How often do you cash? (1 Viewer)

[HanShot1st]

It's been an interesting journey for me this year. I started playing poker actively since June or so. I have two significant wins on the season, one from Feb. and another in July. Which really drove the desire to improve my knowledge and skills. (Also to build my own poker table, and get in on the Majestic chip buy) It's funny that in the process of learning how to try and take my game to another level, i have seen a backslide. I have since figured out a least a couple holes in my game, and believe now that I can get back on the right track, but this made me wonder, what would you consider a successful percentage of games cashed (in the black)? For the moment, i am still up for the year, but with the building of the table and purchase of the chips, my bankroll has to be rebuilt. So instead of setting unrealistic expectations for myself, i was wondering what you think,

 

[DrStrange]

Cash or tournament?

I think you need better data than that though. Look for dollars won per hour or big blinds won per hour rather than how often you have a winning session.

Special caution - don't play trying to "win" the session. +$2 vs -$2 isn't meaningful to whether you are a winning player or not.

DrStrange

 

[Anthony Martino]

Playing live your sample size is going to be awfully small and most likely statistically insignificant, I'm assuming a few sessions/month?

 

[stocky]

So far I've cashed in 2/3 PCF tournaments. So I'm almost pro now.

 

[HanShot1st]

Actually curious about both, cash and tournament. I started using Poker Income Pro on my cell phone at the suggestion of Trihonda, so I am able to track many different things. The one aspect i haven't been keeping track of is each hand. I fell it pulls my attention away from the table, but in all fairness I haven't even tried to keep that info. So I have been left to just recording the results of my sessions. It has room for notes, etc. Mostly where I scold myself for hands i played poorly.

And your right, as far as sample sizes go. It's about 2-4 sessions per month. There was a stretch where I was going at least once a week, but have pulled back on those reigns a bit lately.

 

[Anthony Martino]

Yeah, unfortunately, statistically speaking, 2-4 sessions per month just isn't going to be enough to determine a realistic expectation of your long-term results.

I mean, let's say you do the max of 4 sessions/month. So that's 48 sessions/year. It sounds like a lot, but a professional playing 5 days/week is going to cover that many sessions in roughly 10 weeks (i.e. a little more than 2 months compared to your 12 months).

I'm not sure how many sessions of tournament poker would need to be played to determine a win-rate. But for cash game results the number at a minimum is in the low five-figures worth of hands played (i.e. 20K) and there are plenty of folks who state to arrive with a reasonable expectation based on standard deviations you really need hundreds of thousands of hands played.

There's just no way to accomplish that with 2-4 sessions/month. So by all means keep track if you wish, but I wouldn't agonize over the results if they're bad, or get too excited if they're good, as the sample size is too small to adjust for long-term variance.

 

[DrStrange]

You might be surprised what the numbers will show from samples that small. Something is better than nothing. We can make solid observations about who are the big winners and big losers with a ten session sample playing in big bet cash games.

Statistical significance for LIMIT poker is damned hard to get because the win rates are typically less than 2 big bets per hundred hands and the standard deviation is roughly 13 big bets per hundred hands - in other terms, the noise is ten times the signal.

No limit or pot limit results will be far more vigorous relative to the variation. OP should easily be able to classify himself into a box, i.e. Big loser / loser / break even / winner / big winner with reasonable confidence using a year's worth of data. What OP will not be able to do is figure out if he is getting better or getting worse unless the change is profound - 35 sessions per year is not enough to establish trend AND we can't tease out changes in the other players / changes in table composition from the changes in OP's game.

If we are looking at tournament results - - - then OP is back in the limit poker problem. The noise is much bigger than the win rate.

DrStrange

 

[Anthony Martino]

OP should easily be able to classify himself into a box, i.e. Big loser / loser / break even / winner / big winner with reasonable confidence using a year's worth of data.

I agree, if OP is on a full-time schedule. With 2-4 sessions/month as you stated the noise is just too high to ascertain with any certainty

 

[detroitdad]

I only monitor my based on my poker bankroll. Its currently on the positive side. I'm apparently winning more than I'm losing.

Of course the value I have in poker chips is eating into my bankroll way more than any losses.

 

[bentax1978]

Of course the value I have in poker chips is eating into my bankroll way more than any losses.

Don't think of buying chip as losses to your bankroll; You're just changing the liquidity of your bankroll with an asset that has the potential to increase in value.

 

[detroitdad]

Don't think of buying chip as losses to your bankroll; You're just changing the liquidity of your bankroll with an asset that has the potential to increase in value.

That doesn't usually happen. I usually break even. Basically I'm playing with the chips for free for an x amount of time. I think I've made a decent profit twice. Only lost money a few times.

 

[Shaggy]

Basically I'm playing with the chips for free for an x amount of time.

That's exactly how I look at it.

 

[bentax1978]

That's exactly how I look at it.

