Difference between ACTUAL and EXPECTED results

[matt21]

hi everyone, hope all your counting is going well.

i wanted to ask how peoples' session results vary from expected results. The games that I play vary from casino to casino in terms of 4D/6D as well as playing conditions (pen & # of players) but on the whole I believe i am facing the following expected values:

EV per hour = 1.47 units ($36.72)
SD per hour = 28.98 units ($724)
68% Conf Interval = -$688 , $761
95% Conf Interval = -$1,412 , $1,486

EV for 3 hours = 4.41 units ($110)
SD for 3 hours = $1,255
68% Conf Interval = -$1,145 , $1,365
95% Conf Interval = -$2,400 , $2,620

Standard unit = $25 and total bankroll = $25,000

I have been playing at the $25 level for 37 hours (after completing close to 300 hours at a mix of $5 & $10 standard bets) .

What I am surprised about is that I have not had a single session where I have won or lost more than $1,000 (my sessions vary in length from 1/2 hour to 5 hours). Instead most of my session wins and losses fall somewhere into the $250-$700 range (and the remainder into 0-$200 range). I am thinking that i should be hitting bigger wins and bigger loses in my game when taking into account my calculated theoretical values. There may be an error with my model - I have been meaning to have a look at it together with Kasi. Note my historical return per hour of 1.35 units, is sitting close to the EV.

Additionally I am hitting +4 (using Hi-Low) and higher fairly regularily i.e. i am certainly pushing out big bets at least several times in each session.

What I'd like to know is whether people experience much wilder swings in their sessions (e.g. often winning or losing more than say 40 or 50 units). I would be primarily in players' ACTUAL experiences, not in EXPECTED results

 

[ThodorisK]

My advise? Either they cheat you or they do not cheat you. If they do not cheat you, you are losing time with all these calculations. Keep playing. Why are you not playing now? Time is money.

The only important thing to have in mind so that you dont get disencouraged or confused, is that before you head up to the point that you will never again fall below your very first starting bankroll, you are bound to experience sooner or later a long negative variance (unless you are one of the most lucky persons on earth) where you will fall 30 or 50 or 100 bets, it depends how lucky you are. This is what the risk of ruin formulas tell us for a number of hands which tends to infinity.

If you meet an extreme bad variance in a casino from the first day you started playing there, be suspicious of cheating and seriously think of never playing there again. A very fast dealer knows how to cheat. I can give you examples if you want, and Petter Griffin has mentioned some of them that he experienced.

 

[blackjack avenger]

If You Play It, They Will Come

The longer you play the bigger the swings you will face. Of course your overall results will flatten out over time.

 

[creeping panther]

Matt

If all is nearly equal play the 4d and not the 6d.

Personnaly i think you are doing fine and should take of like a rocket at any time... if you can play all 4d:grin:

CP

 

[kewljason]

hi everyone, hope all your counting is going well.

Note my historical return per hour of 1.35 units, is sitting close to the EV.

What I'd like to know is whether people experience much wilder swings in their sessions (e.g. often winning or losing more than say 40 or 50 units). I would be primarily in players' ACTUAL experiences, not in EXPECTED results

Since your longer term return per hour is close to the EV, it sounds like you are doing all right to me. 37 hours as I am sure you know is still very short term. Eventually you will have those wild swings, both good and bad and when you do, may wish for this brief period back when everything went fairly smoothly. But if you play long enough, you will have plently of everything and it will all even out in the end. The longer you play the closer you should be to EV. I always figure that by the time I hit 500 hours each year, I'm getting into the "long run" I dont get in 100 hands an hour though, more like 60-70, so that may not be the greatest measuring stick.

 

[EasyRhino]

Assuming your max bet gets into $200 or so, then yes, it sounds like you've had unusually sedate swings.

Don't worry, it's probably the last time it will happen.

Is penetration decent? If it's not, there may be few high counts, and it may be dampening your betting level

 

[sagefr0g]

...
What I'd like to know is whether people experience much wilder swings in their sessions (e.g. often winning or losing more than say 40 or 50 units). I would be primarily in players' ACTUAL experiences, not in EXPECTED results

what is your spread?
well anyway for me when i was using a 1-8 spread for six deck shoes about the best i experienced (using hi/lo) was 40 units. it was usually much less.
when i pushed my spread to 1-10 i had a more severe swings. mostly down. but the 1-8 spread i used for around 200 sessions that might average 3 hours play at mostly full tables. the 1-10 spread i didn't stick with very long.

 

[sagefr0g]

.....
If you meet an extreme bad variance in a casino from the first day you started playing there, be suspicious of cheating and seriously think of never playing there again.....

just curious as to why you stipulate the first day and not a more protracted sampling?

 

[matt21]

hi again, and thanks for all your replies - this is very appreciated!!
i'll clarify and comment a little further.

what is your spread?
well anyway for me when i was using a 1-8 spread for six deck shoes about the best i experienced (using hi/lo) was 40 units. it was usually much less.
when i pushed my spread to 1-10 i had a more severe swings. mostly down. but the 1-8 spread i used for around 200 sessions that might average 3 hours play at mostly full tables. the 1-10 spread i didn't stick with very long.

my spread is quite large - from 1 or 2 boxes of 1/2 unit each to 3 boxes of 8 units (24 units total). A couple of points though: i hardly ever wong and secondly i generally play pretty empty tables so play 100 hands per hour or more.

