Craps

Is my SRR better than a Randy's 6.0 ?

Spread the love



While at my practice table collecting data and figuring my SRR for so many throws
I was always bothered by the question of whether the calculated SRR was valid and
whether I had enough throws for confidence in the value. The SRR seemed to vary so much,
never consistent. So I made up this spreadsheet, first for the 6/8,
then for the 7 for a random roller which is the basis of my argument.

Please take a look at this and say what you think.

First, the data collection process must be valid:
using the hardway set throughout, all throws must be included, with all the sevens.

Then, the total number of throws must be large "enough":
in accordance with the statistical three-sigma rule as set out by Skinny
in his post "how can you tell if you are influencing the dice?"

A Randy, we figure, will have an SRR of 6.0 so according to the three-sigma rule,
for a 95% confidence, at our practice table he could fling the dice against the
wall, collect data while figuring the SRR and we will notice from the spreadsheet presented
here for the Randy, that
after 30 throws the SRR would vary from a low of 3.3 and a high of 32.7, a very wide range;
after 60 throws the SRR would vary from a low of 3.8 and a high of 14.2, not a tight range;
after 600 throws the SRR would vary from a low of 5.1 and a high of 7.3, a tighter range;
after 6000 throws the SRR would vary from a low of 5.7 and a high of 6.4, a tighter range;
after 7200 throws the SRR would vary from a low of 5.7 and a high of 6.3, a very tight range.
On the spreadsheet you’ll notice that the higher the total number of throws the tighter the range
and the more credible the calculated value of the SRR.

So if you are practicing to improve your SRR you’ll need to do better than
this Randy. If you are not better, at any point of so many throws on the chart, you are a Randy.
If you threw 600 times and your SRR is no better than 7.3, you’re a Randy.
If you threw 7200 times and your SRR is no better than 6.3, you’re a Randy.

What do you think?

Attached files


Replies:

Posted by: Not2soon on May 30, 2012, 3:15 am

If you threw 600 times and your SRR is no better than 7.3, you’re a Randy.
If you threw 7200 times and your SRR is no better than 6.3, you’re a Randy.

Not quite true. Your chances of being random are better but it does not mean that you are. It means that with 7200 throws a random roller could possibly have an SRR of 6.3 but it does not mean that your SRR of 6.3 is random.

Posted by: the gman on May 30, 2012, 4:04 am

there appears to be much more than just SRR in determining if you have some
control over your outcomes.

I recently lost my data on smart craps for some reason and am starting over. My srr is only 7.14
which is lower than it was, however it stills shows some good trends.

Over my last 1445 rolls i have an SRR of 7.14
I have 31 more 6’s or 15.5 % more than would be expected
I have 39 8’s more than expected or 19.5% more
i have 38 less 7’s than expected by smart craps or 18 % fewer

I sure am no expert on this stuff, but seems to me it shows i have a pretty good advantage
over where i was a couple of years ago as a ramdom thrower, and that if i bet it should be
on the 6 & 8

Smart craps should show you were you are in relation to where you were as a ramdom guy.

gman

Posted by: Finisher on May 31, 2012, 2:16 am

Clll Just did a session that was 103 numbers with a SRR of 5.42 so I am still a randy . the throws were 3-10-13-3-12-13-3-5-41. I had 19 7s with 10 for come out 7s . The roll had 1-2s 6-3s 10 4s 7-5s 15-6s 19 -7s 12 -8s 10 -9s 9 -10s8 -11s and 6 -12s .It seems that I am throwing less 2s then normal so maybe I am on the right track ?
Love your charts.
Good Rolling.

Posted by: fscobe on May 31, 2012, 11:30 am

Finisher, if your SRR results over a few thousand rolls stayed in the 5.49 range that might indicate control. You can have control that goes over 1:6 and also control that goes under 1:6.

On-axis shooters can have a 1:6 and still be controlled shooters. The SRR is one indicator of control but SmartCraps is another. Someone with a low SRR can actually win more money than someone with a higher SRR if that someone has strong axis control.

As you can see, this is a somewhat complicated area but the key is to keep track over time. And here’s another wrinkle — your results will vary. You can get better or worse over time. Your control can beceome stronger or weaker.

Yes, it can drive you crazy!

Posted by: the gman on May 31, 2012, 1:51 pm

last nights play is an example of what Frank is talking about.

i had 6 hands…. the last hand was a 7 out which hurt the over-all SRR which ended
at 8.5 for the evening. I won, but only 25% of my buy in.
I felt i threw better, but my data said i did not. I am a 6 & 8 better and last night i ended up with about
27.8% of my throws were 6 & 8 and a radom thrower should be about 27.7% .
The real data shows that in terms of my play, i had no more control last night than anyone else
at the table.

This was a 16′ table, and while i get lucky from time to time, the data shows i would only
be kidding myself to think i have any degree of advantage on this table.

gman

Posted by: Finisher on May 31, 2012, 5:46 pm

Clll I was using a hard way set of 5-5 on top with 6 to left. It helped to have 10 wins on come out 7s and 4 come out 11s .
the hard ways were 4s 2-6s 3-8s 2-10s. with 10 point wins.
Made it passed the 5 count 5 out of 9 rolls hope to start doing better with that.
Good Rolling.

Posted by: SevenTimesSeven on June 5, 2012, 4:33 am

"the gman" wrote: there appears to be much more than just SRR in determining if you have some
control over your outcomes.

I recently lost my data on smart craps for some reason and am starting over. My srr is only 7.14
which is lower than it was, however it stills shows some good trends.