That's actually how I've always looked at it too, which is why I made sure I highlighted the word "potential" in my post. I would never advocate buying chips with the expectation that they increase in value, especially now when prices seem to a bit on the high side. Obviously the same goes with customs, which in general I would not expect to appreciate (and would very likely be worth less to someone else).

 

[Mr Tree]

That's actually how I've always looked at it too, which is why I made sure I highlighted the word "potential" in my post. I would never advocate buying chips with the expectation that they increase in value, especially now when prices seem to a bit on the high side. Obviously the same goes with customs, which in general I would not expect to appreciate (and would very likely be worth less to someone else).

Customs to me are something you should count on taking a loss on if you sell them. Just too personal.

 

[HanShot1st]

I appreciate the feedback, and once again am faced with the fact I will always be my harshest critic. Relax, kick back, and play some poker.
Thanks!

 

[Anthony Martino]

I appreciate the feedback, and once again am faced with the fact I will always be my harshest critic. Relax, kick back, and play some poker.
Thanks!

I think it was Chris Jesus Ferguson who said on the tournament circuit the best pro's could travel it all year and not make any significant cashes, such is the variance (and the reason why the pro's that tend to stand the test of time often have cash game skills that provide a steady stream of income, but even then they can lose for significant stretches at a time as well)

 

[links_slayer]

if you're looking to improve, maybe focus more on the decisions and less about the results? since you presumably aren't playing poker as a source of income (but rather as a form of entertainment with the possibility of making some $$), just make sure you are buying in for amounts that are within your means.

 

[Trihonda]

if you're looking to improve, maybe focus more on the decisions and less about the results? since you presumably aren't playing poker as a source of income (but rather as a form of entertainment with the possibility of making some $$), just make sure you are buying in for amounts that are within your means.

Exactly!

Monetary statistics are fine (if judged long term to show profitability or long-term success), but if you simply focus on making the right decisions (and having fun, of course), then you're more likely to be profitable (and have fun).

 

[gopherblue]

Must...refrain...from...making...card protector...comment....
:facepalm:

 

[lherron]

I think it was Chris Jesus Ferguson who said on the tournament circuit the best pro's could travel it all year and not make any significant cashes, such is the variance (and the reason why the pro's that tend to stand the test of time often have cash game skills that provide a steady stream of income, but even then they can lose for significant stretches at a time as well)

Seriously, I can't imagine being a pro who plays only tourneys. It would probably take 10 years to get enough data for a really meaningful winrate.

It's been an interesting journey for me this year. I started playing poker actively since June or so. I have two significant wins on the season, one from Feb. and another in July. Which really drove the desire to improve my knowledge and skills. (Also to build my own poker table, and get in on the Majestic chip buy) It's funny that in the process of learning how to try and take my game to another level, i have seen a backslide. I have since figured out a least a couple holes in my game, and believe now that I can get back on the right track, but this made me wonder, what would you consider a successful percentage of games cashed (in the black)? For the moment, i am still up for the year, but with the building of the table and purchase of the chips, my bankroll has to be rebuilt. So instead of setting unrealistic expectations for myself, i was wondering what you think,

Take a look at this: http://pokerdope.com/poker-variance-calculator/

It's not really applicable since it sounds like you're playing tourneys, but it will illustrate the massive number of hands you need to play in order to get a statistically significant understanding of your winrate.

For example, let's say you crush 1/2 for $30 an hour (true win rate), and play 10k hands, roughly 300 hours, or ~45 trips to the casino. There's a 10-15% chance you actually lose money over that period of time. And there's also a good chance you'll be up $5-10k. (This is how I justify to myself that I'm a winning player...I'm in that 10% )

If you really want to improve your poker game and be the best player you can, I'd recommend finding any sort of online poker and start tracking your play through one of the apps on the market. It doesn't need to be for much money at all, but until you do that and log 20-50-100k hands you won't know that you don't defend the SB enough, or you 3bet too light in middle position, or you don't showdown enough TP, etc.

Don't get me wrong you can still improve by reading and playing live and analyzing your play after the fact, but if you're serious about it, there's a deeeeeeep rabbit hole for you to go down. Good luck! It might sound daunting but it's really not once you dive in.

 

[DrStrange]

I think the authors of the software missed the boat with their guidance about win rates and standard deviation. I wonder if the authors ever really played poker in a serious way or perhaps wrote the simulation to convince themselves they weren't loser but just on a really long bad run. I expect they did get the math right, but the suggested inputs are wildly wrong. Start with bad input and you get worthless output, or worse, misleading output.

To start, if you are not collecting data you do not have a clue about your standard deviation. (and for the sake of purity, if the sample size is too small to estimate win rate then the standard deviation can't be trusted either.) The software recommends 100bb/100hands as a standard deviation. That is much too high for limit poker and a too low for NL. My personal values are a standard deviation of 13bb/100 for on-line limit poker and 150bb/100 for live no limit. I play as TAG. I see LAG get values more like 17bb/100 for limit and 300bb/100 for no limit.