Assuming your max bet gets into $200 or so, then yes, it sounds like you've had unusually sedate swings.

Is penetration decent? If it's not, there may be few high counts, and it may be dampening your betting level

Ok, so they are unusually sedate. My max bets are in the range of $400-$500 but spread over 2-3 boxes. So this makes it sounds even more sedate LOL !! And yes penetration is decent - mostly in the 75-83 range in both 4D and 6D.

I have to admit every time that I shift to a new standard unit level e.g. from $5 to $10 and from $10 to $25 i initially am always scared to push out those max bets particularly to keep on pushing them out if i am winning. In terms of units i have had much wilder swings with the $5 and $10 std bets than I have had with the $25 std bets.

So from your comments, it seems that it will just be a matter of time. The 36 unit gain in ONE SHOE today was a good start (the true count hit +3 after only 1 deck in 6D game, then never went below and I got very lucky with all my index plays and the dealer getting really crap cards) :grin::grin::grin: I mean how lucky can you get!

 

[ThodorisK]

just curious as to why you stipulate the first day and not a more protracted sampling?

Yes, this is what I meant, the overall results, but from the day one started playing in a casino.

However, I would run away even if the first day's results consisted extremely rare bad luck.

 

[Traveller]

Hi Matt

I ran some sims for you based on the info i could get from your posts.
This is for 6 decks, DAS, RSA upto 4 hands, no surrender, cut 78. This is based on 500 rounds for each session, or five hours play. It is also based on one spot of 25 below +1, 3 spots of 50 at 1 to 2, 3 x 100 at 2 to 3 and 3 x 150 at counts above +3.

Chances of losing $1000 at the end of the session 29.6% Nadir 53.6%
Chances of Winning $1000 at the end of the session 34.5% Apex of 60.9%
Chances of losing $2000 at the end of the session 16.7% Nadir 28.8%
Chances of winning $2000 at the end of the session 20.5% Apex 35.0%
Chances of losing $3000 at the end of the session 8.1% Nadir 13.1%
Chances of winning $3000 at the end of the session 10.5% Apex 18.2%
Chances of losing $4000 at the end of the session 3.6% Nadir 5.9%
Chances of winning $4000 at the end of the session 5.3% Apex 9.1%
Chances of losing $5000 at the end of the session 1.6% Nadir 2.5%
Chances of winning $5000 at the end of the session 2.2% Apex 3.7%

T.

 

[sagefr0g]

Yes, this is what I meant, the overall results, but from the day one started playing in a casino.

However, I would run away even if the first day's results consisted extremely rare bad luck.

just one day's worth of extremely rare bad luck would be enough?
that's what i thought your point was.
i'm curious as to your rational.
please continue explaination, inquiring minds want to know lol.
and another question, how extremely rarely bad would it need to be on a scale of one to ten?

 

[ThodorisK]

Not "one day" but the first day you have ever played in that casino.

A 5% rare bad luck in this case should make you run away. It could be just real variance, but why risk it? Go back to the casinos where you have made profits instead, or go to the next casino you have never visited.

 

[Kasi]

What I'd like to know is whether people experience much wilder swings in their sessions (e.g. often winning or losing more than say 40 or 50 units). I would be primarily in players' ACTUAL experiences, not in EXPECTED results

Actual unit swings of any unit amount are only "wild" when compared to expected lol. What else is there?

Are your last 37 hours, 64 half-hour sessions and 1 5 hr session? 7 5 hr sessions and 4 half-hour sessions?

All I can suggest is, given different games, number of people at table, possibly changing unit spread when going from $5 or $10 or $25 units, is why not estimate "hands-played" after a session and go from there? Taking it one step further, why not ask yourself this same question after each "session" for whatever game, spread, number of people at table? Why wait 37 hours or 337 hours to ask it and offer, apparently, a weighted blended expectation over all these games put together?

Apparently your numbers assume 100/hds per hour, guess I should say rounds/hr given you spread to multiple hands, which is fine if true.

From where do you get your numbers anyway? A sim? Many different sims for all the different scenarios you have played under?

Next time you play for an hour, or 5 hours, can you come home and answer, approximately anyway lol, what was expected over those 60 hands/hr if it happened to be at a full table or for those 150 hands/hr if it happened to be heads-up, spreading to multiple hands or not, wonging or not, risk etc?

It's great you're doing so well.

I guess my long way of saying, if you haven't, maybe it's time to consider investing a couple "sessions'" EV in some sim software so you can answer this stuff to your own satisfaction after each and every session. If you have the software, why are you asking?

Like, since I don't have too much to worry about anyway, I worry about maybe you only having a 1000 unit roll ($25 unit with $25K roll) in certain scenarios, especially play-all. What risk do you think you are currently playing at, etc?