Over my last 1445 rolls i have an SRR of 7.14
I have 31 more 6’s or 15.5 % more than would be expected
I have 39 8’s more than expected or 19.5% more
i have 38 less 7’s than expected by smart craps or 18 % fewer

I sure am no expert on this stuff, but seems to me it shows i have a pretty good advantage
over where i was a couple of years ago as a ramdom thrower, and that if i bet it should be
on the 6 & 8

Smart craps should show you were you are in relation to where you were as a ramdom guy.

gman

Gman,

Yes, I agree there is much more than just SRR in determining if you have some control over your dice outcomes. But in as much as SRR is one factor, you can compare yourself with a random thrower, using the three-sigma statistical calculations, which I will show correlate with your results out of Smart Craps.
Please see the attached chart "WhatWouldRandyDo.xlr" for reference, which includes your 1445 throws.

For your 1445 throws, your SRR and your performance with the sixes and eights show better than those for a random thrower, for a 95% confidence level:
(a) Your stated 31 more sixes ,15.5% more than expected, calculates to 200
sixes expected (31/15.5%).
Chart shows EXP=200.7 sixes.
Your actual adds up to 231 sixes thrown (200+31).
The chart shows Randy would throw a high of 227 sixes at best.
(b) Your stated 39 more eights (19.5% more than expected) calculates to 200
eights expected (39/19.5%).
Chart shows EXP=200.7 eights.
Your actual adds up to 239 eights (200+39).
The chart shows Randy would throw a high of 227 eights at best.
(c) Your stated SRR of 7.14 calculates to 202 sevens thrown (1445/7.14).
Chart shows EXP=240.8 sevens.
Your stated 38 less sevens calculates to 202 sevens (240-38).
The chart shows Randy would throw a low of 212 sevens for a 6.8 SRR, at best.

I took Gman’s ‘expected’ to indicate the expected mean value(EXP) in the three-sigma statistical calculations.
Don’t these numbers come together nicely and conveniently? LOL
Anyone agrees or disagrees?

Attached files

Posted by: Skinny on June 5, 2012, 5:15 am

SevenTimesSeven,

Nice job, I like what you have done here. I agree you need to look at more than just SRR to see if you are influencing the dice or not. SRR is good enough when you are first starting out and trying to see how well you are catching on to controlling the dice. But after you have developed a throw and have the technique down fairly well, you need to look at more than just SRR.

Your spreadsheet enables you to analyze your results for all the numbers to see how you compare to random. It is a good tool to aid you in your analysis. The closer you come to the 3 sigma level (99%), for all the numbers the better you are performing.

I think your analysis of Gman’s data is spot on and helpful to others on the board. I hope you continue to post your ideas, questions and observations. I for one find them thoughtful, insightful, interesting and informative.

Posted by: the gman on June 5, 2012, 5:07 pm

Shooter name:

New shooter
# of roll sets in Pro Test:

42
# of roll sets excluded:

0
Note: excluded roll sets are not included in Pro Test results below
Total # of rolls:

1435
Pro 1 Test:

Both dice on axis
# of passes: 696 (more is good)
Expected # of passes: 637.78
Pro Test score: 0.11%
Pass/fail: pass (1% or less to pass)
Pro 2 Test:

Primary face hits
# of passes: 188 (more is good)
Expected # of passes: 174.00
Pro Test score: 11.92%
Pass/fail: fail (1% or less to pass)
Pro 3 Test:

Double pitches
# of failures: 129 (less is good)
Expected # of failures: 174.00
Pro test score: 0.00%
Pass/fail: pass (1% or less to pass)
SRR evaluation:

# sevens rolled: 201
# sevens expected: 239.17
SRR: 7.14
Pro Test score of SRR: 0.38%
less than 1% is a pass
Outcomes summary:Outcome # Occurences # Expected Pro Test score
2 35 (less) 39.86 23.50%
3 61 (less) 79.72 1.78%
4 135 (more) 119.58 7.70%
5 158 (less) 159.44 45.50%
6 230 (more) 199.31 1.04%
7 201 (less) 239.17 0.38%
8 238 (more) 199.31 0.18%
9 160 (more) 159.44 48.50%
10 121 (more) 119.58 45.50%
11 73 (less) 79.72 22.50%
12 23 (less) 39.86 0.43%

i dont know what all this means exactly except when i said expected that was just what the
smart craps diagram said.

As i have said this is a small sampling, i keep records for each casino trip and for waht ever reason i
have a much higher SRR in a casino.

For the casino i keep track of my throws….
and 6 & 8’s it seems if i can have 33% or higher on per centage of 6 & 8"s that i throw of the over-all
number of throws i can make money.

Smart craps indicates to me as does my casino records that i have a good advantage on the 6 & 8 and 4. Timmer
recently indicated i need to add 4 to my bets when i have 6 or 8 as my pass line number and that makes sense.

gman

Posted by: Stickman on June 6, 2012, 2:29 pm

Gman,

Your Smart Craps statistics are very good. The only test you did not pass was Pro Test 2 (hardways). Even though you did not pass this test, you are throwing significantly more hardways than random. You are, indeed, controlling the dice.

While the number of rolls may not be enough to validate an SRR, you have plenty of rolls recorded to feel confident that you are controlling your dice by keeping your dice on axis – at least in the home practice environment. Smart Craps is unique in measuring on-axis performance and using the information to determine you have an edge in as little as 100 rolls. You have 14 times that scattered over 42 different roll sets so I would feel good about what Smart Craps tells you about your edge and best dice set(s). Have you checked which set Smart Craps recommends and what your player edge is? I suspect it will be the 3V.

In any case, your stats look great. Use the information the program gives to to bet into your edge.

Jerry

Posted by: the gman on June 6, 2012, 3:47 pm

thanks for the comments Jerry

gman