Same thing with an estimated win rate, if you don't collect the data your best guess isn't likely to be accurate. The simulation defaults to 2.5 bb/100, a rather good rate for a limit player but not so hot for no limit. This works out to be something $5 winnings per session of $1/$2. I totally agree that someone who measures a win rate of $5 per session in a $1/$2 no limit game has no clue if he is a winning or losing player and will never know over his/her life time if they win or lose money five bucks per night. And that doesn't matter from a bankroll perspective, since the money won or lost is meaningless. It is notable that this also means you can't pick out a +$5/100 improvement in your game, that size change is going to be overwhelmed by variance.

We know through our own observations that there are players who win a lot more than five bucks a night and others who lose a lot more. If we start with a TAG who wins "big" (a 100bb/100 winner) we get 95% confident he is a winning player after only ten sessions and get 99.9% confident after 30 sessions. The original poster plays enough to get in over 30 sessions a year - enough to be quite confident of his status if he is one of the big winners or losers.

If we take a TAG who wins $60/session at $1/$2 NLHE, it is going to take longer to reach 95% confidence - a solid 100 sessions. That surely isn't fast but way better than the vast number of sessions that commonly held myths suggest are needed. It truly doesn't take decades of live play to figure out big loser / loser / break even / winner / big winner - I can pick out the extremes in 30 sessions or less and the modest winners and losers in 100 sessions and so can you.

DrStrange

 

[lherron]

I think the authors of the software missed the boat with their guidance about win rates and standard deviation. I wonder if the authors ever really played poker in a serious way or perhaps wrote the simulation to convince themselves they weren't loser but just on a really long bad run. I expect they did get the math right, but the suggested inputs are wildly wrong. Start with bad input and you get worthless output, or worse, misleading output.

To start, if you are not collecting data you do not have a clue about your standard deviation. (and for the sake of purity, if the sample size is too small to estimate win rate then the standard deviation can't be trusted either.) The software recommends 100bb/100hands as a standard deviation. That is much too high for limit poker and a too low for NL. My personal values are a standard deviation of 13bb/100 for on-line limit poker and 150bb/100 for live no limit. I play as TAG. I see LAG get values more like 17bb/100 for limit and 300bb/100 for no limit.

Same thing with an estimated win rate, if you don't collect the data your best guess isn't likely to be accurate. The simulation defaults to 2.5 bb/100, a rather good rate for a limit player but not so hot for no limit. This works out to be something $5 winnings per session of $1/$2. I totally agree that someone who measures a win rate of $5 per session in a $1/$2 no limit game has no clue if he is a winning or losing player and will never know over his/her life time if they win or lose money five bucks per night. And that doesn't matter from a bankroll perspective, since the money won or lost is meaningless. It is notable that this also means you can't pick out a +$5/100 improvement in your game, that size change is going to be overwhelmed by variance.

We know through our own observations that there are players who win a lot more than five bucks a night and others who lose a lot more. If we start with a TAG who wins "big" (a 100bb/100 winner) we get 95% confident he is a winning player after only ten sessions and get 99.9% confident after 30 sessions. The original poster plays enough to get in over 30 sessions a year - enough to be quite confident of his status if he is one of the big winners or losers.

If we take a TAG who wins $60/session at $1/$2 NLHE, it is going to take longer to reach 95% confidence - a solid 100 sessions. That surely isn't fast but way better than the vast number of sessions that commonly held myths suggest are needed. It truly doesn't take decades of live play to figure out big loser / loser / break even / winner / big winner - I can pick out the extremes in 30 sessions or less and the modest winners and losers in 100 sessions and so can you.

DrStrange

I assume you’re joking that the authors created this tool this to make themselves feel better about losing at poker but I do agree it lacks some practical advice on answering the question of “are you a winning player or not, and to what degree”. The equations themselves are indisputable, but of course it makes many assumptions about the games and players that are definitely not 100% accurate (consistent game difficulty, opponent behaviors, etc). Calculating your precise win rate isn’t something most of us should be too concerned with, but if you’re a pro considering stepping up to a higher limit and pushing your bankroll to the limit, this type of calculation is essential if you want to understand the risks.

And I agree that different games and player types are going to yield very different stdevs. But to illustrate to the OP how many hands he needs to play to get real info on his win rate, you can use any number between 100-200bb/100 for live NL and get a pretty good idea.

I also think you’re right on with your last sentence, 30/100 sessions for extreme/average winners/losers. Because if you are a huge winner by true skill but have one of these "95% unlikely horrible runs”, you’ll probably notice that you keep getting coolered left and right, and overall running like complete garbage It will be obvious. Again, this goes to providing a little real life context with all the math, which the author could definitely do.

In the end, it’s in everyone’s best interest for the winners and losers not to be able to readily identify themselves in the short term and win/lose all the money over just a few sessions…

I would also love to see different stdevs for each major game and player type. I imagine LAG PLO is totally insane.