Just late night thoughts. As always, if you think I can help, you got my number. Happy to try anyway lol.

 

[Kasi]

A 5% rare bad luck in this case should make you run away. It could be just real variance, but why risk it? Go back to the casinos where you have made profits instead, or go to the next casino you have never visited.

A 1 in 20 chance of "bad luck" is "rare"? You could be running out of casinos to play at pretty soon lol.

 

[ThodorisK]

Well if you keep playing you are bound to meet this 5% sooner or later, but usually after you have had some positive swings too (above the starting bankroll you began with at that particular casino). But if I fall 50 bets below that starting bankroll without ever having gone above 10 bets above it, I will run away. In how many casinos this will happen if none of them cheats?

 

[matt21]

Actual unit swings of any unit amount are only "wild" when compared to expected lol. What else is there?

Are your last 37 hours, 64 half-hour sessions and 1 5 hr session? 7 5 hr sessions and 4 half-hour sessions?

All I can suggest is, given different games, number of people at table, possibly changing unit spread when going from $5 or $10 or $25 units, is why not estimate "hands-played" after a session and go from there? Taking it one step further, why not ask yourself this same question after each "session" for whatever game, spread, number of people at table? Why wait 37 hours or 337 hours to ask it and offer, apparently, a weighted blended expectation over all these games put together?

Apparently your numbers assume 100/hds per hour, guess I should say rounds/hr given you spread to multiple hands, which is fine if true.

From where do you get your numbers anyway? A sim? Many different sims for all the different scenarios you have played under?

Next time you play for an hour, or 5 hours, can you come home and answer, approximately anyway lol, what was expected over those 60 hands/hr if it happened to be at a full table or for those 150 hands/hr if it happened to be heads-up, spreading to multiple hands or not, wonging or not, risk etc?

I guess my long way of saying, if you haven't, maybe it's time to consider investing a couple "sessions'" EV in some sim software so you can answer this stuff to your own satisfaction after each and every session. If you have the software, why are you asking?

Hi Kasi - thank you for your reply!!

I have actually been doing many of the suggestions you made. For example I am now tracking my EV for each individual session - i.e. i mentally take note of the game rules, the average penetration, the average number of players per table and the amount of time played. I then plug these numbers into my probability model (whihc i would like to act together with you!!) which also incorporates some sim figures. So I am glad i have started doing these BEFORE you suggested them LOL :grin: i guess that provides me with some confidence that i am on the right track. The main reason i did this was to objectively compare the playing conditions of the various casinos i play it - rather than by going by feeling.

In doing this i have just found that my actual results do not deviate from my EV very much - nearly always staying within 0.5 standard deviations - thats what led me to asking the question whether other counters are often experiencing 1 or 2 standard deviation results in their sessions.

By the way, this week has been the best ever in terms of results. I had seven consecutive winning sessions and made $3,000 (120 units) in 3 days. I was just really lucky with my big hands. :grin:

 

[zengrifter]

What I'd like to know is whether people experience much wilder swings in their sessions (e.g. often winning or losing more than say 40 or 50 units). I would be primarily in players' ACTUAL experiences, not in EXPECTED results

YES. zg

 

[Kasi]

In doing this i have just found that my actual results do not deviate from my EV very much - nearly always staying within 0.5 standard deviations - thats what led me to asking the question whether other counters are often experiencing 1 or 2 standard deviation results in their sessions.

By the way, this week has been the best ever in terms of results. I had seven consecutive winning sessions and made $3,000 (120 units) in 3 days. I was just really lucky with my big hands. :grin:

Well, I probably can't help much with 1/2 unit to 1-2 boxes and 8 units to 3 boxes, etc lol.

Not sure but maybe CVDATA would call betting 3 boxes 1 "round". Maybe CVCX would call it 3 "hand"? Could be a factor maybe lol.

Hey, if you have confidence in your numbers, fine lol. But maybe one SD # might not fit for all the different things you may be doing lol. So what if I think 28 units/hr might be low. Maybe it's not. I don't know.

Guess it works the other way too lol - was the $3K win in 3 days never outside 0.5 SD in a session? How "really lucky" was it?

But, yes, to your original point, I'd say never having any sessions results exceeding 0.5 SD either way could be an indication of something.

Whatever, why re-invent the wheel? It's time to maybe blow a max bet on some software? One sim for one game at a time lol.

Easy for me to say :grin:

Continued good luck to you.

 

[matt21]

i am putting a bit more time analysing all of session using CV data for simulations now. this is proving to be very useful. Definitely recommend CV Data to anyone who is serious about all this.

and talking about swings - they are coming home to me baby!

just last week alone i was down more than $1,000 (40 units) three times but recovered in each case - in one situation my chip pile went from $2,000 (4 buy-in's of $500) to $500 and then to $2,500 - i basically won about $2,000 in the space of three shoes - that was fun :grin: !! hence although i had big swings my final session results just said +1000, -250 and +200.

but this activity is definitely mentally draining.

All in all, up $730 in 30 hours of play for the week.