saturdays are like that. In my first 200 hours, Saturday was not a big profitable day for me, winning 5 bb/hr less than average. But now that I ve reached 600 hours or so, I am winning 2 bb/hr more which does make sense. It could still be statistical noise though.

I played 1800+ hours in 2016 and Saturday followed by Friday were my least profitable days of the week (they were both badly negative and I had a good winning year). I only logged 425 hours across those two days because #sociallife but the point remains.

As someone who has probably played more hours than 99% of the active posters on this forum over the last 2+ years I am more convinced as time goes by that nobody has any clue what a statistically significant sample size is. Variance in live poker is remarkably enormous to the point that you can't explain it to someone that hasn't experienced the full cycle of highs and lows.

I used to be one of those people that thought "breaking even for 100 hours" was running bad. And then I played more and more and more and saw more and more and more and have begun to come around to the realization that poker is not the never ending ATM you think it is when you are running good.

Most importantly, it's made me realize that when you think you are running "average" you are actually running pretty good, when you think you are running "bad" you are most likely running average, and when you are cognizant enough to admit you are running "good" you are most likely sun running so hot that the thousands of hours that succeed that period will leave you wondering why the "good times" never return.

Just my .02 (not bad for post 3333)

For the reasons JohnnyB said I say the smell test is better than 1/7 of a small sample size. The best day certainly depends on the market. At Foxwoods it was Sunday as much of the clientele was the weekend tourist who had his last chance to get even Sunday night. At CAZ the weekend brings a horde of tight collar professionals who don't have time to play during the week.

Consider the STD instead of winrate, it might give you an idea of how much is being wagered.

While most winners perceptions on winrate potential are probably skewed to make them think they can win higher than possible long term like you say, some people run way worse than 95% of people ever do for some periods, like you probably did. Your views are probably skewed the other way.
Id be interested in hearing about your best, and worst 500 hr stretches (or even 1000) if you wanna share, just to see the variance that a good player can have

+1

I've played 3285 hours at my 1/3 NL game, which means an "average" sample size on any particular day would only be 469 hours, which is pretty much meaningless.

It's also no coincidence that of my day-by-day breakdowns that the day the ends up being closest to my overall winrate is the day I've put in the largest sample size (Sun @ $22.77/hr over 1144 hours, which is closest to my overall $22.26).

I would actually argue that in some markets (mine being one of them) that day / time-of-day makes zero difference, as it is just a consistent wall-to-wall reggish line up (i.e. there are zero random Friday/Saturday night casino goers with beer-in-hand sitting down at poker tables in my room). In fact, if anything, it would be the opposite at my room (as Friday/Saturday the 5 table room is usually so packed that there is a multi-hour wait list, preventing any random casino noob from ever sitting down, whereas on slower nights a random casino noob would have a better chance at getting on a table quicker).

GcluelessnoisenoobG

When I've done day to day breakdowns I've seen the same thing. Too much noise even in the 500 hour samples. Same applies to comparing different rooms.

But that doesn't negate our eyes. A room with more traffic on a given day should afford us greater table selection and a higher cap if we are good enough to exploit it. Same with the average stack on the table or average pot size (easy to observe). Different rooms and markets are going to have different hot times. And as GG says, sometimes a room busy enough to have a waitlist will turn away the fishiest players. I've seen that happen often too.

Are you referring to me not showing up the other night because of the list?


I prefer the never ending school of fish and chips offered by the MGM buffet these days.